Investigating patterns of prehistoric dispersal in
Eastern Polynesia: a commensal approach using
complete ancient and modern mitochondrial genomes
of the Pacific rat, Rattus exulans.
Katrina West
A thesis submitted in fulfillment
of the requirements for the Degree of
Master of Science in Genetics
at the University of Otago, Dunedin,
New Zealand
June 2016
Abstract
The pattern of prehistoric human dispersal through Eastern Polynesia,
representing one of the last major migrations in human history, remains highly
contested with conflicting archaeological evidence. The commensal approach,
using the phylogenies of commensal animal and plant species has been
increasingly applied since the late 1990s as a proxy for human movement.
Previous genetic commensal research using the Pacific rat, Rattus exulans, a
species transported across Remote Oceania as part of the Polynesian expansion,
has been largely unsuccessful in distinguishing finer-scale movements between
Remote Oceanic islands, with the majority of rat samples adhering to the same
mitochondrial control region haplotype. It was anticipated that with greater
molecular resolution provided by complete mitochondrial genome sequencing, a
greater number of Pacific rat lineages would be distinguished and could be used as
a proxy to investigate the origins and dispersals of Eastern Polynesian people.
Archaeological rat specimens were obtained from the earliest occupational
contexts across Western and Eastern Polynesia. Complete mitochondrial
sequencing of both ancient and modern specimens of the Pacific rat was
undertaken. Nine ancient and twenty-five modern haplotype lineages were
identified. A central haplotype, derived from an ancestral haplotype in Western
Polynesia, is ancestral to all Eastern Polynesian rat populations and reflects a
previously proposed central East Polynesian homeland region from which eastern
expansion occurred. An Easter Island and Tubuai (Austral Islands) grouping of
related haplotypes suggests that both islands were established by the same
colonisation wave, proposed to have originated in the central homeland region
before dispersing through the south-eastern corridor of Eastern Polynesia. The
application of second-generation sequencing in generating complete and credible
mitochondrial genomes from archaeological and modern commensal specimens
provides greater molecular resolution to investigate prehistoric dispersal and
interaction spheres across the Pacific.
i
Acknowledgements
It is with immense gratitude that I would like to acknowledge the people who have
made this research project possible.
Firstly, I would like to thank my supervisor, Lisa Matisoo-Smith, for your
invaluable advice and support over the last year. It has been an absolute pleasure
to have worked with you and to have been part of your lab. To Olga Kardailsky and
Catherine Collins
I am extremely appreciative of all the time you have spent with
me, imparting your knowledge from sample processing to bioinformatics. You are
both great teachers and I am deeply grateful for the skills you have taught me, of
which I am sure I will carry with me into my next endeavour. I would also like to
mention James Boocock - your assistance in bioinformatics was greatly
appreciated. I am also grateful to all the archaeologists and anthropologists who
excavated/trapped the Pacific rat samples for which I used in this project.
To Jessica, Kate, Edana and Denise
without you. You guys are amazing and I look forward to future catch-ups around
the world!
Most importantly to my family your continued support and faith in me, even from
afar, has been instrumental to my goals and success.
Of course, none of my work would have been possible without financial support
from the university. I am incredibly grateful for the University of Otago Masters
Scholarship, which has supported me during my time in New Zealand.
ii
Table of Contents
Abstract....................................................................................................................................... i
Acknowledgements .............................................................................................................. ii
Table of Contents .................................................................................................................. iii
List of Tables ............................................................................................................................ v
List of Figures .........................................................................................................................vi
List of Abbreviations .......................................................................................................... vii
1 Introduction ......................................................................................................................1
1.1
Overview of Pacific colonisation and dispersal models.......................................... 1
1.2
Interaction spheres in post-settlement Polynesia ................................................. 10
1.3
The commensal approach ............................................................................................... 12
1.4
aDNA: its use and limitations......................................................................................... 19
1.4.1
Preservation................................................................................................................................ 19
1.4.2
Contamination............................................................................................................................ 21
1.4.3
DNA Damage ............................................................................................................................... 22
1.4.4
New methodologies ................................................................................................................. 23
1.5
Research Objectives .......................................................................................................... 26
2 Methods ........................................................................................................................... 27
2.1
Modern DNA sequencing ................................................................................................. 27
2.1.1
Sample Collection ..................................................................................................................... 27
2.1.2
DNA Extraction .......................................................................................................................... 29
2.1.3
PCR amplification ..................................................................................................................... 29
2.1.4
Library preparation & sequencing .................................................................................... 29
2.2
Ancient DNA sequencing ................................................................................................. 32
2.2.1
Sample Collection ..................................................................................................................... 32
2.2.2
DNA Extraction .......................................................................................................................... 35
2.2.3
Library preparation ................................................................................................................. 36
2.2.4
Bait production & preparation ............................................................................................ 39
2.2.5
Hybridisation capture & sequencing ................................................................................ 40
2.3
Bioinformatics processing .............................................................................................. 41
2.3.1
Modern DNA processing ........................................................................................................ 41
2.3.2
Ancient DNA processing ........................................................................................................ 42
iii
2.4
Phylogenetic Analysis....................................................................................................... 42
3 Results .............................................................................................................................. 46
3.1
DNA preservation and sequence recovery ............................................................... 46
3.2
Sequence authenticity ...................................................................................................... 46
3.3
Phylogenetic analysis ....................................................................................................... 50
4 Discussion ....................................................................................................................... 64
4.1
DNA preservation in Pacific archaeological contexts ........................................... 64
4.2
The application of complete mitochondrial genome sequencing .................... 65
4.3
Implications for Pacific rat dispersal.......................................................................... 68
5 Conclusion....................................................................................................................... 76
5.1
Summary ............................................................................................................................... 76
5.2
Future Directions ............................................................................................................... 77
6 Supplementary material ............................................................................................ 78
7 Reference List ................................................................................................................ 89
iv
List of Tables
2.1
Tissue samples collected for modern Pacific rat DNA
27
processing.
2.2
a) Long-range PCR amplification for modern samples.
30
b) Long-range PCR amplification for bait.
2.3
Archaeological samples collected for ancient mitochondrial
33-34
DNA processing.
3.1
Summary statistics for successful library amplification and
47-48
post-sequencing processing of ancient R. exulans samples.
3.2
High-coverage ancient samples used in the phylogenetic
49
analyses.
6.1
Variable positions across 203bp of the control region
78-81
(from position 15,392 15,595).
6.2
Archaeological sample information.
82-85
6.3
Modern sample information.
86-88
v
List of Figures
1.1
Map depicting Near and Remote Oceania.
2
1.2
The geographical boundaries of Melanesia, Micronesia
3
and Polynesia.
1.3
A neighbour-joining tree and corresponding map depicting
15
the distribution of the mitochondrial control region
haplogroups I, II, IIIA and IIIB.
2.1
Geographical distribution of modern and ancient samples
28
collected for DNA processing.
2.2
Targeted overlapping fragments for PCR amplification.
31
2.3
Archaeological sample of Pacific rat (R. exulans) femur.
35
2.4
Overview of the hybridisation capture on beads method.
37
3.1
51
to the ancient sample MS10592.
3.2
Median-joining network of control region haplotypes.
52
3.3
Distribution of nucleotide diversity ( ) across a) modern
54
and b) ancient complete mitochondrial sequences.
3.4
Rooted maximum likelihood (ML) tree inferred from the
56
modern (complete mitochondrial) dataset.
3.5
Median-joining network of the modern (complete
57
mitochondrial) dataset.
3.6
Rooted maximum likelihood (ML) tree inferred from the
59
ancient dataset.
3.7
Median-joining network of the ancient (complete
60
mitochondrial) dataset.
3.8
Median-joining network of the modern and ancient
(complete mitochondrial) datasets, including acquired
GenBank sequences.
vi
63
List of Abbreviations
~
Approximately
±
Plus-minus
Five prime
Three prime
l
Microlitre
M
Micromolar
aDNA
Ancient DNA
BAM
Binary version of SAM file
bp
Base pairs
B.P.
Years before Present
COI or COX1 Cytochrome c oxidase subunit I
COII
Cytochrome c oxidase subunit II
COIII
Cytochrome c oxidase subunit III
Cyt b
Cytochrome b
DNA
Deoxyribonucleic acid
dNTP
Deoxyribonucleotide triphosphates
EDTA
Ethylenediaminetetraacetic acid disodium salt dihydrate
g
Gram
HVR I
Hypervariable I region
ISEA
Island Southeast Asia
kb
Kilo base pairs
km
Kilometre
M
Molar
mg
Milligram
MJ
Median-joining
ml
Millilitre
ML
Maximum-likelihood
mtDNA
Mitochondrial DNA
NADH
Nicotinamide adenine dinucleotide
ND4L-4
NADH Dehydrogenase subunit 4L
SGS
Second generation sequencing
vii
NJ
Neighbour-joining
PCR
Polymerase chain reaction
qPCR
Quantitative PCR
rpm
Rotations per minute
rRNA
Ribosomal RNA
SNP
Single nucleotide polymorphism
sq
Square
TE
Tris-EDTA
TET
Tris-EDTA with Tween
tRNA
Transfer RNA
U
Unit
VCF
Variant call file
viii
1 Introduction
1.1 Overview of Pacific colonisation and dispersal models
The expansion of modern humans into the Pacific, commencing with the mid-late
Pleistocene migration through south-east Asia to Australia and New Guinea, and
culminating with the colonisation of Eastern Polynesia within the last 1000 years,
represents one of the earliest and one of the last major migrations in human
history (Matisoo-Smith 2015; Duggan et al. 2014). Following the out of Africa
migration, a group of early modern humans in Sundaland (Southeast Asia)
migrated onto the island continent of Sahul (Pleistocene New Guinea and
Australia) approximately 55-40,000 B.P. (Summerhay
Allen 2004; Groube et al. 1986). This would have required crossing Wallacea, a
sizeable biogeographical ocean barrier separating Sundaland from Sahul. Multiple
open-water crossings, each over a minimum distance of 70km (Matisoo-Smith
2015), would only have been possible with the construction of seaworthy
watercraft. The crossing onto Sahul is arguably the earliest example of purposeful,
longSahul, humans quickly dispersed over the continent reaching southern Australia
and north-eastern New Guinea by 50with successive water crossings reached the Bismarck Archipelago by 40,000 B.P.
(Leavesley et al. 2002) and the Solomon Island chain by 28,000 B.P. (Wickler &
Spriggs 1988).
This region of early human settlement across what is now Australia, New Guinea,
the Bismarck Archipelago and the Solomon archipelago is increasingly referred to
1991). This is to distinguish
the region
much later period and are culturally distinct from those in Near Oceania (see
Figure 1.1). Historically, Oceania was classified into three cultural/geographical
regions 1.2). The Polynesian classification has been frequently used, as it is a meaningful
unit for a culturally homogenous region. The use of Melanesia and to a lesser
1
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
Figure' 1.1' Map' depicting' Near' and'Remote' Oceania.!These!regions!are!separated!by!
the!red!dotted!line.!The!Polynesian!Triangle!is!overlaid!over!parts!of!Remote!Oceania!that!
are! affiliated! with! the! Polynesian! culture.! Within! Polynesia,! there! is! a! distinct! cultural!
separation! between!Western!and!Eastern!Polynesia,!which!is!denoted!geographically!by!
the!blue!dotted!line.!Base!map!courtesy!of!Les!O’Neill.!
2!
!
!
!
!
Figure' 1.2' The' geographical' boundaries' of' Melanesia,' Micronesia' and' Polynesia.! These!
regions!are!denoted!in!green,!yellow!and!blue,!respectively.!Base!map!courtesy!of!Les!O’Neill.!
!
3!
extent Micronesia however, has been heavily criticised as the communities across
both regions are culturally, biologically, and linguistically diverse (Green 1991); a
reflection of population development over a 36,000 year time span. The use of
Remote and Near Oceania as both a cultural and geographical divide in Oceania
-island
distances. The extensive ocean distance separating Near and Remote Oceania was
never breached by the original inhabitants of the Solomon Islands. This crossing,
instigated by a new wave of human migration, was not to occur for another 25,000
years.
Around 3,400 B.P., Austronesian speakers from Island Southeast Asia (ISEA)
moved into the Bismarck Archipelago (Summerhayes et al. 2010), settling
alongside the original inhabitants. A mixed culture developed into what is now
known as the Lapita Cultural Complex (Kirch 2000). Lapita settlements are
commonly distinguished by unique earthenware ceramics and the introduction
and utilisation of new plant and animal species, such as the Pacific rat, Rattus
exulans (Matisoo-Smith et al. 2004). New watercraft technologies such as the
invention of outrigger canoes enabled the Lapita people to voyage out beyond the
Solomon Islands and make long distance open-sea crossings into the wider Pacific
(Irwin 2008). Their success instigated a phase of long-distance voyaging and
migration over the next few centuries.
Eastward expansion beyond Near Oceania led to the colonisation of the Reef and
Santa Cruz groups by 3200 B.P. (Green et al. 2008), Vanuatu and Fiji by 3100 and
3000 B.P. respectively (Denham et al. 2012), Tonga by 2800 B.P. (Burley et al.
2015) and Samoa by 2750 B.P. (Petchey 2001; Clark et al. 2016). After some
period of time, the descendants of these Lapita migrants proceeded to move
further east into what is now denoted as
Polynesian Triangle
sq. km stretch of ocean comprising of over 500 islands scattered between Hawaii,
Easter Island (Rapa Nui) and New Zealand (Matisoo-Smith 2015). Nearly every
habitable island in this region was colonised, some subsequently used for
permanent settlements and others abandoned (Kirch 2000). The timing and
sequence of settlement across many of the major archipelagos of Eastern Polynesia
4
remain largely unresolved, with conflicting estimates of initial colonisation varying
by more than 1000 years (Wilmshurst et al. 2008).
relied upon comparative material, cultural and linguistic analyses to develop early
dispersal models in Eastern Polynesia (Burrows 1938; Buck 1944; Emory 1946).
The prevailing scenario proposed that the Society Islands were the major dispersal
centre leading to the colonisation of the rest of the Polynesian Triangle.
Preliminary radiocarbon dating of Pacific samples during the late 20th century
however, gave way to new theories on Polynesian settlement. The Marquesas
Islands were identified as the first group of East Polynesian islands to be settled,
after they produced the earliest radiocarbon dates at approximately 1650 B.P.
(Emory & Sinoto 1965; Sinoto 1970, 1983). Subsequent radiocarbon dating
indicated that Easter Island was settled as early as 1550 B.P., the Hawaiian Islands
by 1200 B.P., the Society Islands by 1150 B.P., the Cook Islands by 1050 B.P., and
New Zealand between 1150-950 B.P. (Sinoto 1970, 1983; Simmons 1973;
Bellwood 1978; Jennings 1979).
Artefact assemblages characterised by diagnostic types of wooden hand clubs,
basalt adzes, compound-shank fishhooks and pearl pendants from early deposits,
indicated a direct cultural link between the Marquesas and the Society Islands
(Sinoto 1983). Early material types in the Henderson Islands, Mangareva and Cook
Islands also had similar characteristics to artefacts affiliated with the early
Marquesan culture. Other archaic artefacts found in the Society Islands, such as
whale-tooth pendants and diagnostic harpoon heads were likewise uncovered in
early sites in New Zealand, signifying a direct cultural link between these two
regions (Sinoto 1983). Reversed-triangular adzes from a later phase in the Society
Islands occupation, dated to around 600 B.P., were discovered in Hawaii, indicating
a later link between Hawaii and the Society Islands (Emory & Sinoto 1965; Sinoto
1983). In association with the early radiocarbon dates and cultural/material
analyses, an orthodox scenario emerged whereby 1) an ancestral Polynesian
society developed in the West Polynesian archipelagos of Tonga and Samoa over a
1500 year period; 2) the Marquesas Islands were colonised from Western
5
served as a major dispersal centre towards other East Polynesian islands; and 3)
the Society Islands served as a secondary dispersal centre, leading to the
colonisation of New Zealand and initiating a secondary movement to Hawaii
(Emory & Sinoto 1965; Bellwood 1978; Jennings 1979; Sinoto 1983).
The orthodox scenario however attracted substantial criticism. Excavation
sampling error as a result of limited and patchy archaeological activity on other
Eastern Polynesian islands alluded to the prematurity of the orthodox scenario as
an accurate East Polynesian dispersal model (Irwin 1981). A lack of shared
artefacts between the earliest assemblages uncovered in the Marquesas and those
in West Polynesia (i.e. Samoa and Tonga) also challenged the idea that the
Marquesas was the first group of islands to be colonised in Eastern Polynesia
(Bellwood 1970; Davidson 1976). If the Marquesas was directly colonised via
Western Polynesia, early assemblages from the Marquesas should resemble those
from a phase in Western Polynesian culture (Davidson 1976).
Given that Eastern Polynesian languages share many lexical and phonological
variants, and that Eastern Polynesian culture in general shares many similarities,
Kirch (1986) argued that these variants must have developed after an East/West
Polynesian split, but prior to further splits within the Eastern Polynesian
community. If the Marquesas was settled by 1650 B.P. and was succeeded with a
rapid migration to Easter Island and the Society Islands, it is unlikely that that an
ancestral East Polynesian language and other cultural innovations would have had
the time to develop prior to the rapid migration and establishment of communities
across the Polynesian Triangle (Kirch 1986). Alternatively, sustained inter-island
groups (Hunt 1979), which would have allowed shared language and cultural
innovations to develop concurrently (Kirch 1986). Similar artefact assemblages,
such as those between the Marquesas and the Society Islands, may therefore not
be representative of a one-way migration and settlement of early Marquesan
culture on the Society Islands, but of a shared culture between the archipelagos,
sustained by frequent exchange. The settlement sequence depicted in the orthodox
6
scenario therefore may not be as simplistic, but rather involved numerous interisland migrations and interactions (Biggs 1972).
Kirch (1986) also noted that the absence of diagnostic artefacts associated with
-tooth pendants, harpoon heads and
compound-shank fishhooks) in Hawaiian deposits, indicated that Hawaii may have
been settled prior to the development of archaic material culture in Central
Polynesia. Several C14 age determinations, suggesting Hawaiian occupation as
early as the fourth century A.D (~1600-1700 B.P.) (Kirch 1985) during which the
Marquesas was allegedly colonised, placed further strain on the orthodox scenario.
Kirch (1986) considered it unlikely that Hawaii was rapidly colonised after the
purported settlement of the Marquesas, given that there would not have been
enough time for the development of a shared ancestral East Polynesian language.
Kirch (1986) re-evaluated a number of radiocarbon dates and suggested that the
Marquesas was colonised during the late 1st millennium B.C (prior to 1950 B.P.),
and both Easter Island and Hawaii by 1550 B.P. He argued that the shared artefact
assemblages between the Marquesas, Society Islands and New Zealand are not
representative of an archaic material culture, but of a
consistently dated between 1250-850 B.P., which was after the
settlement of Easter Island and Hawaii.
Spriggs & Anderson (1993) conducted a meta-analysis of 147 previously published
radiocarbon dates in an attempt to revise the conflicting chronologies. Samples
and their corresponding age approximations were rejected as part of a
context, materials were associated with short Pacific chronologies or high inbuilt
age, samples were of mixed isotopic fractionation, if a stratigraphic context
provided no overlap in dating, and if samples were inadequately treated prior to
dating. The results from the Spriggs & Anderson (1993) analysis suggested a later
chronology for the settlement of East Polynesia, with the Marquesas settled by
1650-1350 B.P., Hawaii after 1350 B.P., Society Islands after 1200 B.P., the Cook
Islands by 1150 B.P. and Easter Island towards the end of the 1st millennium A.D
7
(prior to 950 B.P). This built on previous work by Anderson (1991) focusing on the
initial settlement of New Zealand, which was dated to between 950-750 B.P.
Subsequent radiocarbon studies using short-lived materials, e.g. short-lived plants
chronometric hygiene protocol, have generally supported a short and rapid
colonisation of East Polynesia from 1050 B.P. to 750 B.P. The short chronology
model, incorporating an 1800 year pause between the settlement of Western and
Eastern Polynesia, suggests the colonisation of the Society Islands and Marquesas
to have commenced between 1050-750 B.P. (Anderson & Sinoto 2002), 950-550
B.P. across the Cook Islands (Kirch et al. 1995; Allen & Wallace 2007), and by 750
B.P. in both Easter Island and New Zealand (Hunt & Lipo 2006; Wilmshurst et al.
2008).
Conve
model for an earlier and gradual colonisation of Eastern Polynesian islands from
2500 B.P. to 750 B.P. The paleoenvironmental approach uses palynological,
paleobotanical and sedimentological analyses of island ecosystems to identify
human disturbance in relation to forest clearance, erosion and alluvial deposition,
and the introduction and extinction of species, as a proxy for human occupation.
Stratigraphic and geochemical evidence from Mangaia, Cook Islands, reveal
dramatic changes to the erosional/depositional cycle and a decline in the organic
content in sediments, attributed to human occupation as early as 2000 B.P. (Kirch
& Ellison 1994). Palynological changes such as an increase of charcoal, in
conjunction with a reduction in forest taxa concentrations attributed to slash-andburn horticulture, is dated to around 2500 B.P. (Kirch 1996). In Hawaii, changes in
vegetation, the presence of charcoal, the introduction of Polynesian plants and the
Pacific rat, Rattus exulans, have been consistently dated to the period 900-1000
B.P. (Burney & Kikuchi 2006; Burney et al. 2001). Rat-gnawed seeds, reflecting the
anthropogenic introduction of Rattus exulans in New Zealand, were dated to
approximately 750 B.P. (Wilmshurst & Higham 2004).
8
The long or short chronologies for human colonisation in Eastern Polynesia offer
conflicting estimates that vary by more than 1000 years. A recent meta-analysis
incorporating 1,434 published radiocarbon dates from across the major
archipelagos of Eastern Polynesia, indicated that arguments for colonisation prior
to the first millennium A.D (before 1950 B.P.), have largely been based on
materials providing imprecise calibrations (Wilmshurst et al. 2011). Materials
such as long-lived plant, unidentified charcoal, bone and marine shell were often
unreliable, introducing substantial in-built age and measurement errors. Dates
from these types of materials extended back to 2450 B.P. and showed a larger
spread of ages from all island sites. In comparison, dates from short-lived plant
and terrestrial bird eggshell provided a much narrower and reliable calibration
range from 925-430 B.P. (Wilmshurst et al. 2011). Exclusion of unreliable
materials produced a shortened chronology and a two-phase sequence of
settlement
establishment in the Society Islands ~925-830 B.P. before a
widespread radiation towards the Marquesas, ~750-673 B.P.; Easter Island, ~750697 B.P.; Tokelau, 750-550 B.P.; Hawaii, ~731-684 B.P.; New Zealand, ~720-668
B.P.; Southern Cooks, ~700-669 B.P.; and the Line Islands, ~675-657 B.P.
(Wilmshurst et al. 2011; Petchey 2010). Recent radiocarbon dating research has
focused on selecting short-lived materials (Rieth et al. 2011; Molle & Conte 2011;
Kahn 2012), which have generally supported the model of Eastern Polynesian
expansion in the early second millennium A.D (after 950 B.P.).
A robust chronology and dispersal model for Eastern Polynesian expansion is
crucial in order to address initial colonisation events, social development and
interactions, trade, resource use and the anthropogenic impact on Pacific
ecosystems (Rieth et al. 2011). As discussed above, the use of radiocarbon dating
in conjunction with associated material assemblages often forms the basis for
dispersal models. However, these chronologies are heavily reliant upon finding
suitable samples from the earliest occupation layers across different sites.
Furthermore, coastal submergence can greatly affect the visibility of archaeological
sites (Dickinson & Green 1998). With many Polynesian islands lacking sustained
archaeological research, island groups may be inaccurately represented in
dispersal models.
9
1.2 Interaction spheres in post-settlement Polynesia
Following the initial settlement of the Polynesian Triangle, widespread voyaging
and interaction was evident within island groups and between remote
archipelagos (Rolett 2002). Early inter-island interactions may have been crucial
to widen marriage spheres, develop political and economic influence, and facilitate
the trade of plants and animals that some island populations lacked (Kirch 1988;
Clark et al. 2014). The unequal distribution of important tool materials, such as
basalt
a volcanic rock used for making adzes, may also have facilitated initial
trade (Best et al. 1992). High-precision methods such as X-ray fluorescence,
electron microprobe analysis and petrography have been used to assess the
composition and subsequently the source location of traded basalt in Polynesia.
These analyses have provided conclusive evidence for the regional trade of basalt
adzes, from the Austral Islands and Society Islands into Rarotonga and Mangaia
(Cook Islands) (Sheppard et al. 1997), Eiao (Marquesas Islands) into Moorea
(Society Islands) and Mangareva (Gambier Islands) (Weisler 1998), and Pitcairn
Island into Mangareva (Conte & Kirch 2004). Basalt and ceramic was also widely
exchanged between the Western Polynesian archipelagos of Fiji, Tonga and Samoa
(Clark et al. 2014; Dye & Dickinson 1996), with an eastern extension into the Cook
Islands (Walter 1998; Kirch 1995; Allen & Johnson 1997). However, artefact
transfer between Fiji, Tonga and Samoa was relatively low for the first 1500 years
post-colonisation, until 1200 B.P. when inter-archipelago basalt trading began to
increase in frequency, peaking around 800-900 B.P (Clark et al. 2014, Cochrane &
Reich 2016). Interestingly, this coincides with the expansion into Eastern
Polynesia. However, the maintenance of a single language (that eventually evolved
into Proto-Polynesian) in the Samoa-Tongan region for the first 1000 years postcolonisation does signify some level of continuous interaction between these two
archipelagos (Pawley 2015). A reduction in inter-archipelago interaction between
1700 B.P. and 1200 B.P. likely led to the regional development of the dialects
Tongic and Nuclear Polynesian.
Rolett (2002) postulated a model of interaction spheres across central-east
Polynesia, with prehistoric contact occurring between most islands, except for the
10
most isolated. Rolett argued that based on current evidence, a core interaction
zone existed between the Southern Cooks, Society Islands and Austral Islands.
Trade links also connected Samoa and Tonga with the Southern Cooks, indicating
interaction across the Western and Eastern Polynesian cultural divide. A smaller
interaction sphere existed between the Marquesas and the Society Islands, and
between Mangareva and the Pitcairn group on the eastern fringes of Polynesia
(Rolett 2002). Linguistic evidence also suggests that a secondary interaction
sphere may have existed between the Austral Islands, Mangareva (Gambier
Islands) and the Pitcairn islands to the eastern Tuamotus and the Marquesas
(Kirch 2000). However, given limited archaeological fieldwork in this region of
Eastern Polynesia, there has been no material analysis linking these islands to
support this claim.
An abrupt decline in imported materials reported in later deposits, indicated that
widespread trade in interaction spheres contracted after 500 B.P. (Rolett 2002). As
populations grew larger and more self-sufficient, the benefits of long-distance
trading may have grown smaller (Kirch 1988). Island deforestation as a result of
human settlement may also have limited the ability to construct voyaging canoes,
as the availability of timber declined over time. It is argued that this contributed to
the isolation of Mangareva (Weisler 1994) and Easter Island (Van Tilburg 1994).
Heightened hostilities between Tahiti and Moorea in the Society Islands (Haddon
& Hornell 1975) may have contributed to a breakdown of the Societies as a
regional hub in the core interaction sphere, which would have had indirect
repercussions on the outlying archipelagos (Rolett 2002). Ethnohistoric evidence
also suggests that inter-chiefdom hostilities were present between Marquesan
islands, constraining inter-island voyaging both within the Marquesas and with
other island groups (Robarts 1974).
The investigation of Polynesian interaction spheres and the subsequent
breakdown of these exchange networks after 500 B.P. is critical for understanding
the development, through both sustained interaction and isolation, of Polynesian
chiefdoms. However, like the chronological research on colonisation, the research
on Polynesian trade and interaction relies heavily upon archaeological fieldwork
11
and the subsequent retrieval of comparative material assemblages. The
commensal approach, utilising recent advances in genome sequencing and the
construction of comparative databases, is increasingly being applied to study
dispersal and interaction patterns across the Pacific.
1.3 The commensal approach
Molecular research has been applied to the study of Pacific origins and settlement
exhibited a high frequency of globin gene mutations, which was also found in
populations in Near Oceania. The first mitochondrial research in the Pacific
itochondrial DNA (mtDNA)
markers found in high frequency in Polynesia) within populations in Island
Southeast Asia (ISEA) (Melton et al. 1995; Redd 1995). This provided support for a
Polynesian origin in ISEA. Later research found that modern Polynesian
populations showed a high frequency of a Near Oceanic derived Y-chromosome
(that developed prior to the Austronesian intrusion) even in populations
exhibiting a near fixation of ISEA derived mtDNA lineages (Hage and Marck 2003;
Kayser et al. 2000; Kayser 2010). This indicated that substantial mixing between
the indigenous non-Austronesians and Austronesian-speaking people took place as
the latter moved into the Bismarck Archipelago. Hage and Marck (2003) argued
that the high frequency of an indigenous non-Austronesian Y-chromosome in
modern Polynesian populations is indicative of matrilocal residence patterns and a
matrilineal descent structure in the ancestral Austronesian-speaking people. It was
proposed that Indigenous non-Austronesian males would have moved in with their
s
that eventually moved into Remote Oceania. This is consistent with evidence of
matrilineal descent in some modern populations found across Micronesia and
Island Melanesia (Jordan et al. 2009; Allen 1984).
Polynesian populations are however, not well represented in many Pacific-wide
genetic studies. Technical and cultural concerns towards the sequencing, storage
and interpretation of indigenous human DNA in the Pacific have limited human
12
genetic research in this region. Subsequently this has led to the use of commensal
species (plant and animal species transported by humans) as a proxy for human
migration (Matisoo-Smith 2015). Introduced by the Lapita people, animal species
such as the dog, chicken, pig and the Pacific rat have been found in the earliest
archaeological deposits across Polynesia (Matisoo-Smith 2015). Plant species such
as kava, breadfruit, taro, yam, ti and the paper mulberry were also introduced
(Whistler 2009). Phylogeographic analyses of these commensal species have
provided invaluable information on geographic origins and dispersals in
conjunction with human movement across the Pacific (Matisoo-Smith 1994;
Matisoo-Smith et al. 1998, 2004; Matisoo-Smith & Robins 2008; Barnes et al. 2006;
Larson et al. 2007; Storey et al. 2007, 2010; Oskarsson et al. 2012; Roullier et al.
2013; Thomson et al. 2014; Greig et al. 2015).
The Pacific rat (Rattus exulans) was the first commensal species investigated in
relation to human migration in Oceania (Matisoo-Smith 1994). Argued to have
originated in Flores, Indonesia (Thomson et al. 2014), Pacific rats were
transported by ancestral Polynesians and are now widely distributed across the
region. Large deposits of skeletal remains in early Lapita sites signify that Pacific
rats may have been exploited as a food item, and therefore intentionally carried
across the Pacific. Unintentional transport as a stowaway however cannot be
excluded (Matisoo-Smith et al. 1998). Rattus exulans inability to interbreed with
the recently introduced European rats (Rattus rattus and Rattus norvegicus) makes
it an ideal species to study models of Lapita origins and dispersals, as its genetic
history has not been compromised. Furthermore, there is little evidence of
transportation in historic vessels; therefore it has been proposed that the modern
Pacific rat populations are direct descendants of the original rats carried by the
Polynesian colonists (Matisoo-Smith 1994). Conversely, the dogs, pigs and
chickens introduced by the Lapita peoples were the same species as those later
introduced by Europeans (Matisoo-Smith et al. 2004). Substantial admixture with
the European introduced populations obscures genetic patterns in modern
populations that are solely associated with the Lapita movement.
13
Initial research investigating genetic variation in 480bp (base pairs) of the
hypervariable mitochondrial control region in modern Pacific rats, identified a
central East Polynesian interaction sphere comprised of the southern Cooks and
Society Islands (Matisoo-Smith et al. 1998). This broad interaction sphere was
central to a northern (i.e. the Marquesas and Hawaii), and a southern Polynesian
sphere (i.e. Kermadec Islands and New Zealand). The authors suggested that this
likely reflects a central homeland region from which major Eastern Polynesian
islands were colonised. This signified that that Marquesas was not part of the
central homeland as previously suggested, and given the monophyletic nature of
the Marquesan samples in this study, was considered relatively isolated postcolonisation. The Marquesas was however a stepping-stone towards the
colonisation of Hawaii (Matisoo-Smith et al. 1998), which coincides with previous
linguistic and archaeological research (Green 1966; Elbert 1982; Allen 2014). The
appearance of non-Marquesan Pacific rat lineages in Hawaii also suggests multiple
introductions either from an early secondary colonisation or later trade links with
the central homeland. Within the southern sphere, there is evidence of multiple
introductions into New Zealand, both from the Kermadecs and directly from the
central homeland. The monophyletic nature of the Pacific rats sourced from the
Chatham Islands however, indicate long-term isolation post-colonisation from
New Zealand (Matisoo-Smith et al. 1998). This is also consistent with previous
linguistic and archaeological research (Harlow 1979; Sutton 1980).
A later phylogeographic analysis targeted 240bp of the mitochondrial control
region in both ancient and modern Pacific rats across the Pacific (Matisoo-Smith et
al. 2004). This study indicated the existence of three major control region
haplogroups; the first, an isolated group within Southeast Asia; the second, a
dispersed group from Southeast Asia to Near Oceania; and the third, a primarily
Remote Oceanic group (see Figure 1.3.). The isolation of Haplogroup I consisting of
samples from the Philippines, Borneo and Sulawesi is indicative of an interaction
sphere confined within Southeast Asia. This group has no relation to the Oceanic
haplogroups following an ancestral split. Haplogroup II is distributed across both
Southeast Asia and Near Oceania from the Philippines, through the Bismarck
Archipelago and into the Solomon Islands and Santa Cruz Islands. The distribution
14
Figure 1.3 A neighbour-joining tree and corresponding map depicting the
distribution of the mitochondrial control region haplogroups I, II, IIIA
and IIIB (Matisoo-Smith et al. 2004). Bootstrap values are presented on the
main branches. The map symbols are as follows: haplogroup I, filled circle;
haplogroup II, asterisk; haplogroup IIIA, open triangle; haplogroup IIIB, filled
triangle. Haplotype names are given on the radiating branches of the NJ tree.
15
of Haplogroup II is suggestive of an encompassing interaction sphere that
coincides with previous archaeological research indicating the existence of
obsidian trading, commensal species translocations and post-Lapita interactions in
this region (Flannery 1995; Kirch 1997; Friedlaender et al. 2002).
Haplogroup III has been largely found in Remote Oceanic populations and is
associated with the Lapita/Polynesian movement through Remote Oceania. Its
distribution extends from Vanuatu across to Central and East Polynesia, to Hawaii,
New Zealand, and from back migrations, into the Caroline and Marshall Islands.
The only western (Near Oceanic) samples belonging to Haplogroup III are from
Halamhera, Wallacea, which has been identified as the most likely point of origin
for this haplogroup. The lack of Near Oceanic samples in Haplogroup III was
surprising, given evidence for the Lapita movement through both Near and Remote
Oceania. The authors noted however, that their Near Oceanic samples largely came
from large islands (e.g. New Guinea and Bougainville) that were previously
occupied and therefore not targeted by the Lapita populations as they moved
through New Oceania. If rats of Haplogroup III exist in Near Oceania, it is likely that
they are residing on the smaller, offshore island regions that were uninhabited
pre-Lapita (Matisoo-Smith et al. 2004). Incomplete sampling may therefore have
undetected two major introductions into Oceania
the first from an initial pre-
Lapita introduction into Near Oceania, and the second with the Lapita movement
through offshore island regions in Near Oceania before a further dispersal into
Remote Oceania. Further research targeting islands in the Bismarck Archipelago
found some overlap between Haplogroup II and III (Matisoo-Smith & Robins
2008).
Haplogroup III is further divided into subgroups IIIA and IIIB, the latter derived
from IIIA. These subgroups overlap in Western Polynesia, however only subgroup
IIIB has been found in Eastern Polynesia (Matisoo-Smith et al. 2004). Samples from
the haplogroup IIIB either belong or are derived from the haplotype, R9, which has
been found across central East Polynesia. Unfortunately, given that most East
Polynesian samples adhere to this R9 haplotype, there has been limited success in
determining the dispersal route across this region. This was evident in a study on
16
ancient Pacific rat samples from Anakena, Easter Island (Barnes at al. 2006). All
samples presented the R9 haplotype and therefore could not indicate precisely
where the Easter Island population originated in Eastern Polynesia. However, the
limited variation in the ancient Easter Island samples did indicate a single or
limited introduction of the species, followed by extreme isolation (Barnes et al.
2006). Interestingly, the introduction of the Pacific rat onto Easter Island
ecologically devastated the island, which in turn is theorised to have contributed to
the isolation and cultural collapse of the Rapa Nui people (Hunt & Lipo 2012).
Other research of commensal species has provided an insight into the complexity
of human migration and commensal translocations across Southeast Asia and the
Pacific. Larson et al. (2007) found evidence for three introductions of pig, the first
involving the translocation of Sus celebensis to Flores, Timor and Sulawesi by
Neolithic settlers; the second involving the introduction of another species (Sus
scrofa) originating from mainland Southeast Asia and now found in Taiwan, the
Philippines and Micronesia; and the third, the translocation of an independent
Pacific lineage of Sus scrofa by the Lapita people across Near and Remote Oceania.
This Pacific lineage was traced to Vietnam. However, the absence of the Pacific
lineage in Taiwan, the Philippines, Borneo and Sulawesi, suggests that if the
model, the Austronesians did not initially transport this lineage. However it was
picked up along the way and carried by their Lapita descendants across the Pacific
(Larson et al. 2007).
Ancient mtDNA analyses investigating the translocation of dogs (Canis familiaris)
into the Pacific indicate at least two introductions, which are distinct from the New
Guinea Singing dog (NGSD) found in Papua New Guinea, and the dingo lineage that
was already established in Australia by 3500 B.P. (Savolainen et al. 2004; Milham
& Thomson 1976). Archaeological evidence of dogs in early Lapita contexts is
relatively rare, although they are found in many early sites across Central and East
Polynesia, except for Easter Island. Two distinct lineages are evident in ancient
Polynesian dogs - Arc1, which outside of Oceania is predominately found in
mainland and western ISEA; and Arc2, found across all of ISEA (Oskarsson et al.
17
2012; Savolainen et al. 2004). Given that neither of these haplotypes were found in
Taiwan or the Philippines, the translocation of these lineages must not have passed
through the northeastern route into the Pacific. The presence of a specific
haplotype found in ancient New Zealand dogs and in modern dogs largely from
Indonesia, provides evidence for a southwestern dispersal through Indonesia
(Greig et al. 2015). When and how these dogs dispersed across the Pacific however
will require greater sampling and phylogenetic resolution.
Chickens are a commensal species that have been found in the earliest Lapita sites
across both Near and Remote Oceania (Storey et al. 2010). Recent ancient DNA
(aDNA) analyses targeting a 150bp segment of the mitochondrial D-loop, have
found evidence for two introductions of chicken into the Pacific, from mainland
and Southeast Asia (Storey et al. 2007). Interestingly, this coincides with the two
introductions of dogs from similar locations in Southeast Asia. A chicken bone
uncovered in a Pre-Columbian archaeological site in El Arenal, Chile produced a
sequence identical to prehistoric chickens uncovered in Tonga and American
Samoa. The bone sample was radiocarbon dated to 622 ± 35 B.P., which in
conjunction with the genetic data provides firm evidence for a Polynesian
migration and subsequent translocation of chickens from Oceania to South
America (Storey et al. 2007). Thomson et al. (2014) indicated a strong reservation
towards this finding, claiming modern chicken contamination of the El Arenal
sample. Storey and Matisoo-Smith (2014) insist however that their samples had
not been compromised by contamination and that Thomson et al. (2014) did not
provide any proof to justifiably dispute their evidence.
Further evidence for prehistoric contact between South America and Polynesia is
in the widespread presence of sweet potato (Ipomoea batatas), a plant originating
from South America that has been found in many pre-European archaeological
sites across Polynesia. Targeting both chloroplast DNA and nuclear microsatellites
in modern and herbarium samples collected from the Americas, Oceania and
Southeast Asia, Roullier et al. (2013) provided strong support for a pre-historic
translocation of sweet potato from the Peru-Ecuador region of South America into
Polynesia. Whether the multiple lineages in prehistoric Polynesia are
18
representative of one of more introductions, is at the moment unknown. Roullier
et al. (2013) did note however that sharing of prehistoric clones between Eastern
Polynesian and the Western Pacific was quite rare, reflecting a cultural and
subsequent trading boundary between these regions. Conversely in the Western
Pacific, a lack of geographic structure suggests a wide circulation, presumably
facilitated by trade of several clones. Sharing of some clones within Eastern
Polynesia also indicate circulation within this region. Later European contact and
modern movements led to multiple new lineages that have recombined and
reshuffled the prehistoric sweet potato lineages across Polynesia (Roullier et al.
2013).
1.4 aDNA: its use and limitations
The field of ancient DNA (aDNA) is a relatively new area of molecular research,
increasingly applied in fields such as anthropology, zoology and plant biology. The
sequencing of ancient genetic material sourced from multiple geographic and
temporal populations has provided invaluable insights into past demographic
history and evolutionary processes (Slatkin & Racimo 2016; Ramakrishnan &
Hadly 2009). There are however, many technical issues that arise in working with
ancient DNA. Low preservation, contamination and post-mortem damage can
affect the amplification, sequencing and authenticity of genetic data. Improved
extraction/amplification methodologies and sequencing technologies are however
addressing many of the limitations that have plagued previous aDNA research.
1.4.1 Preservation
Ancient DNA refers to any DNA retrieved from preserved tissue (e.g. bones, teeth,
hair and skin) and other materials (e.g. seeds and pollen). DNA degradation in
preserved tissue is largely affected by age, density, environmental factors (e.g.
temperature fluctuation and acidity) and post-excavation procedures (Robins et al.
2001). Across the Pacific, Pacific rat samples recovered from older archaeological
sites (e.g. from sites in the Santa Cruz Islands, Tonga and Niue dated between
2000-3000 B.P.) have previously produced a low polymerase chain reaction (PCR)
amplification success rate for mtDNA (Robins et al. 2001). In comparison, samples
19
from contexts dated to within 1000 B.P. (e.g. Cook Islands, Hawaii, New Zealand
and Easter Island) have produced PCR success rates between 60-100%. The site
context (i.e. whether it is open, within a rock shelter, or cave) also influences aDNA
preservation, with protected sites, such as caves providing a stable temperature
and calcareous environment that limits the growth of microbial activity and
subsequent degradation of bone apatite (Nielsen-Marsh et al. 2000). Conversely,
open sites such as dirt mounds and dunes, in association with higher moisture,
acidity and temperature can lead to faster DNA degradation.
Pacific islands vary from tropical to temperate climates. All-year round warm-hot
temperatures and humidity in tropical parts of the Pacific suggests that DNA
preservation will be poor. However, there has been very little research in assessing
temperature at various depths in archaeological sites across the Pacific
which
with lower temperatures would be more conductive to DNA preservation (Robins
et al. 2001). Unfortunately, given that many archaeological sites within Remote
Oceania are situated within beach dunes (Robins et al. 2001), low temperatures
are likely meaningless given that water percolation from the nearby ocean
dissolves bone apatite, leading to the growth of microorganisms that metabolise
DNA (Bollongino et al. 2008). In conclusion, Robins et al. (2001) indicated that
protected sites (i.e. caves and rock shelters) from more temperate islands (e.g.
New Zealand, New Caledonia and Easter Island) have produced samples with the
highest success rate of DNA recovery, regardless of contextual age, but within the
last 4000 years. In tropical environments, both site protection and age have an
influence on DNA degradation, with a higher PCR success rate on samples from
rock shelters and caves, dated to within the last 2000 years. Samples recovered
from open and/or coastal environments produced the lowest PCR success rates
and are highly influenced by contextual age (Robins et al. 2001).
Post-excavation procedures also influence DNA degradation, as the environmental
conditions radically change, and in hot excavations, samples may be left exposed to
direct sunlight for hours (Smith et al. 2003). UV light can induce free radical
damage to DNA (Bollongino et al. 2008). An analysis on the effect of postexcavation treatments on DNA amplification indicated that washing, treatment
20
with chemicals and long-term storage at room temperature (typical in museum
collections) is detrimental to DNA preservation (Pruvost et al. 2007). Recently
excavated and unwashed samples provided twice as much authentic DNA than
samples that underwent post-excavation treatment (Pruvost et al. 2007). If
samples are to be stored long-term, they should be kept unwashed and ideally be
placed in a cold, dry environment, such as a freezer (Yang et al. 2005).
1.4.2 Contamination
The incorporation of microorganisms into a sample from the site environment,
touching of samples without gloves and washing of samples with tap water, both
on an excavation and in the laboratory, are the biggest sources of exogenous DNA
contamination. Well-preserved samples can be decontaminated in the laboratory,
however degraded samples such as porous bone and teeth, are more vulnerable to
the penetration of contaminants into the tissue structure itself (Bollongino et al.
2008). The hypersensitivity of PCR reactions whereby amplifying minute amounts
of DNA will indiscriminately amplify large amounts of contaminant DNA, can
potentially overwhelm the amplification of the authentic DNA template. A number
of guidelines are now widely implemented to minimise contamination in the field
and in the laboratory, and to decontaminate samples prior to DNA extraction in
order to reduce the amplification of exogenous PCR products.
The wearing of gloves and the changing of tools is encouraged when handling
material in the field to minimise human contamination (Yang et al. 2005). It is also
recommended that samples are no longer washed in the field to reduce
contamination from the water source. Sample decontamination and DNA
extraction must be conducted in a dedicated and isolated laboratory, away from
routine DNA amplifications, such as in a modern genetics laboratory. Gloves, face
masks, hair nets and in some laboratories
are routinely used in the laboratory to minimise contamination from the
researcher. There are a number of decontamination strategies to remove
exogenous DNA, such as removing surface layers, employing UV irradiation,
washing the sample in ethanol, bleach and/or purified water. The immersion of
21
samples into 6% bleach for a period of 15min has been assessed as the most
practical and cost-effective method for removing surface contamination (Kemp &
Smith 2005). The inclusion of blanks (a control treated as a real sample) should be
processed in conjunction with extraction and amplification steps to evaluate crosscontamination. Independent replication is also necessary to discount intralaboratory contamination (Cooper & Poinar 2000).
1.4.3 DNA Damage
Enzymatic repair mechanisms that maintain genome stability in living cells, cease
to function after death (Lindahl 1993). This leaves DNA vulnerable to strand
fragmentation by endonuclease activity and microorganisms that inhabit decaying
tissue. Successful sequencing of samples aged between 4 to 13,000 years generally
shows DNA fragmentation of between 40-500 bp (Paabo 1989). Preservation in a
cold environment can inhibit substantial destruction by these degradative
processes, however other chemical processes, such as hydrolytic deamination and
oxidative damage can impair the integrity and the ability to amplify and sequence
aDNA (Dabney et al. 2013).
Hydrolytic damage to both phosphodiester and glycosidic bonds can result in
single-stranded nicks and abrasic sites where the base has been de-attached from
the sugar-phosphate backbone (Hofreiter et al. 2001). This in turn, can lead to
greater strand breakage and aDNA fragmentation, which is the prominent form of
aDNA damage (Mitchell et al. 2005). The effect of hydrolytic deamination (i.e. the
removal of amine groups) of cytosine bases, is commonly found on the ends of
aDNA fragments. Deamination of cytosine produces a uracil base that during DNA
replication will incorporate adenine (A), an analogue of thymine (T). This will
result in ancient DNA fragments characterised by transitions of cytosine (C) to
thymine (T) and on the complementary strand, guanine (G) to adenine (A)
(Dabney et al. 2013). C to T transitions can originate in hot spots across both ends
of ancient DNA library preparation, as the exonucleolytic activity of T4 DNA
22
own C to T transitions) are removed and the new ends produced during blunt end
complemented by a G to A transition after rebuilding
complementary strand (Sawyer et al. 2012; Briggs et al. 2007; Brotherton et al.
2007). The development of miscoding lesions from hydrolytic deamination alters
the DNA template, leading to nucleotide misincorporation during amplification and
sequencing.
Oxidative damage, alkylation and Maillard reactions (between reducing sugars and
amino groups) can also lead to the development of nucleotide modifications and
crosslinks, formed between DNA strands or different DNA fragments. These
blocking lesions obstruct DNA polymerases, preventing amplification and
subsequent sequencing (Dabney et al. 2013; Willerslev & Cooper 2005). This can
lead to a proportionally larger amount of amplified contaminant DNA over
endogenous DNA.
1.4.4 New methodologies
The
initiated early aDNA research as it became possible for the first time to target and
amplify small, specific aDNA sequences (Golenberg et al. 1990; DeSalle et al 1992).
Many of these early studies were however later disregarded, due to the
unregulated amplification of exogenous contaminants (Rizzi et al. 2012). Stricter
protocols to minimise contamination during both sampling and processing in the
laboratory (Poinar & Cooper 2000), and in assessing the authenticity of sequences
in bioinformatics processing have since been widely implemented. Ancient DNA is
usually quite short and fragmented, constraining amplification to fragments no
larger than 500bp (Willserslev & Cooper 2005). This is quite small, considering
that the human mitochondrial and nuclear genomes are approximately 16,500 bp
and 3x109 bp, respectively. The traditional aDNA methodology involved amplifying
23
small target fragments (typically up to 200bp), generating several clones for each
amplified fragment (applying Sanger sequencing) and meticulously aligning and
comparing any overlap from different clones to construct a finalised consensus
sequence (Rizzi et al. 2012). The recent emergence of second-generation
sequencing however has revolutionised the field of ancient DNA. Secondgeneration sequencing in conjunction with developments in bioinformatics
processing have made it possible to simultaneously generate billions of reads from
multiple samples in parallel, which are then overlapped and assembled into larger
DNA sequences, i.e. complete mitochondrial genomes or nuclear genes (Knapp &
Hofreiter 2010). Second-generation sequencing was first implemented in ancient
DNA research in 2006, with the generation of a 13 million bp nuclear sequence
from an extinct woolly mammoth (Poinar et al. 2006). Prior to the incorporation of
this sequencing technology, the largest ancient dataset obtained was a 27,000 bp
nuclear sequence from a Pleistocene cave bear (Noonan et al. 2005).
Another increasingly applied methodology in ancient DNA research, is the use of
hybridisation capture for target DNA enrichment prior to second-generation
sequencing (Ng et al. 2009). This new method, limiting the overuse of PCR
amplification (which can amplify large amount of exogenous contaminant DNA),
involves hybridising target DNA fragments to a known genomic region
(alternatively referred to as bait) that has been immobilised on magnetic,
streptavidin covered beads. After washing (to remove any non-target libraries e.g.
contaminants) the target libraries are eluted from the streptavidin beads and
amplified using ligated adapters, prior to sequencing. If libraries are to be pooled
for multiplex sequencing (simultaneous sequencing of multiple samples), then a
barcode (unique sequencing tag) can be added to each sample during library
preparation. The sequencing reads pertaining to each individual sample can then
be identified during the bioinformatics processing (Knapp & Hofreiter 2010).
Second-generation sequencing can provide direct measures of deamination,
depurination, fragmentation and blocking lesions, contributing new insights into
degradation and providing a means to authenticate aDNA (Overballe-Peterson et
al. 2012; Dabney et al. 2013). The development of polymerase extension profiling,
24
using the high-throughput of second generation sequencing, can identify the
locations of blocking lesions, such as crosslinks (Heyn et al. 2010), providing
valuable information on biochemical processes in aDNA. Second-generation
sequence datasets can be analysed for aDNA deamination patterns using the
program mapDamage 2.0 (Jonsson et al. 2013), which provides a quantitative
estimate of damage parameters. The appearance of cytosine deamination,
characterised by transition
across sequence reads and can be used in conjunction with other measures to
authenticate aDNA. Previous research has recorded a maximum rate of 0.05 for
misincorporation within modern samples (<117 years of age) and a minimum
frequency of 0.11 for samples older than 500 years (Sawyer et al. 2012). Increasing
coverage during sequencing or applying enzymatic treatments to deaminated
residues during blunt-end repair can reduce nucleotide misincorporation as a
result of deamination (Briggs et al. 2010; Overballe-Peterson et al. 2012). Direct
estimates of exogenous contamination can also be calculated by mapping acquired
reads to reference genomes (i.e. from humans, pigs, chickens, cows) (Green et al.
2009). A lack of contamination in conjunction with sequences exhibiting typical
aDNA damage patterns can be used to authenticate endogenous DNA sequences.
Third-generation sequencing technologies performing single-molecule DNA
sequencing will likely be widely utilised in the near future. Unlike secondgeneration technologies, single-molecule DNA sequencing does not require the
preparation of DNA libraries (that can increase the level of nucleotide
misincorporations) and PCR amplification (that can unintentionally amplify
contaminant DNA) (Orlando et al. 2011; Derrington et al. 2010; Pushkarev et al.
2009). Preliminary research has indicated a higher yield of endogenous DNA using
single-molecule DNA sequencing on Helicos HeliScope platforms, than using
second-generation technologies on Illumina sequencing platforms (Orlando et al.
2011). It is anticipated that the development of third-generation sequencing will
outcompete current technologies, providing greater yield, sensitivity and accuracy
of aDNA sequencing.
25
1.5 Research Objectives
The main aim of this thesis was to investigate the origins and dispersals of Eastern
Polynesian people using complete ancient and modern Pacific rat (Rattus exulans)
mitochondrial phylogenies. Hybridisation capture, second-generation sequencing
and subsequent bioinformatics processes were applied to recover complete
ancient Pacific rat mitochondrial sequences, obtained from early archaeological
deposits across Western and Eastern Polynesia. A modern complete mitochondrial
Pacific rat database was also generated from contemporary tissue samples
collected across Eastern Polynesia.
Specific objectives were to:
1. Assess whether complete mitochondrial sequencing provides a greater
molecular resolution (than control region analyses) to distinguish island
populations/lineages across Eastern Polynesia.
2. Examine
how the distribution
of prehistoric
mtDNA haplotypes
corresponds with archaeological and linguistic evidence of human dispersal
across Polynesia.
3. Identify the immediate origin of Easter Island Pacific rats.
4. Assess whether modern Pacific rat populations are representative of
prehistoric populations.
5. Evaluate prehistoric Pacific rat DNA preservation across Polynesia.
26
2 Methods
2.1 Modern DNA sequencing
2.1.1 Sample Collection
A total of 97 modern Pacific rat tissue samples from the Society Islands, Cook
Islands and Samoa were amassed for mitochondrial genome extraction and
sequencing (see Table 2.1; Figure 2.1). These samples were collected by Lisa
Matisoo-Smith, James Russell, Lucie Falquier, Gerald McCormack and Terry Hunt
between 1992-2009. Rats were trapped using basic snap-type rat traps. Heart,
liver and tail samples were removed and either immersed in ethanol at room
temperature or placed within a -80°C freezer for long-term storage.
Table 2.1 Tissue samples collected for modern Pacific rat DNA processing.
Location
Island
Society Islands
Huahine
Date of
Collected by:
collection
1993
Lisa Matisoo-Smith
Tahiti
1993
Lisa Matisoo-Smith
7
Raiatea
1993
Lisa Matisoo-Smith
16
Tetiaroa
2009
Aitutaki
1992
Gerald McCormack
4
Pukapuka
1992
Gerald McCormack
7
Atiu
1992
Gerald McCormack
3
Apia, Upolu
1992
Terry Hunt
3
Cook Islands
Samoa
James Russell and
Lucie Falquier
(n)
21
36
97
TOTAL
27
!
Figure'2.1'Geographical'distribution'of'modern'and'ancient'samples'collected'for'DNA'processing.!Blue!and!green!filled!
triangles!denote!modern!and!ancient!samples,!respectively.!
!
28!
2.1.2 DNA Extraction
The extraction and library preparation of modern Pacific rat samples was
conducted in a modern DNA laboratory in the Anatomy Department at the
University of Otago. DNA was extracted from heart and liver tissue samples using a
MagJET Genomic DNA Kit (Thermo Fisher Scientific, Inc), according to the
manufacturers instructions. Purified DNA samples were quantified using a
NanoDrop 2000 (Thermo Fisher Scientific, Inc) and stored at 4°C until further use.
2.1.3 PCR amplification
Whole mitochondrial genomes were amplified using a modified PCR protocol; the
original outlined in Robins et al. (2010). Four overlapping fragments, ranging from
4-6kb in length, were amplified separately using targeted primers (see Table 2.2a
and Figure 2.2). Long-range polymerase chain reactions (PCR) were conducted
using a KAPA LongRange HotStart PCR Kit. PCR amplification reactions in 30 l
contained: 1x KAPA LongRange Buffer (without Mg2+ ); 1.75-3mM MgCl2; 0.3mM
dNTPs; 0.5 M of each forward and reverse primers; 0.75U KAPA LongRange
HotStart DNA Polymerase; and 2 l of DNA template. The following amplification
protocol was used: 94°C, 3 min, followed by 10 cycles of: 94°C, 30 s; various
annealing temperatures (47°C - 50°C) (see Table 2.2a), 30s; 68°C, 8min, 25 cycles
of: 94°C, 30 s; various annealing temperatures (47°C - 50°C), 30s; 68°C, 8min with
a 10 s time increment per cycle, followed by 1 cycle of 72°C, 10 min and a hold at
10°C. Amplification of PCR products was analysed by electrophoresis on a GelRed
(Biotium) stained 1% agarose gel using a KAPA Universal DNA Ladder (KAPA
Biosystems) for comparison. PCR products that successfully amplified were
purified using Agencourt AMPure XP beads (Beckman Coulter Inc) and quantified
using a NanoDrop 2000 (Thermo Fisher Scientific, Inc).
2.1.4 Library preparation & sequencing
Double-stranded barcoded libraries were constructed from the modern PCR
fragments for paired end sequencing on an Illumina MiSeq platform (San Diego,
CA, USA). Modern libraries were built to contain primer-binding sequences, a
29
Table 2.2 a) Long-range PCR amplification for modern samples. b) Long-range PCR amplification for bait.
a)
Region amplified
12S
COI
COI
COIII
COIII
Cyt b
b)
Cyt b
16S
Region amplified
COI
Cyt b
Cyt b
COI
Primer pairs
Av175312SF
AAACTGGGATTAGATACCCCACTAT
& R6036R
ACTTCTGGGTGTCCAAAGAATCA
BatL5310
CCTACTCRGCCATTTTACCTATG
& LR2RB
CTGATTGGAAGTCAGTTGTATTTTT
L10647F
TTTGAAGCAGCAGCCTGATAYTG
& RCb9H
TACACCTAGGAGGTCTTTAATTG
RGlu2L
CAGCATTTAACTGTGACTAATGAC
& Long
16SR
TGATTATGCTACCTTTGCACGGTCAGGATACC
Primer pairs
BatL5310
CCTACTCRGCCATTTTACCTATG
& RCb9H
TACACCTAGGAGGTCTTTAATTG
RGlu2L
CAGCATTTAACTGTGACTAATGAC
& R6036R
ACTTCTGGGTGTCCAAAGAATCA
30
Amplicon
Annealing
length
temperature
5.5 kb
50°C
4.1 kb
47°C
5.5 kb
50°C
4.2 kb
50°C
Amplicon
Annealing
length
temperature
9.5 kb
47°C
8.3 kb
49°C
!
Figure' 2.2! Targeted' overlapping' fragments' for' PCR' amplification.! The! red!
arrows! indicate! where! four! overlapping! fragments! were! amplified! for! modern!
mitochondrial!DNA!sequencing.!!The!blue!arrows!indicate!where!two!overlapping!
fragments!were!amplified!as!part!of!the!R.#norvegicus!bait!preparation,!for!ancient!
hybridisation!capture.!
!
barcode! sequence! for! sample! identification! in! bioinformatics! processing,! and!
adapter!sequences!that!will!bind!to!the!MiSeq!flow!cell.!For!each!sample,!the!four!
overlapping!PCR!fragments!were!pooled!into!equimolar!ratios!with!the!addition!of!
PCR! grade! water! up! to! 100μl.! Pooled! DNA! fragments! were! then! sonicated! for! 9!
cycles! of! 15sec! on,! 45sec! off,! using! a! Picoruptor! (Nippongene).! Sonication! was!
necessary! to! break! the! longMrange! DNA! fragments! into! smaller! sizes,! prior! to!
!
31!
further processing. DNA ends were blunted using T4 DNA Polymerase (New
England BioLabs Inc), phosphorylated using T4 Polynucleotide Kinase (New
England BioLabs Inc) and purified using Agencourt AMPure XP beads (Beckman
Coulter Inc), in preparation for A-tailing and adapter ligation. Fragments
underwent A-tailing using the enzyme Klenow (New England BioLabs Inc), which
subsequent ligation of adapters. DNA fragments were once again purified using the
AMPure XP beads prior to adapter ligation. Illumina compatible double stranded
adapters (ds_AdapterP5 and ds_AdapterP7) were prepared by hybridising each
d reverse complementary strands, using the following
thermal profile: 95°C, 10s; ramp to 14°C (-0.1°C/s). The adapter sequences
(ds_AdapterP5 and dsAdapterP7) were then ligated to the ends of fragments,
followed by purification using the AMPure XP beads. A high fidelity PCR was
conducted using a KAPA HiFi PCR Kit (KAPA Biosystems). PCR amplification
reactions in 30 l contained: 1x KAPA HiFi Buffer; 0.3mM dNTPs; 0.3 M of the
primer P5_extension; 0.3 M of a barcoded index primer P7_extension that
contains a short, unique sequence allowing sample identification in subsequent
bioinformatics processing; 0.6U KAPA HiFi DNA Polymerase; and 19 l of DNA
template. The following amplification protocol was used: 94°C, 3 min, 20 cycles of:
94°C, 25 s; 60°C, 15s; 72°C, 15s; followed by a final extension of 72°C for 15 min.
This was followed by another purification using the AMPure XP beads. Libraries
were quantified using a Qubit 2.0 Fluorometer (Invitrogen) and pooled in
equimolar amounts (40ng of DNA per library). Libraries were sequenced using 2 x
250 base pair-end runs on an Illumina MiSeq platform, conducted by New Zealand
Genomics Limited (NZGL) at Massey University, New Zealand.
2.2 Ancient DNA sequencing
2.2.1 Sample Collection
A total of 77 ancient Pacific rat archaeological samples from Tonga, Tokelau,
American Samoa, Easter Island, Austral Islands, Cook Islands and the Society
Islands were amassed for mitochondrial genome extraction and sequencing (see
Table 2.3; Figure 2.1). These samples were collected by Dave Burley, Simon Best,
32
Table 2.3 Archaeological samples collected for ancient mitochondrial DNA processing. Age of material or associated
dates: 1 = Present 500 B.P., 2 = 500 1000 B.P., 3 = 1000 2000 B.P., 4 = 2000 3000 B.P. * refers to unknown.
Location
Site, Island
Age
Site type
Date of
collection
Collected by:
Excavation report
(n)
Tonga
Pukotala
4
Open
1997
Dave Burley
Burley et al. 1999
2
Faleloa, Foa
4
Open
1997
Dave Burley
Burley et al. 1999
2
Mele Havea,
4
Open
1997
Dave Burley
Burley et al. 1999
1
Fakaofo
2
Open
1986
Simon Best
Best 1988
6
Ofu Village, Ofu
4
Open (coastal)
2012
Jeffrey T. Clark
Quintus et al. 2015
6
Olosega Village,
Olosega
4
Open
2015
Jeffrey T. Clark
Unpublished
6
Anakena
2
Open (coastal)
2004
Terry Hunt
Hunt & Lipo 2006
11
2
Open (coastal)
2007
Robert Bollt
Tokelau
American
Samoa
Easter Island
Austral Islands Atiahara, Tubuai
33
Worthy & Bollt
2011
5
Location
Site, Island
Age
Site type
Date of
collection
Collected by:
Cook Islands
Ureia, Aitutaki
3
Open
1987
Melinda Allen
Mangaia
*
Open (coastal)
*
Kazumichi
Katayama
Leach et al. 1994
4
Terre Teahutapu,
,
Maupiti
1
2012
Jennifer Kahn
Kahn et al. 2015
14
Terre Namotu,
Maupiti
1, 2
2012
Jennifer Kahn
Kahn et al. 2015
4
Haranai Valley,
Maupiti
1, 2
Open
2012
Jennifer Kahn
Kahn et al. 2015
4
Haumi
archaeological site,
Moorea
1, 2
*
2014/15
Jennifer Kahn
Unpublished
6
Society
Islands
Open
(coastal)
Open
(coastal)
TOTAL
Excavation report
Allen & Steadman
1990
(n)
6
77
34
Jeffrey T. Clark, Terry Hunt, Robert Bollt, Melinda Allen, Kazumichi Katayama and
Jennifer Kahn from archaeological excavations between 1986 and 2015. These
archaeological samples were unwashed and placed within a dark and dry
environment for long-term storage.
2.2.2 DNA Extraction
All aDNA extractions and library building prior to PCR amplifications was
conducted in a purpose-built Ancient DNA laboratory in the Richardson Building at
the University of Otago. Access and use of the aDNA laboratory requires adherence
to strict protocols in place to minimise contamination risks, particularly from areas
where amplified DNA is a present, i.e. modern DNA laboratory. Femurs were
common in the acquired Pacific rat bone assemblage (see Figure 2.3) and were
targeted for DNA extraction.
Figure 2.3 Archaeological sample of Pacific rat (R. exulans) femur.
35
Femurs are the largest bone in the Pacific rat skeleton, weighing on average 0.1g
(Matisoo-Smith et al. 1997). Given it is not appropriate to pool bone samples in
order to prevent more than one individual from being analysed, the femur
provided the best chance of retrieving preserved DNA from an individual rat. Prior
to DNA extractions, bone samples were soaked in 5% bleach for 10 minutes, rinsed
several times with ultrapure water and left to dry overnight. Each bone sample
was ground into a fine powder using a sterile mortar and pestle and digested
overnight in an extraction buffer containing 0.45M of EDTA and 0.25mg/ml of
Proteinase K (Rohland & Hofreiter 2007). Samples were processed in sets of nine
in conjunction with one extraction blank. Extraction blanks were processed
alongside samples throughout extraction and library preparation, and sequenced
in a separate run consisting of all blanks. This was necessary to identify any crosscontamination between samples that would jeopardise the integrity of each
sequence in a set.
Following the overnight digestion, the supernatant from each digested sample
(and blank) was added to 2.5ml of binding buffer and then combined with 100 l of
silica suspension. Samples were rotated for 3hrs in the dark at room temperature
to bind the DNA to the silica surface. After binding, the silica was extracted,
repeatedly cleaned with a wash buffer to remove PCR-inhibiting agents and DNA
eluted in 1x TE buffer. Each sample was then spun down and the supernatant,
containing the DNA extract, was transferred to siliconised tubes and stored at 20°C (Rohland et al. 2010).
2.2.3 Library preparation
Double-stranded barcoded libraries were generated from aDNA extracts (Knapp et
al. 2012) in preparation for hybridisation capture and paired end sequencing on an
Illumina MiSeq platform (see Figure 2.4a). Blunt-end repair was performed using
T4 DNA Polymerase (New England Biolabs Inc) and phosphorylated using T4
Polynucleotide Kinase (New England Biolabs Inc), allowing for subsequent adaptor
ligation. Post-blunt end repair, DNA samples were purified with a MinElute PCR
36
!
!
!
Figure' 2.4! Overview' of' the' hybridisation' capture' on' beads' method.! a)! The!
production! of!indexed! (barcoded)! libraries!from! individual!bone!samples.!b)!The!
production!of!immobilised!bait!using!a!fresh!tissue!sample!from!the!species!Rattus&
norvegicus! c)' Hybridisation! of! indexed! libraries! onto! baited! beads.! Captured!
libraries! are! then! washed! before! being! denatured! from! the! beads,! eluted! and!
amplified!prior!to!multiplex!sequencing.!
!
37!
Purification Kit (QIAGEN), with the following modification: DNA eluted in 0.1x TE +
0.05% Tween.
Illumina compatible adaptors, Sol_adap_P5 and Sol_adap_P7-BIO, were ligated to
the ends of the target sequences and purified with the MinElute PCR Purification
Kit (QIAGEN) with the following modifications: 2x PE wash steps and elution of
DNA in 25 l of 1xTE. The Sol_adap_P7-
nd,
allowing the DNA library to immobilise onto streptavidin-coated beads. The beads
were then washed to remove unincorporated adaptors, before being re-suspended
in a master mix containing Bst Polymerase, an enzyme that fills in any nicks to
complete adaptor attachment. A quantitative PCR (qPCR) was then performed to
determine the number of cycles required for the sufficient amplification of each
library in the immortalisation step. qPCR reagents and target libraries were
prepared in the ancient DNA lab, before being transferred to the modern DNA lab
for PCR amplification. Each qPCR amplification reaction in 25 l contained: 1 l of
adaptor ligated library; 10 M each of the universal adaptors, Sol_quant_P5 and
Sol_quant_P7; and 12.5 l of SYBR Green Master Mix. The following qPCR
conditions were used: 95°C, 10 min, 40 cycles of: 94°C, 30 s; 58°C, 30s; 72°C, 1min,
followed by a final extension at 72°C, 10 min on a Stratagene MxPro 3000P. Each
qPCR amplicon was then visualised by electrophoresis on a GelRed (Biotium)
stained 1% agarose gel using a KAPA Universal DNA Ladder (KAPA Biosystems)
for comparison to check the success of library preparation and for the presence of
adapter dimers. If libraries presented adaptor dimers, a second attempt at library
preparation was made or they were removed from further processing.
Preparation of libraries for immortalisation was conducted in the ancient DNA
laboratory, before being transferred to the modern DNA laboratory for PCR
amplification. Each immortalisation amplification reaction in 50 l contained: 19 l
of adaptor ligated library; 1xTaq Buffer; 2.5mM MgCl2; 1mM of dNTPs; 3.75U of
AmpliTaq Gold polymerase (ThermoFisher Scientific); 0.2 M of extension primer,
Sol_ext_P5; and 10 M of a barcoded index primer, Sol_ext_P7. The following
immortalisation conditions were used: 95°C, 12 min, # cycles (determined by
when the plateau was reached during the qPCR) of: 94°C, 30 s; 58°C, 30s; 72°C,
38
1min, followed by a final extension of 72°C for 10 min. Following immortalisation,
samples were purified with a MinElute PCR Purification Kit (QIAGEN) with the
following modification: 2x PE washes. Immortalised libraries were then amplified
using a KAPA HiFi PCR Kit (KAPA Biosystems) in preparation for the capture, each
50 l reaction containing: 1 l of immortalised library; 1x KAPA HiFi Buffer; 0.3mM
dNTPs; 0.3mM each of amplification primers, Sol_amp_p5 and Sol_amp_p7; and 1U
of KAPA HiFi DNA Polymerase. The following amplification conditions were used:
94°C, 5 min, 10 cycles of: 94°C, 20 s; 55°C, 55s; 72°C, 15s, followed by a final
extension of 72°C for 5 min. Following amplification, samples were purified with a
MinElute PCR Purification Kit (QIAGEN) with the following modification: 2x PE
washes.
2.2.4 Bait production & preparation
Baited beads were produced for hybridisation capture (see Figure 2.4b) using
tissue from a laboratory rat (Rattus norvegicus), sourced from the Department of
Anatomy, University of Otago. A preserved tissue sample of an R. exulans, originally
sourced from Near Oceania, was initially used for bait production however the
sample provided heavily degraded mitochondrial DNA that could not be amplified
in two overlapping fragments. The R. norvegicus sample was therefore used to
mass-produce the 8-9.5kb overlapping fragments. DNA extractions for the bait
production
methodology. Whole mitochondrial genomes were amplified using a modified PCR
protocol as outlined in the modern DNA amplifications above. The mitochondrial
genome was amplified in two overlapping fragments using the following targeted
primers (see Table 2.2b and Figure 2.2). PCR products were analysed by
electrophoresis and if successfully amplified, were purified with a MinElute PCR
Purification Kit (QIAGEN) and quantified using a NanoDrop 2000 (Thermo Fisher
Scientific, Inc).
Bait preparation was conducted following the protocol detailed in Maricic et al.
(2010). The two long-range fragments were pooled into equimolar amounts and
sonicated for 9 cycles of 15sec on, 45sec off, using a Picoruptor (Nippongene).
39
Sonication was necessary to break the long-range DNA fragments into smaller
sizes, prior to further processing. Sonicated bait was then aliquoted into small PCR
Each bait sample underwent blunt end repair followed by purification with a
MinElute PCR Purification Kit (QIAGEN). Each blunt-ended sample was then
-T/B adapters, purified with a MinElute PCR Purification
Kit (QIAGEN) and quantified with a NanoDrop 2000 (Thermo Fisher Scientific).
Five hundred nanograms of Bio-T/B adapter-ligated bait DNA was then aliquoted
into PCR tubes and denatured at 98°C. Bait DNA was combined with purified
streptavidin beads and rotated at room temperature for 20 minutes to allow
binding. Post-binding, the baited beads were collected and re-suspended in TET
buffer until ancient samples were ready for hybridisation capture.
2.2.5 Hybridisation capture & sequencing
Ancient libraries were captured using a modified hybridisation protocol, the
original outlined in Maricic et al. (2010). A hybridisation mixture containing: 2 g
of barcoded library; 1 M each of the blocking oligonucleotides: BO_p5_f, BO_p5_r,
BO_p7_f(1), BO_p7_r(1), BO_p7_f(2), BO_p7_r(2); 0.6x Agilent blocking agent; and
0.6x Agilent hybridisation buffer was prepared for each sample and treated to the
following conditions: 95°C, 3 min; 37°C, 30 min. The hybridisation mixture was
then added to the baited beads and rotated at 12rpm for 48hrs in a 65°C
hybridisation oven (see Figure 2.4c). Following the capture, the baits beads, now
containing captured libraries, were isolated and washed as described by Maricic et
al. (2010), before being re-suspended in 15ul of 1x TE solution and heated at 95°C,
3 min, denaturing the captured libraries from the beads. The supernatant
containing the captured libraries was transferred to a new siliconised tube and
stored at -20°C.
Captured libraries were then re-amplified using a KAPA HiFi PCR Kit (KAPA
15 l of post-captured library; 1x
KAPA Buffer; 0.3mM dNTPs; 0.3mM each of the amplification primers, Sol_amp_p5
and Sol_amp_p7; and 1U of KAPA HiFi DNA Polymerase. The following
40
amplification conditions were used: 94°C, 5 min, 20 cycles of: 94°C, 20 s; 55°C, 55s;
72°C, 15s, followed by a final extension of 72°C for 5 min. Following amplification,
samples were purified with a MinElute PCR Purification Kit (QIAGEN), quantified
Qubit 2.0 Fluorometer (Invitrogen) and stored at -20°C. Samples were pooled in
equimolar concentrations per run, resulting in 80-150ng of each library being
sequenced. Libraries were sequenced using 2 x 75 base pair-end runs on Illumina
MiSeq, conducted by New Zealand Genomics Limited (NZGL) at Massey University,
New Zealand. A negative run containing all the blank samples was also sequenced.
2.3 Bioinformatics processing
2.3.1 Modern DNA processing
The raw reads from the modern samples were mapped to a reference genome (R.
exulans, GenBank NC_012389.1) (Robins et al. 2008) in BWA v.0.7.12 (Li & Durbin
2009), implementing the maximal exact matches (MEM) command (Li 2013). PCR
(http://broadinstitute.github.io/picard/). Variant calling was conducted using the
GATK Haplotype Caller (McKenna et al. 2010) and the subsequent variant call file
(vcf)
converted
to
a
(https://github.com/theboocock/ancient_dna_pipeline,
FASTQ
Boocock
file
2014)
for
phylogenetic analyses.
To assess the impact of contamination from laboratory reagents, reads were also
mapped to mitochondrial reference genomes from cow (Bos Taurus, GenBank
accession NC_006853.1), pig (Sus scrofa, NC_0012095.1), human (Homo sapiens,
GenBank NC_012920.1), chicken (Gallus gallus, GenBank NC_001323.1) and dog
(Canis lupus familiaris, GenBank NC_002008.4). A contamination ratio was
calculated by the ratio of all mapped reads to a reference (quality > =20) in
comparison to those mapping to the R. exulans mitochondrial genome. Sequencing
coverage was evaluated in Geneious (v.8.1.8) (http://www.geneious.com, Kearse
et al. 2012) and only sequences exhibiting more than 90% coverage were retained.
41
2.3.2 Ancient DNA processing
The raw reads from the ancient samples were initially processed in
AdapterRemoval (v.2) (Lindgreen 2012) to remove adapters, merge paired-end
fragments (overlapping by at least 11 base pairs), and remove stretches of Ns,
bases with low quality scores (<30) and short reads (<25). Reads were then
mapped to a reference genome (R. exulans, GenBank KJ530564.1) (Tsangaras et al.
2014) in BWA v.0.7.12 (Li & Durbin 2009), following recommended settings for
ancient DNA, i.e. n 0.03 (allow for more substitutions), -o 2 (allow more gaps at
the beginning) and l 1024 (deactivate seed mapping) (Schubert et al. 2012). As
outlined previously, reads were also mapped to the mitochondrial genomes of cow,
human, pig, chicken and dog to calculate a contamination ratio for each sample.
PCR dup
MarkDuplicates tool and from merged reads using script from PaleoMix (Schubert
et al. 2014). Merged and unmerged reads were combined into one BAM file for
each sample using Samtools (Li et al. 2009). The program mapDamage (v.2.0)
(Jónsson et al. 2013) was then implemented to assess damage patterns and lower
characteristic aDNA damage patterns were produced for each sample. INDEL
realignment was conducted using the GATK Best Practices (McKenna et al. 2010).
A variant call file was generated and a consensus sequence produced, that was
SN
Viewer (IGV) (Robinson et al. 2011; Thorvaldsdóttir et al. 2013). Sequencing
coverage was evaluated in Geneious (v.8.1.8) (http://www.geneious.com, Kearse
et al. 2012) and only sequences exhibiting more than 80% coverage were used in
subsequent phylogenetic analyses.
2.4 Phylogenetic Analysis
Prior to the phylogenetic analyses of the complete mitochondrial genomes, a
control region analysis was conducted to categorise each sample into the three
major control region haplogroups
I, II, III (A & B) as described in Matisoo-Smith
42
et al. (2004). Complete modern and ancient mitochondrial sequences were aligned
to four 203bp control region fragments (from position 15,392
15,595) using
MUSCLE (Edgar 2004) in Geneious (v.8.1.8) (Kearse et al. 2012). These four
control region fragments were obtained from GenBank (AY604204, AY604209,
AY604231, AY604228; Matisoo-Smith et al. 2004) and represent haplogroups I
(haplotype R1), II (R2), IIIA (R15) and IIIB (R9). Three publicly available complete
mitochondrial Pacific rat sequences sampled from Thailand, Papua New Guinea
and New Zealand (Genbank EU273709.1, EU273710.1, EU273711.1; Robins et al.
2008) were also added to the alignment, as they are analysed in conjunction with
the Polynesian assemblage in downstream complete mitochondrial phylogenetic
analyses. After the full mitochondrial sequence alignment, base positions outside
of 15,392 15,595 were removed to export a complete 203bp sequence alignment.
A median-joining network was generated in PopArt (Leigh & Bryant 2015; Bandelt
et al. 1999). However as PopArt removes variable sites where one or more
sequences provide alignment gaps or missing data, incomplete sequences were
removed from the median-joining network analysis. Instead, these samples were
individually assessed for variable positions in Geneious (v.8.1.8) and if possible
assigned to their respective control region haplogroups.
Estimates of nucleotide diversity, ( ), the average number of nucleotide variations
per site between two samples, were calculated across complete modern and
ancient mitochondrial sequences in DnaSP (v.5.10.1) (Librado & Rozas 2009). This
was conducted to evaluate the level of variation outside of the control region and
therefore assess the benefit of sequencing full mitochondrial genomes for
phylogenetic analyses. Separate modern and ancient alignments were generated
using MUSCLE in Geneious (v.8.1.8). Only sequences exhibiting more than 90%
coverage were used for this analysis, as the program removes sites where one or
more sequences provide alignment gaps or missing data.
For the complete mitochondrial phylogenetic analyses, three alignments (modern,
ancient and modern + ancient) were generated using MUSCLE in Geneious
(v.8.1.8). The best-fit model for nucleotide substitution in the modern and ancient
datasets
was
determined
43
using
jModeltest2
(https://github.com/ddarriba/jmodeltest2, Guindon & Gascuel 2003; Darriba et al.
2012) with 11 substitution schemes. Model selection was computed using the
Bayesian and Akaike information criteria (BIC and AIC).
A Bayesian inference analysis of phylogeny was conducted using the MrBayes
plugin (v.2.2.2: Huelsenbeck et al. 2001; Ronquist & Huelsenbeck 2003)
implemented in Geneious (v.8.1.8). This consisted of running four chains with the
heated chain temperature set at 0.2 to allow sufficient chain-swapping over 20
million generations, with a burn-in length of 100,000 and sampling every 2000
generations. The F81 substitution model (Nst=1) (Felsenstein 1981) was applied
for the modern sequence analysis and the HKY85 substitution model (Nst=2)
(Hasegawa et al. 1985) with invariable rate variation (rate variation command
BIC and AIC calculations conducted in jModeltest2. Gaps or missing/ambiguous
characters are treated as missing data in MrBayes. Unlike PopArt and DnaSP, sites
with valid characters are retained for analysis, regardless of whether another
sequence produces missing data at that site. Complete mitochondrial sequences
from Thailand, Papua New Guinea and New Zealand (Genbank EU273710.1,
EU273709.1, EU273711.1; Robins et al. 2008) were added for the purposes of
outgrouping (Papua New Guinea) and also added to assess where these sequences
group in relation to the Polynesian assemblage. The resulting Bayesian consensus
trees were visualised in Geneious (v.8.1.8). Convergence was assessed using the
Trace tool in Geneious; there were no obvious trend lines to suggest that the
MCMC was still converging and there were no large-scale fluctuations that would
suggest poor mixing. Effective sample sizes (ESS) for all traces exceeded 100.
Maximum likelihood (ML) analyses (Felsenstein 1981) were conducted using the
PhyML plugin (v.3.0: Guindon et al. 2010) implemented in Geneious (v.8.1.8).
Substitution rate categories were set to 4 and bootstrap support was estimated
using 100 replicates. The F81 substitution model (Felsenstein 1981), with the
proportion of invariable sites fixed to zero and software-optimised estimates of
the gamma shape parameter and the transition/transversion ratio was applied to
the modern dataset. The HKY85 substitution model (Hasegawa et al. 1985), with
44
software-optimised estimates of the proportion of invariable sites, the gamma
shape parameter and the transition/transversion ratio was applied to the ancient
dataset. Like MrBayes, PhyML does not exclude sites where one or more sequences
provides missing data, but will ignore missing data for a particular sequence. The
complete mitochondrial sequences from Thailand, Papua New Guinea and New
Zealand (Genbank EU273710.1, EU273709.1, EU273711.1; Robins et al. 2008)
were also added. The Papua New Guinea sample was used as an outgroup.
Median-joining networks were generated using the aligned modern, ancient and
modern + ancient datasets in PopArt. The combined modern + ancient haplotype
analysis was conducted to visually assess how all the sequences associate in a
spatial and temporal sense. The sequences from Thailand, Papua New Guinea and
New Zealand (Genbank EU273710.1, EU273709.1, EU273711.1; Robins et al.
2008) were added as part of the modern + ancient haplotype analysis.
45
3 Results
3.1 DNA preservation and sequence recovery
A total of 47/97 modern samples produced adequately preserved DNA for
fragment amplification and subsequent sequencing. The overall PCR success rate
across modern samples was 48.5%. This low PCR rate could be attributed to the
degradation of DNA in some of the tissue samples, given all tissue samples were
retrieved from various states of preservation implemented over the last few
decades. Some samples were preserved in ethanol at room temperature, whilst
others had been placed in an -80°C freezer for long-term storage. Fifteen modern
samples were removed post-sequencing as a result of low coverage (<90%) and
read depth (<2) when mapped to a Rattus exulans mitochondrial reference
genome. Thirty-two samples remained, providing an overall success rate of 33%
from modern DNA extraction to downstream data analyses. These high coverage
modern samples all originate from the Cook Islands and Society Islands. A total of
40/77 ancient libraries produced adequately preserved DNA that successfully
amplified (success rate: 52%; see Table 3.1). These amplified libraries were then
subjected to hybridisation capture and sequencing. Twenty-seven samples were
removed post-sequencing as a result of low coverage (<80%) and read depth (<2)
when mapped to a Rattus exulans mitochondrial reference genome. Thirteen
ancient samples remained (see Table 3.1 and 3.2), providing an overall success
rate of 17% from ancient DNA extraction to downstream data analyses.
3.2 Sequence authenticity
To estimate the degree of reagent contamination, reads from each sample were
mapped to mitochondrial reference genomes from a pig, human, cow, chicken and
dog. This was compared to reads mapped to a rat mitochondrial genome. In the
ancient sequences, only one sample (MS10605) exhibited a small amount of
human contamination equating to approximately 0.03%. A lack of exogenous DNA,
particularly from cow, pig and chicken, indicates that reagents used in this study
were free from contamination. It has been previously asserted that laboratory
46
Table 3.1 Summary statistics for successful library amplification and post-sequencing processing of ancient R. exulans
samples. Successful DNA amplification of (n) libraries, prepared from ancient bone samples, is given by the amplification success
rate. Amplified libraries were then subjected to hybridisation capture and Illumina sequencing. The overall success rate is of (n)
remaining samples, after low coverage removal during bioinformatics processing. Age of material or associated dates: 1 = Present
500 B.P., 2 = 500 1000 B.P., 3 = 1000 2000 B.P., 4 = 2000 3000 B.P. * refers to unknown.
Amplification
success rate
(n) after low
coverage
removal
Overall
success rate
Site Type
Age
(n)
libraries
Pukotala
Open
4
2
50%
0
0%
Faleloa, Foa
Open
4
2
100%
1
50%
Mele Havea,
Open
4
1
100%
0
0%
Tokelau
Fakaofo
Open
2
6
66.7%
2
33.3%
American Samoa
Ofu Village, Ofu
Open (coastal)
4
6
33.3%
0
0%
Open
4
6
0%
0
0%
Open (coastal)
2
11
90.9%
6
54.5%
Location
Site
Tonga
Olosega Village,
Olosega
Easter Island
Anakena
47
Site Type
Age
(n)
libraries
Amplification
success rate
(n) after low
coverage
removal
Overall
success rate
Open (coastal)
2
5
60%
2
40%
Open
3
6
0%
0
0%
Mangaia
Open (coastal)
*
4
0%
0
0%
Terre Teahutapu,
,
Maupiti
Open (coastal)
1, 2
14
64.3%
0
0%
Terre Namotu,
Maupiti
Open (coastal)
1, 2
4
50%
1
25%
Haranai Valley,
Maupiti
Open
1, 2
4
75%
0
0%
*
1, 2
6
50%
1
16.7%
77
52%
13
17%
Location
Site
Austral Islands
Atiahara, Tubuai
Cook Islands
Ureia, Aitutaki
Society Islands
Society Islands
Haumi
archaeological site,
Moorea
TOTAL
48
Table 3.2 High-coverage ancient samples used in the phylogenetic
analyses. Associated layers dates are as reported in archaeological excavation
publications (see Table 2.3), except for the Society Islands (Jennifer Kahn,
personal communication).
Site
Tonga
Faleloa, Foa
MS10596
2600 ± 50 B.P.
Tokelau
Fakaofo
MS10592
1090 ± 60 B.P.
MS10591
1090 ± 60 B.P.
MS10606
610
560 B.P.
MS10610
610
560 B.P.
MS10609
610
560 B.P.
MS10602
610
560 B.P.
MS10605
610
560 B.P.
MS10604
610
560 B.P.
MS10601
769
675 B.P.
MS10600
769
675 B.P.
Terre Namotu,
Maupiti
MS10502
495
320 B.P.
Haumi
archaeological site,
Moorea
MS10523
750
550 B.P.
Easter Island
Austral Islands
Society Islands
Sample code
Associated
Location
Anakena
Atiahara, Tubuai
49
layer date
reagents, such as dNTPs, can contain DNA from domestic and laboratory animals
that may obscure endogenous DNA amplification (Leonard et al. 2007; Thomson et
al. 2014). No blank processed with any of the modern and ancient samples
exhibited contamination, providing no indication of cross-contamination during
the processing of samples. The deamination patterns exhibited in the reads from
each ancient sample were consistent with expected damage in ancient DNA
sequences. For example, sample MS10592 exhibited an elevated frequency of T
within the first ten
and of A nucleotide misi
the maximum rate of 0.05 for samples less than 117 years old, and is consistent
within samples older than 500 years (Sawyer et al. 2012). Conclusively, this
indicates that the sequences are endogenous mtDNA of rats obtained from early
archaeological deposits across Polynesia.
3.3 Phylogenetic analysis
An analysis of the control region was undertaken to categorise samples into the
three major haplogroups
I, II, III as described in Matisoo-Smith et al. (2004). The
majority of Polynesian samples (both modern and ancient) clustered within
Haplogroup IIIB, in particular adhering to the R9 haplotype (see Figure 3.2). The
samples that were missing sites and subsequently removed from the medianjoining network analysis, were analysed individually for variable positions that
would indicate haplogroup membership. For each sample, variable positions
across the 203bp fragment of the control region are shown in the Appendix Table
6.1. The samples MS10602, 606, 609 (Easter Island); MS10591, 592 (Fakaofo,
Tokelau); MS10596 (Foa, Tonga); MS10600, 601 (Tubuai, Austral Islands) and
MS10523 (Moorea, Society Islands) contained missing sites in the control region
and were therefore assessed individually. MS10606, 609 and 592 were assigned to
haplogroup IIIB. MS10600, 602 and 523 were missing some key variable sites but
could be assigned to haplogroup III. MS10596, 591 and 601 were missing a
50
a)
b)
Figure 3.1
of reads pertaining to the
ancient sample MS10592. a) The grey brackets, indicative of strand breaks,
incorporate the start and end of the mitochondrial sequence. The frequencies of
bases A, G, C and T are shown for the first and last ten nucleotide sites. b)
Nucleotide misincorporations across the first and last 25 nucleotide sites. The
coloured lines are indicative of the following nucleotides - red T; grey C; blue A;
purple G.
51
substantial amount of variable sites and could not be assigned haplogroup
membership. As depicted in Figure 3.2, the samples Raia28, Raia45 and Raia16
(Raiatea, Society Islands) radiated by one mutation off the haplotype R9. It was
unclear whether Raia28 and Raia45 were part of haplogroup IIIA or IIIB, however
after individual assessment all three samples were assigned to haplogroup IIIB.
Therefore all samples, except for those with substantial damage and the Puk001
(Pukapuka, Cook Islands) sample, were assigned to haplogroup IIIB. The Puk001
sample was clustered within haplogroup II, alongside the Papua New Guinean
Figure 3.2 Median-joining network of control region haplotypes. Colours
represent one of four central haplotypes chosen to represent each haplogroup (I,
II, IIIA and IIIB). The diameter of each circle is proportional to the frequency of a
specific haplotype.
52
sequence (GenBank EU273709.1). Interestingly, the Thailand sequence (GenBank
EU273710.1), clustered outside all three major haplogroups, indicating the
existence of potentially another major R. exulans haplogroup.
Nucleotide variation across both complete modern and ancient mitochondrial
genome datasets (>90% coverage) was evaluated using the program DnaSP
(Librado & Rozas 2009). In the modern dataset, a total of 113 variable sites were
identified. The average number of nucleotide differences (k; Tajima 1983) between
two modern sequences was 9.2 ± 18.92. Figure 3.3a illustrates the variation in
nucleotide diversity ( ; Nei 1987) across the modern mitochondrial dataset, using
overlapping windows of 100 bp (step size: 25) that are centred at the midpoint. In
the modern dataset, the highest nucleotide diversity was found at the end of the
tRNA-pro coding region and the beginning of the control region from the
nucleotide positions 15339-15690, peaking at 0.00402. Conversely, the
hypervariable I region (HVR I) of the control region (positions 16024-16383)
produced a lower level of nucleotide variation, peaking at 0.00156. Across the
coding region in modern samples, the highest nucleotide diversity was found in the
protein-coding genes COX1 (peak of 0.00286) and ND4L-4 (peak of 0.00261).
In the ancient dataset, a total of 20 variable sites were identified. The average
number of nucleotide differences (k; Tajima 1983) between two ancient sequences
was 6.5 ± 11.5. Figure 3.3b illustrates the variation in nucleotide diversity ( ; Nei
1987) across the ancient mitochondrial dataset. In ancient sequences, the highest
level of diversity was found at the beginning of the protein-coding region ND5
(from nucleotide position 11713-12130), with a peak of 0.01. The control region in
comparison shows a lower level of nucleotide variation, with a peak of 0.002 in the
HVR I region.
The consensus maximum likelihood tree for modern samples, supported by
Bayesian posterior probabilities, indicated two major groupings
the first
consisting of sequences from Papua New Guinea and Pukapuka (indicative of
control region haplogroup II) and Thailand, and the second consisting of rats from
53
a)!
b)!
!
!
Figure' 3.3! Distribution' of' nucleotide' diversity' (π)' across' a)' modern' and' b)'
ancient' complete' mitochondrial' sequences.' Nucleotide! diversity,! (π),! represents!
the! average! number! of! nucleotide! variations! per! site! between! two! samples.! This! is!
distributed!in!windows!of!100bp!(step!size!of!25!and!centered!at!the!midpoint)!across!
the!entire!mitochondrial!genome,!represented!in!a!linearised!genetic!map.!
!
!
!
54!
across Remote Oceania (haplogroup III) (see Figure 3.4). Using the substitution
rate on the maximum likelihood tree, approximately 100 mutational steps separate
the modern Near Oceanic and Remote Oceanic populations. Across the Remote
Oceanic populations, there was high Bayesian posterior probability support for a
group containing New Zealand and Huahine (Society Islands) sequences, distinct
from the rest of the Polynesian samples. Other Huahine lineages however, were
present in the wider Central Polynesian assemblage. Overall there was very little
bootstrapping and posterior probability support for phylogeographic structure.
Whilst many islands have produced multiple lineages, there are no supported
island groupings whereby all lineages are grouped together and are distinct from
the rest of the Polynesian assemblage. The median-joining network of modern
sequences revealed a major central East Polynesian haplogroup, consisting of
many Cook Island and Society Island linages radiating from an ancestral haplotype
comprising of five samples from Tahiti, Raiatea (Society Islands) and Aitutaki
(Cook Islands) (see Figure 3.5). The radiating lineages are distinct from the
ancestral haplotype by one to nine mutational steps. The Pukapuka sequence is
widely distinct from the central East Polynesian haplogroup, which is to be
expected given that it has been confirmed as a haplogroup II rat.
The consensus maximum likelihood (ML) tree for the ancient samples, supported
by Bayesian posterior probabilities, also indicated the same two major groupings,
indicative of Papua New Guinea (haplogroup II) and Thailand, and all other
Remote Oceanic (haplogroup III) rats (see Figure 3.6). Using the substitution rate
on the maximum likelihood tree, the Near Oceanic populations are separated by 73
mutational steps from the deepest branch in the ancient Remote Oceanic group.
Tonga and Tokelau, which have previously been identified as part of an
overlapping region between the control region haplogroups IIIA and IIIB (MatisooSmith et al. 2004), produced ancient samples (MS10596 & MS10592) that
clustered on the deeper branches within the Polynesian assemblage. The Tongan
sample was unfortunately too damaged in the control region to evaluate whether it
belonged to either IIIA or IIIB. The Tokelau sample (MS10592) however was
assigned to haplogroup IIIB. The maximum-likelihood analysis indicated that 9
mutational steps separate the Tonga and Tokelau samples. Conversely, the median
55
!
!
!
!
!
!
Figure!3.4!Rooted!maximum!likelihood!(ML)!tree!inferred!from!the!modern!
(complete! mitochondrial)! dataset.!Values!above!each!node!represent!Bayesian!
posterior! probabilities! (obtained! using! MrBayes)! and! ML! bootstrap! values!
(obtained!from!PhyML).!Only!posterior!probabilities/bootstrap!values,!where!one!
or!both!are!above!0.7!and!70!respectively,!are!included.!Individual!sample!names!
are! included! and! geographic! populations! defined! by! colour.! Scale! bar! represents!
amino!acid!substitutions!per!site.!Additional!R.#exulans!sequences!(Thailand,!Papua!
New! Guinea! (outgroup)! and! New! Zealand;! Genbank! EU273709.1,! EU273710.1,!
EU273711.1)!(Robins!et!al.!2008)!were!included.!!
!
56!
!
!
!
Figure!3.5!Median?joining!network!of!the!modern!(complete!mitochondrial)!
dataset.! Colours! represent! geographic! sampling! locations! across! central! east!
Polynesia.!The!diameter!of!each!circle!is!proportional!to!the!frequency!of!a!specific!
haplotype.!
!
!
57!
-joining (MJ) network indicates they are only separated by one mutational step
(see Figure 3.7). However PopArt does remove varying sites from the MJ analysis if
one or more sequences provide missing data and therefore cannot be considered a
reliable source for mutational steps between sequences.
Median-joining networks do not identify ancestor-descendent relationships but
instead are based on overall similarity. However the ancient median-joining
analysis has produced a network that shows a direct link between Western
(Tonga) and far Eastern Polynesia (Easter Island), where the Tokelau sample
(MS10592) could be considered a transitional haplotype. It can therefore be
argued that that the central haplotype corresponding to MS10592 (and thereafter
all Polynesian haplotypes) are directly descended from this ancestral haplotype
uncovered in Tonga. This coincides with what we know of Pacific rat and
associated human dispersal across the Pacific, which has overwhelming headed in
an easterly direction.
A shallower branch on the ancient maximum likelihood tree revealed a Society
Island grouping (Moorea & Maupiti islands), with high bootstrap and Bayesian
support. This is distinct from the rest of the Polynesian assemblage that included
samples from New Zealand, Tokelau (MS10591), Easter Island and the Austral
Islands. The median-joining network indicates the existence of two haplotypes
corresponding to the samples from Maupiti and Moorea. These haplotypes radiate
off the central haplotype (MS10592) by several mutations, according to the ML
tree. Given their MJ network position, it would seem that these two Society Island
haplotypes are not directly ancestral to the rest of the Polynesian assemblage.
Whilst we cannot exclude the possibility that the central haplotype corresponding
to MS10592 could also exist in the Society Islands, there is no conclusive evidence
from this study to indicate an Eastern Polynesian expansion from the Society
Islands. The other ancient sample from Fakaofo, Tokelau (MS10591) clustered
with the rest of the Polynesian assemblage (New Zealand, Easter Island and
Austral Islands) with high Bayesian support on the maximum-likelihood tree.
According to the ML tree, MS10591 is separated by 14 mutational steps from the
central haplotype.
58
!
!
Figure' 3.6!Rooted' maximum' likelihood' (ML)' tree' inferred' from' the' ancient'
(complete' mitochondrial)' dataset.!Values!above!each!node!represent!Bayesian!
posterior! probabilities! (obtained! using! MrBayes)! and! ML! bootstrap! values!
(obtained!from!PhyML).!Only!posterior!probabilities/bootstrap!values,!where!one!
or!both!are!above!0.7!and!70!respectively,!are!included.!Individual!sample!names!
are! included! and! geographic! populations! defined! by! colour.! Scale! bar! represents!
amino!acid!substitutions!per!site.!Additional!R.#exulans!sequences!(Thailand,!Papua!
New! Guinea! (outgroup)! and! New! Zealand;! Genbank! EU273709.1,! EU273710.1,!
EU273711.1)!(Robins!et!al.!2008)!were!included.!
!
59!
Figure'3.7!Median?joining' network' of'the' ancient'(complete'mitochondrial)'
dataset.' Colours! represent! geographic! sampling! locations! across! West! and! East!
Polynesia.!The!diameter!of!each!circle!is!proportional!to!the!frequency!of!a!specific!
haplotype.'
!
!
!
!
!
!
60!
On the shallowest branches of the ancient maximum-likelihood tree, there is high
Bayesian support for a distinct Easter Island/Tubuai (Austral Islands) grouping.
The corresponding median-joining network revealed a large far Eastern
Polynesian haplotype (consisting of 5 Easter Island samples) that radiates off the
central haplotype (MS10592). According to the ML tree, they are separated by
approximately 8 mutations from the central haplotype. Another Easter Island
(MS10606) and two Tubuai haplotypes radiate from this large East Polynesian
haplotype.
The combined ancient and modern complete mitochondrial median-joining
network is illustrated in Figure 3.8. The GenBank sequences corresponding to
Papua New Guinea, Thailand and New Zealand were also included in this network
analysis. Coinciding with the modern median-joining network, many of the modern
Society Island and Cook Island (Aitutaki) samples form a large grouping of
haplotypes, consisting of a central haplotype (of eight samples) and 15 haplotypes
that directly radiate off this node. The ancient Society Island samples form part of
this group, with the Maupiti sample adhering to the central haplotype and the
Moorea sample radiating off the central node. The ancient Society Island
populations could therefore be considered directly ancestral to the modern Society
Island populations.
The modern and ancient Society Islands/Cook Islands haplotypes are directly
connected to the ancient sample from Fakaofo, Tokelau (MS10592), which is the
central haplotype from which all East Polynesian samples are derived. The central
haplotype is derived from the ancient Tongan haplotype, which then links back by
73 mutations to the haplogroup II rats (Papua New Guinea and Pukapuka) and the
Thailand sample. The modern New Zealand haplotype directly radiates off the
central haplotype (MS10592), inferring direct ancestry from this haplotype. The
other sequence obtained from Fakaofo, Tokelau (MS10591) also radiates off the
central haplotype as described previously. The large far Eastern Polynesian group
consisting of samples from Easter Island and Tubuai, Austral Islands directly
radiates off the central haplotype. Unlike in the ancient only MJ tree, an Austral
Island sample has been incorporated into the larger haplotype. This can be
61
attributed to the use of a larger sample assemblage in PopArt. As PopArt removes
variable sites from consideration where one or more samples may be missing a
site or produce a gap, it is likely that the mutation that previously distinguished
the Austral Island sample from the larger Easter Island-only haplotype has been
removed from the analysis. Interestingly, two modern Huahine (Society Island)
samples directly radiate off the shared ancient Easter Island/Austral Island
haplotype. These two samples (Hua27 and Hua24) grouped with the New Zealand
sequence in the modern-maximum likelihood analysis.
62
!
Figure' 3.8' Median/joining' network' of' the' modern' and' ancient' (complete'
mitochondrial)' datasets,' including' acquired' GenBank' sequences.' Colours!
represent! geographic! sampling! locations! across! the! Pacific.! The! diameter! of! each!
circle!is!proportional!to!the!frequency!of!a!specific!haplotype.!
!
63!
4 Discussion
4.1 DNA preservation in Pacific archaeological contexts
DNA preservation and recovery from post-mortem samples is largely dependent
on environmental factors (i.e. temperature, moisture and acidity of the
surrounding context) and age of the sample. As discussed in Chapter 1, Pacific
preservation varies across tropical to temperate climates, and whether the sample
was recovered from a protected, open and/or coastal context (Robins et al. 2001).
Bones recovered from open sites and that fall within the age range 1000-3000 B.P.,
have generally produced low quality DNA. This includes early samples from the
previously produced 0% PCR success rates (Robins et al. 2001). In this study,
assemblages obtained from open sites dated to between 1000-3000 B.P. produced
a range of library amplification success rates (prior to hybridisation capture) from
0 to 100%. Interestingly, assemblages from Foa,
amplification success rates of 100, 100 and 50% respectively. However after the
removal of low coverage genomes during bioinformatics processing, this success
rate abruptly declined. Only one genome (obtained from Foa, Tonga) from the age
range 1000-3000 B.P. was retained for phylogenetic analyses.
Robins et al. (2001) observed that open site samples, dated up to 1000 B.P.,
produced higher PCR success rates between 60-100%. This included samples from
the Cook Islands, New Zealand, Chatham Islands, Norfolk Island and Hawaii. PCR
success in these open sites however, was generally higher in islands under a
temperate climate than a tropical one. In this study, assemblages from open sites
dated up to 1000 B.P. produced a higher library amplification success rate ranging
from 50 to 90.9%. Easter Island produced both the highest amplification success
rate (90.9%) and overall success rate (54.5%), which can likely be attributed to its
younger contextual age and temperate climate. The other assemblages, all
obtained across tropical Polynesia, produced overall success rates between 0 to
40%.
64
Previous commensal studies using archaeological specimens across the Pacific
have produced varying PCR success rates. Larson et al. (2007) targeted three
120bp fragments of the mitochondrial control region in ancient pig samples
collected across Island Southeast Asia and Oceania. Amplifiable DNA was present
in 23/57 ancient bones and teeth. Of these, only 5 archaeological specimens were
obtained from Pacific prehistoric sites, which dated from 925 to 300 B.P. These
five samples originated from Tubuai, Hanamiai (Marquesas), the Reef Islands
(Solomon Islands), Tanagatatau Rock Shelter on Mangaia (Cook Islands) and
Mussau Island (Papua New Guinea). Many of the samples that failed to produce
amplifiable DNA were from prehistoric contexts ranging from ~9000-600 B.P. All
of these sites were situated in tropical/humid climates i.e. Micronesia, Taiwan,
China, Indonesia, Philippines and Papua New Guinea. Conversely, prehistoric dog
remains retrieved from across Polynesia have consistently produced amplifiable
DNA (Savolainen et al. 2004, Greig et al. 2015). Savolainen et al. (2004)
successfully sequenced 263bp of the mitochondrial control region from all 19
collected prehistoric archaeological samples. These were obtained from the Cook
Islands (sites dated from 1000 B.P.), New Zealand (sites dated from 700-400 B.P.)
and Hawaii (undated prehistoric site). A high level of DNA preservation has
earliest archaeological site (Greig et al. 2015). All 16 dog samples produced
enough endogenous DNA for hybridisation capture and complete mitochondrial
sequencing. Fourteen samples produced high coverage (over 95%) and were
retained for phylogenetic analyses. DNA preservation in archaeological deposits
across the Pacific seem to be highly influenced by the surrounding environment,
attributed to both climate and site context. In turn, the susceptibility to
degradation, particularly in humid open-site contexts, is compounded by age.
4.2 The application of complete mitochondrial genome sequencing
The majority of commensal studies in the Pacific, particularly those using
archaeological specimens, have targeted fragments of 100-300bp in the
mitochondrial control region. Mitochondrial hypervariable regions accumulate
mutations at a faster rate than other mitochondrial regions (Stoneking 2000),
producing a greater amount of population variation that can be utilised for
65
phylogenetic research. Pacific commensal studies in the last decade have generally
followed aDNA protocols such as Matisoo-Smith et al. (1997) and Shapiro et al.
(2004), where short overlapping fragments are amplified and pieced together after
Sanger sequencing by capillary electrophoresis. The cloning of PCR products was
also used to assess template damage, detect numts (nuclear copies of mtDNA) and
evaluate accuracy and contamination. As ancient DNA is highly fragmented, it was
practical to target fragments of the mitochondrial control region that produced the
most usable variation. The emergence of second-generation sequencing however
has made it viable to generate longer DNA sequences, e.g. complete mitochondrial
genomes, from fragmented aDNA (Knapp & Hofreiter 2010). The use of
hybridisation capture for target DNA enrichment prior to second-generation
sequencing (Ng et al. 2009), is also beginning to supersede the traditional primerbased PCR amplification that can often amplify large amounts of exogenous
contaminant DNA (Greig et al. 2015; Li et al. 2013; Knapp et al. 2012; Burbano et
al. 2010; Hodges et al. 2007). Recent Pacific commensal studies using complete
mitochondrial genomes have reported a higher level of molecular resolution that
can distinguish phylogenetic groups and individuals with greater accuracy (Greig
et al. 2015; Miao et al. 2013; Duggan et al. 2014).
Matisoo-Smith et al. (2004) study of modern and ancient Pacific rat phylogenies
across ISEA and the Pacific targeted ~240bp of the mitochondrial control region.
This produced 32 haplotypes from 27 variable sites. Notably, the majority of
Remote Oceanic samples (Haplogroup IIIB) either belonged to or radiated off the
R9 haplotype, making it very difficult to distinguish dispersal routes. A follow up
study by Barnes et al. (2006) of Easter Island rats, found that all ancient Pacific rat
samples belonged to the R9 haplotype. It was therefore impossible to establish the
specific source population for Easter Island rats. The control region analysis
conducted in this study found that the majority of the Polynesian samples (both
modern and ancient) clustered within Haplogroup IIIB, in particular adhering to
the R9 haplotype. The use of this control region fragment alone did not produce
enough variation to distinguish island groups or specific lineages.
66
Nucleotide diversity across the complete mitochondrial genomes varied between
modern and ancient samples. In the ancient dataset (comprising of 1 Western and
12 Eastern Polynesian samples), 20 variable sites were identified from which 9
haplotypes were determined. Ancient nucleotide diversity was highest in protein
coding regions, outside of the control region. In the modern dataset (comprising 31
Eastern Polynesian samples and a haplogroup II sample), 113 variable sites were
identified from which 24 haplotypes were present in the median-joining analysis.
In the modern dataset, the highest nucleotide diversity was found at the beginning
of the control region, but notably outside of the hypervariable I region. The low
variation in the hypervariable regions of both modern and ancient Pacific rat
genomes may reflect a number of bottlenecks in the establishment of populations.
Pacific rats, carried by migrating humans, would have been established on Pacific
islands in a stepping-stone manner. A subset of the population (and therefore
haplotypes) would have been carried on each migratory journey, as humans
moved eastwards into the far reaches of Polynesia. Furthermore, as populations
were established on islands, gene flow would have been heavily restricted and
subsequently prolonged low variation. As nucleotide diversity in the HVR I region
of both ancient and modern samples is similarly low, it would seem that
hypervariable variation has not yet recovered. However, overall there was more
hypervariable variation in the modern sequences than in the ancient, signifying
that mutations have accumulated since the initial establishment of island
populations.
The varying distribution of nucleotide diversity across complete ancient and
modern mitochondrial genomes indicates that targeting one region (e.g. HVR I)
may not provide enough variation across all samples to sufficiently distinguish
populations. Furthermore, the effect of bottlenecks would have significantly
decreased genetic variation. In conjunction with previous Pacific commensal
research using complete mitochondrial genomes (Duggan et al. 2014; Greig et al.
2015; Miao et al. 2013), my results suggest that complete mitochondrial
sequencing is necessary to assess all mitochondrial variation and gain the highest
molecular resolution for subsequent phylogenetic analyses.
67
4.3 Implications for Pacific rat dispersal
The three major control region haplogroups, first identified by Matisoo-Smith et al.
(2004) show clear geographic patterning associated with various Pacific rat
dispersal and interaction spheres across Southeast Asia and the Pacific. As
discussed previously, Haplogroup I rats are restricted to the Southeast Asian
region, Haplogroup II is distributed across both Southeast Asia and Near Oceania,
whilst Haplogroup III has been largely confined to Remote Oceania and is
associated with the Lapita/Polynesian movement through this region. An analysis
of a 203bp fragment of the control region was conducted to categorise all samples
to their respective haplogroups prior to further phylogenetic analyses. A complete
mitochondrial sequence from Thailand clustered outside of the three haplogroups,
revealing the existence of a potentially new major Pacific rat haplogroup in
mainland Southeast Asia.
The identification of a modern Pukapukan sample as part of Haplogroup II was
surprising, given no sample further east than the Santa Cruz Islands has been
associated with this haplogroup (Matisoo-Smith et al. 2004). As the Pukapuka
sample is part of the modern assemblage, it is possible that its represents a recent
translocation of rats from ISEA/Near Oceania. This could have occurred as a result
of modern shipping/travel between islands in Near and Remote Oceania, however
it has been previously asserted that Pacific rats do not travel on modern ships, nor
have been found in ports (Matisoo-Smith 1996). It is also uncertain whether
modern translocations of rats can gain a foothold in previously colonised islands
unless original populations have undergone extinction. Under ideal conditions, rats
are capable of doubling their population every 47 days (Hunt & Lipo 2012). Within
several months, populations can reach over one million, which would make it very
difficult for a newly translocated haplotype to gain in frequency. Furthermore, wild
rats are highly territorial (Barnett 1955) using olfactory cues to distinguish
familiar from foreign conspecific rats (Alberts & Galef 1973). This would assert a
large influence on the reproductive success of newly introduced rats.
Alternatively, the Pukapuka sample could be representative of a prehistoric
Haplogroup II population established as a result of early interaction and trade with
68
Western Polynesia that has links into Near Oceania. Whilst Pukapuka, as part of
the Cook Island group, technically falls within the Eastern Polynesian region, the
island seems to be linked to Western Polynesia. For example, the language spoken
on Pukapuka does not form part of the Central Eastern Polynesian linguistic group
from which the Rarotongan and Penrhyn (Cook Island) languages are derived
(Trudgill 2004). Instead, Pukapuka is one of the main Samoic languages related to
Tokelauan, Tuvaluan, Uvean, Futunan and Samoan, found across Western
Polynesia. The Samoan and Eastern Polynesian group (which further subdivided
into Central Eastern and Rapanui, Easter Island) are said to have divided from
Nuclear Polynesian, an ancestral language that originated after the human
settlement of Samoa (Trudgill 2004). There has been limited archaeological
research on the colonisation and post-settlement interactions regarding Pukapuka.
However, an analysis on the composition of basalt adzes retrieved from across
Polynesia, found that a Pukapuka sample clustered with basalt samples from
Western Samoa, Fiji, Tonga, Tokelau, Solomon Islands and Tuvalu (Best et al
1992). This may be representative of basalt trade between Pukapuka and Western
Western Polynesia, it is possible that a Haplogroup II rat population originating
from Near Oceania and transferred through post-settlement interactions in
Western Polynesia was established on Pukapuka.
It was not possible to assign the ancient Tongan haplotype (sample MS10596) to a
control region haplogroup, as a result of DNA deterioration/damage. As the sample
MS10596 is only separated by ten mutational steps from MS10592 (Fakaofo,
Tokelau) however, it was safely assumed that MS10596 forms part of haplogroup
III that is found across Remote Oceania. MS10596 was uncovered from an early
archaeological context on Foa, dated to 2600 ± 50 B.P. (Burley et al. 1999).
Presently, the colonisation of Tonga and Samoa is conceded to have occurred by
2800 and 2750 B.P. respectively (Burley et al. 2015; Petchey 2001; Clark et al.
2016). Given only a 50 year time span between the colonisation of these two major
Western Polynesian archipelagos, it would be expected that the original
haplotype/s carried by the rats into both Tonga and Samoa would be the similar
and likely contain the haplotype corresponding to MS10596.
69
Archaeological and linguistic evidence suggests that Tonga and Samoa form the
Western Polynesian homeland from which Eastern Polynesian populations are
derived (Green 1966, 1981; Pawley 1966; Groube 1971; Emory 1946). MatisooSmith et al. (1998) study of modern Pacific rat variation across Near and Remote
Oceania, found that a Western Polynesian (Samoan) sample clustered with Society
Island and Cook Island samples on the deeper branches of a neighbor-joining tree.
analysis, the Foa (Tonga) sequence
(MS10596) occupied the deepest branch, outside of all other Remote Oceanic
samples. The clustering of the Samoan and Tongan sequences on the deeper
branches of their respective analyses signifies that they are sister lineages to those
in Eastern Polynesia. In conjunction with the ancient median joining network,
illustrating a direct Eastern Polynesian radiation off MS10596, it is possible that
the Tongan haplotype uncovered in this study is ancestral to all those found in
Eastern Polynesia.
All East Polynesian sample/groups identified as part of haplogroup IIIB, radiate off
the central haplotype (relating to the sample MS10592) that was uncovered in
Fakaofo, Tokelau. Given that this haplotype directly radiates off the Tongan
haplotype in the median joining network, it is likely a descendent haplotype
carried by Tongan and/or Samoan populations into central East Polynesia. The
identification of this haplotype in Tokelau alone however, does not suggest that all
East Polynesian populations are descended from an early Tokelauan population.
Further ancient sampling across Central Polynesia would be required to assess the
distribution of this central haplotype. Given that the earliest radiocarbon dates
associated with human settlement in central East Polynesia are found in the
Society Islands (~925-830 B.P.; Wilmshurst et al. 2011), it is highly probable that
this central haplotype would be found in this region. Furthermore, Matisoo-Smith
et al. 1998 identified that the southern Cooks/Society Islands formed a central-east
interaction sphere central to a northern and a southern sphere. This central-east
interaction sphere, linked to all major East Polynesian island groups, likely reflects
a central homeland region from which eastern expansion occurred.
70
The central haplotype is separated from another Tokelauan haplotype (sample
MS10591) by 14 mutational steps. The samples MS10591 and MS10592 were
uncovered from the same early occupation layer, although from different trenches
at the archaeological site (Best 1988). This indicates that the central and diverged
haplotypes existed concurrently. The derived haplotype may have diverged off the
central haplotype either prior or after human establishment on Fakaofo. If it
diverged prior to migration and was part of a central-east population that
contained both haplotypes, then it would be expected that both haplotypes could
be found in the earliest cultural layers in Fakaofo. If the derived haplotype
diverged after establishment, then the layer context in which both samples were
found may not be representative of an early human cultural layer.
The ancient Society Island sequences obtained from Maupiti and Moorea
(MS10502 and MS10592) are derived from the central haplotype but differ by
several mutational steps. These samples and their associated layers were dated to
495-320
B.P.
and
750-550
B.P.
respectively
(Jennifer
Kahn,
personal
communication). Therefore the samples from Maupiti and Moorea do not
represent the oldest archaeological layers that date to initial colonisation in the
Society Islands. However, widespread radiation from the Society Islands may not
have occurred until 750 B.P., given that no Eastern Polynesian archipelago, except
for the Society Islands, have been dated prior to this (Wilmshurst et al. 2011;
Petchey 2010). As no other recovered ancient central East Polynesian haplotypes
are derived from the haplotype found on Moorea, which would have existed at the
time of widespread expansion, or Maupiti, it could be argued that the continued
East Polynesian expansion did not originate from these islands. The haplotype of
the ancient sample from Maupiti however, is the central node of which many
modern Society Island and Cook Island samples share or are derived. Further
sampling of ancient Pacific rats across the Society Islands and Cook Islands would
be required to assess whether the central haplotype was established in these
central island groups and whether it perhaps existed concurrently with derived
haplotypes, such as those found in Moorea and Maupiti. If early archaeological
contexts in the Society Islands are devoid of samples carrying the central
71
haplotype, then the Society Islands may not be the central hub from which all other
Eastern Polynesians populations are derived.
The ancient maximum-likelihood analysis provided high Bayesian support for a
distinct Easter Island/Tubuai (Austral Islands) grouping, separated by 8
mutational steps from the central haplotype. Previous research that targeted
239bp of the mitochondrial hypervariable region in ancient Easter Island Pacific
rats, were unsuccessful in identifying an immediate origin for this population
(Barnes et al. 2006). This was because all ancient Easter Island samples belonged
to the R9 haplotype, which is widely found across Remote Oceania. The absence of
variation was attributed to a single or limited introduction of rats, followed by
long-term isolation. Hagelberg et al. (1994) targeted the mitochondrial DNA from
ancient human remains, confirming a Polynesian origin for the Rapanui people.
This is consistent with linguistic evidence of an ancestral Eastern Polynesian
language group that split into Central Eastern Polynesian languages (e.g.
(Easter Island) (Trudgill 2004). This early divergence of the Rapanui language
would be consistent with an early Easter Island colonisation followed by extreme
isolation from all other Eastern Polynesian groups. The immediate origin of the
prehistoric people of Easter Island is however still unknown.
Wilmshurst et al. (2011) re-analysis of Eastern Polynesian radiocarbon dates,
indicated that Easter Island was colonised by ~750-697 B.P. Following
establishment in the Society Islands around 925-830 B.P, there was a fast and
widespread radiation to all major Eastern Polynesian archipelagos between ~750650 B.P. The grouping of the Tubuai (Austral Is) and Easter Island haplotypes in
this study suggests that both island groups were established by the same
colonisation wave, which carried the haplotype from which Easter Island and
Tubuai rats share or are diverged. Whilst haplotype networks, such as the medianjoining network do not indicate ancestor-descendent relationships and therefore
the direction of migration, it is unlikely that the Tubuai haplotypes are derived
from the Easter Island population. Geographically, Easter Island is situated in the
eastern-most corner of Polynesia. Migratory voyages are likely to have passed
72
through other island archipelagos in a stepping-stone manner before reaching
Easter Island. Therefore it is highly likely that the Austral Islands were colonised
prior to Easter Island. Preliminary radiocarbon dating on Tubuai suggests that it
had been occupied earlier (Worth & Bollt 2011), contemporaneous with sites in
the southern Cooks (Walter 1998). This is the first commensal study to indicate a
potential immediate origin for the Easter Island population, providing support for
the colonisation of Easter Island through a south-eastern voyaging corridor,
originating from the Austral Islands and potentially moving through the Gambier
Islands and Pitcairn Islands. Further sampling would be required in the Gambier
and Pitcairn Islands however to establish haplotype continuity in this southeastern corridor.
As discussed above, Rapanui diverged from the ancestral Eastern Polynesian
language group, distinct from all other Central Eastern Polynesian languages. This
likely reflects the isolation of the Rapa Nui people post-colonisation. The
adherence of all six ancient Easter Island rats sampled in this project to two
haplotypes, representing the initial and a derived haplotype, suggests a limited
introduction of rats onto the island. This coincides with Barnes et al. (2006)
finding of limited variation, which was attributed to extreme isolation after the
initial colonisation. However given that the data is based on rats and not humans,
it is possible that there were human arrivals post-colonisation that did not carry
Pacific rats with them.
The progression of a shared Central East Polynesian language, exempting Rapanui,
is suggestive of one or more interaction spheres that effectively maintained
language continuity across Central East Polynesia. As discussed in Chapter 1, there
is archaeological, linguistic and genetic evidence for a number of interaction
spheres across Eastern Polynesia (Matisoo-Smith 1998; Sheppard et al. 1997;
Conte & Kitch 2004; Rolett 2002; Kirch 2000). Rolett (2002) argued that a core
interaction zone existed between the Southern Cooks, Society Island and Austral
Islands, central to many other smaller interaction spheres. Kirch (2000) also
suggested that based on linguistic evidence, one of these smaller interaction
spheres may have existed between the Austral Islands, Mangareva (Gambier
73
Islands) and the Pitcairn islands to the eastern Tuamotus and the Marquesas. The
inclusion of the Austral Islands into both the central and south-eastern interaction
spheres indicates that it was potentially the link between these two regions.
Interestingly, two modern Huahine (Society Island) sequences were found to
radiate off the ancient Eastern (Austral Is/Easter Is) Polynesian haplotype. These
two lineages are not derived from any of the haplotypes currently found in the
Society Islands. The establishment of these unusual lineages on Huahine may
reflect a post-colonisation translocation of rats from the Austral Islands into the
Society Islands as part of the core interaction zone.
The Society Island and Cook Island populations were the only groups that
successfully produced both ancient and modern data. The ancient Maupiti sample
represents a haplotype from which nearly all modern and ancient Society
Island/Cook Island populations share or are derived. Modern Pacific rat
populations in the Society Islands and Cook Islands are therefore largely
representative of prehistoric populations that were established with initial human
settlement. It has been previously reported that modern Pacific rat populations in
the Chatham Islands are representative of prehistoric populations (Matisoo-Smith
2002). In New Zealand however, not all ancient lineages were found in the modern
populations. Widespread extinction on mainland New Zealand likely lead to the
loss of ancient lineages that are no longer represented in extant populations on the
offshore islands (Matisoo-Smith 2002). Extinction of lineages may also have
occurred in other island groups and subsequently are no longer represented in
modern populations. Conversely, modern populations may have incorporated
lineages that are not represented in prehistoric assemblages. This may have
resulted from post-colonisation trade movements, however it has not yet been
critically assessed whether these movements were associated with further rat
dispersal. Given the alleged difficulty of foreign lineages to integrate and increase
in frequency within established populations (Barnett 1955; Alberts & Galef 1973),
introduced lineages after initial colonisation may have remained at a low
frequency or went extinct.
74
Without the use of both ancient and modern specimens concurrently, it is difficult
to make accurate interpretations on origins, migrations and post-colonisation
interactions. Whilst the results of this project can confirm that the modern Society
Island/Cook Island populations are representative of their prehistoric ancestors
and as such could be used in further analyses as a proxy, there is limited
information about temporal haplotype continuity from other regions in Oceania. It
is recommended that any further research involving the sampling of Pacific rats in
Oceania incorporate the use of both prehistoric and modern Pacific rat samples.
75
5 Conclusion
5.1 Summary
Differences in the variation of nucleotide diversity across complete ancient and
modern Pacific rat mitochondrial genomes indicated that targeting one region (e.g.
hypervariable I region) of the mitochondrial genome, may not provide enough
variation across both ancient and modern samples to sufficiently distinguish
haplotype lineages. Complete mitochondrial sequencing provided greater
molecular resolution, distinguishing new haplotype lineages within the
encompassing control region R9 haplotype of Eastern Polynesia. Recent
advancements in target capture and second-generation sequencing have made it
accessible and more reliable to retrieve complete mitochondrial genomes from
ancient specimens. There is great opportunity for the wider application of
complete mitochondrial sequencing as part of commensal research across the
Pacific.
Varying rates of DNA preservation in archaeological deposits across the Pacific
however does limit the ability to extract amplifiable DNA. Ancient library
amplification success rates in this study coincide with previous reports of low DNA
preservation in hot, humid environments, particular if the archaeological site is
open and older than 1000 B.P. The highest overall success rate (after low coverage
removal) was produced from the archaeological site of Anakena, Easter Island.
Notably, Easter Island falls under a more temperate climate than most Polynesian
islands and has only been colonised within the last 750 years. This is consistent
with high success rates reported in other temperate islands such as New Zealand
and the Chatham Islands, which were also colonised within the last 750 years.
The ancient Eastern Polynesian haplotypes, representative of samples sourced
from some of the earliest archaeological layers across Eastern Polynesia, indicate
descent from a Western haplotype uncovered in Foa, Tonga. This coincides with
archaeological and linguistic evidence of a Western Polynesian homeland in Tonga
and Samoa, from which Eastern Polynesian populations are derived. A central
Eastern Polynesian haplotype that was uncovered in Fakaofo, Tokelau, is ancestral
76
to all other haplotypes found across Eastern Polynesia. The central haplotype
reflects a previously proposed central East Polynesian homeland region from
which eastern expansion occurred. Radiocarbon evidence indicates that the
Society Islands is the most likely candidate, however the ancient Society Island
haplotypes reported in this project are not ancestral to other Eastern Polynesian
populations. The ancient Maupiti (Society Islands) haplotype is however ancestral
to nearly all modern lineages found across both the Society Islands and the Cook
Islands. The modern Society Island/Cook Island populations are therefore largely
representative of prehistoric populations in this region. Easter Island and Tubuai
(Austral Islands) share related haplotypes that are derived from the central
haplotype in central East Polynesia. This suggests that both islands were
established by the same colonisation wave that originated in the central homeland
region and moved through the south-eastern corridor, culminating with the
colonisation of Easter Island. This is the first commensal study to provide a
potential immediate origin for the settlement of Easter Island.
5.2 Future Directions
Further sampling of ancient specimens across central East Polynesia would be
required to assess the distribution of the central haplotype. Given archaeological
evidence that posit that the Southern Cooks/Society Island were the first Eastern
Polynesian islands to have been colonised from Samoa/Tonga, it is likely that the
central haplotype exists in this region. Further sampling in the south-eastern
corridor, from the Austral Islands through to the Pitcairn and Gambier Islands
would also be recommended to assess the continuity of the Easter Is/Austral Is
haplotype, distributed as part of a proposed south-east colonisation wave out of
central East Polynesia. Given the high molecular resolution afforded by complete
mitochondrial sequencing, it would be interesting to expand this study into Hawaii
and New Zealand to investigate multiple reported colonisations and where they
originate. An expansion of modern sampling outside of the Society Islands/Cook
Islands would also be useful in investigating multiple post-colonisation interaction
spheres across the region. Commensal studies can provide invaluable genetic
evidence that used in conjunction with archaeological, linguistic and ethnographic
data can evaluate models of human origins and dispersals across the Pacific.
77
6 Supplementary material
Table 6.1 Variable positions across 203bp of the control region (from
position 15,392
15,595). Colour coding: Adenine, red; Guanine, Blue; Cytosine,
yellow; Thymine, Green.
Amplified Haplo- Haplo- Papua Puk001 Thaifragment group group New
land
position
I (R1) II (R2) Guinea
15,435
15,478
15,498
15,508
15,510
15,528
15,544
15,552
15,553
15,555
15,562
15,572
15,579
15,582
15,583
C
T
T
A
C
C
A
T
A
T
T
C
T
C
A
T
C
T
A
C
T
A
C
A
C
A
C
T
T
A
T
T
A
C
A
A
C
A
A
A
A
C
A
C
A
A
A
Amplified
fragment
position
15,435
15,478
15,498
15,508
15,510
15,528
15,544
15,552
15,553
15,555
15,562
15,572
15,579
15,582
15,583
MS10596
MS10591
MS10601
MS10600
C
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
T
C
N
N
N
N
N
N
N
N
N
N
N
N
C
A
N
N
N
N
N
N
N
N
N
N
MS10602
N
N
N
N
N
N
N
N
N
N
78
Haplogroup
IIIA
(R15)
C
C
T
G
C
T
G
C
A
T
A
C
T
T
C
Haplogroup
IIIB
(R9)
C
C
T
G
C
T
G
C
G
T
A
C
T
T
C
Raia45 Tah8
C
C
T
A
C
T
G
C
G
T
A
C
T
T
C
MS10606
C
N
N
Raia28
A
Amplified
fragment
position
15,435
15,478
15,498
15,508
15,510
15,528
15,544
15,552
15,553
15,555
15,562
15,572
15,579
15,582
15,583
MS10523
Raia16 MS10592
N
C
C
T
G
T
N
G
C
G
T
A
C
T
T
C
Amplified
fragment
position
15,435
15,478
15,498
15,508
15,510
15,528
15,544
15,552
15,553
15,555
15,562
15,572
15,579
15,582
15,583
Tah3
Tah4
N
N
N
MS10609
MS10610
MS10604
MS10605
MS10502
N
N
N
Hua40 Tah2
C
C
T
G
C
T
G
C
G
T
A
C
T
T
C
79
Raia27 Hua24 Tet60.6 Hua23
Amplified
fragment
position
15,435
15,478
15,498
15,508
15,510
15,528
15,544
15,552
15,553
15,555
15,562
15,572
15,579
15,582
15,583
Hua19 Hua27 Tah5
Tah1
Rino1
Hua32 Hua12 Raia21
Amplified
fragment
position
15,435
15,478
15,498
15,508
15,510
15,528
15,544
15,552
15,553
15,555
15,562
15,572
15,579
15,582
15,583
Raia47 Hua47 Hua44 Hua43 Hua41 Raia17 Rama
24
C
C
T
G
C
T
G
C
G
T
A
C
T
T
C
C
C
T
G
C
T
G
C
G
T
A
C
T
T
C
80
Rama
3
Amplified
fragment
position
15,435
15,478
15,498
15,508
15,510
15,528
15,544
15,552
15,553
15,555
15,562
15,572
15,579
15,582
15,583
Rama
14
Hua13
Hua29
New
Zealand
C
C
T
G
C
T
G
C
G
T
A
C
T
T
C
81
Table 6.2 Archaeological sample information. If samples were successfully amplified for sequencing they were
given a database code. Only samples above 80% coverage were retained for phylogenetic analysis.
Location
Society Islands
Site
Terre Teahutapu, Plage
Terre Namotu, Maupiti
Haranai Valley, Maupiti
Archaeological Sample Code
Database
Code
MAU-1 N99E100 B1 UB 80
MS10451
MAU-1 N99E160 C1 UB 827-828
MAU-1 N99E100 A2 UB 59
MAU-1 N97E98 B5 UB 347
MAU-1 N100E100 B2 590-592
MAU-1 N99E100
MAU-1 N99E96 B1
MAU-1 N100E100 B3
MAU-1 N99E96 B3
MAU-1 N98E98 C1
MAU-1 N96E98 B3
MAU-1 N96E98 B4
MAU-1 N98E98 B3
MAU-1 N97E98 B7
MAU-5 N105 E98.5 C2
MAU-5 N100 E100 C2
MAU-5 N105 E98.50 C4
MAU-5 N105 E98.5 C1
MAU-11 TP1 B4
MS10452
MS10453
MS10454
MS10455
MS10457
82
Coverage
> 80%
MS10463
MS10478
MS10482
MS10503
MS10502
MS10506
Yes
Haumi archaeological site,
Moorea
American Samoa
Ofu Village, Ofu
Olosega Village, Olosega
Cook Islands
Aitutaki
MAU-11 TP2 A4
MAU-11 TP1 B8
MAU-11 TP2 A3
MS10511
MS10507
SCMO350 N95 E113 A2
MS10531
SCMO350 N95 E118 A1
SCMO350 N95 E116 C2
SCMO350 N95 E120 C4
SCMO350 TP2 A3
SCMO350 N95 E119 C2
AS-13-41 XU-4 Layer V #57
AS-13-41 XU-4 Layer III Level 6 #63
AS-13-41 XU-4 Layer V Level 3 #46
AS-13-41 XU-3 Layer X Level 1 #7
AS-13-41 XU-4 Layer VI Level 7 #67
AS-13-41 XU-3 Level VI Level 4 #15
AS-12-3 13W/10N Layer V Level 10
AS-12-3 14W/10N Layer V Level 9
AS-12-3 11W/10N Layer V Level 8
AS-12-3 1E/21N Layer IV Level 6
AS-12-3 14W/10N Layer VI Level 12
AS-12-3 IE/21N Layer V Level 8
Ait10 TP7 VII/7 #232
Ait10 TP1 VI,VII/6 #22
Ait10 TP5 IV/4 #146
Ait10 TP10 V15 #277
83
MS10523
MS10518
MS10549
MS10543
Yes
Mangaia
Tokelau
Fakaofo
Tonga
Faleloa, Foa
Mele Havea, Ha'afeva
Austral Islands
Atiahara, Tubuai
Ait10 TP3 V16 #102.1
Ait10 TP8 VII/6 #245
AI 737
AI 740
AI 739
AI 742
Fakaofo Sq.4 Layer D26
Fakaofo Sq.4 Layer C1
Fakaofo Sq.2 Layer C4
Fakaofo Sq.2 Layer C2
Fakaofo Sq.4 Layer D5
Fakaofo Sq.4 Layer D4
Tonga:Ha'ano:Ha1 Pukotala site unit 14
Level 7
Tonga:Ha'ano:Ha1 Pukotala site unit 14
Level 6
Tonga:Foa:Fo1 Faleloa site Unit 18 Level 8
Tonga:Foa:Fo1 Faleloa site Unit 18 Level 2
Tonga:Ha'afena:Hf1 Mele Havea site Unit
2 Level 7
Tub452 AT1.1 Sq.B6 Layer 4 134cm BD
Tub452 AT1.1 Section B Layer 7 AB14 166
BD
Tub452 AT1.1 Section A Sq.09 Layer 5
123-30 BD
Tub452 AT1.1 Sq.D5 L4 137.46cm BD
Tub452 AT1.1 Section A Sq.M5 152 BD
84
MS10592
MS10593
MS10591
MS10594
Yes
Yes
MS10595
MS10596
MS10597
Yes
MS10598
MS10600
Yes
MS10601
Yes
MS10599
Easter Island
Anakena
Layer 5
Anakena North Unit 1 UH ID#0033
Anakena North Unit UH ID#0015
Anakena North Unit 1 UH ID#0005
Anakena North Unit 1 UH ID#0030
Anakena North Unit 1 UH ID#0028
Anakena North Unit 1 UH ID#0022
Anakena North Unit 1 UH ID#0013
Anakena North Unit 1 UH ID#0018
Anakena North Unit 1 UH ID#0024
Anakena North Unit 1 UH ID#0010
Anakena North Unit 1 UH ID#0020
85
MS10602
MS10603
MS10604
MS10605
MS10606
MS10607
MS10608
MS10609
MS10610
MS10611
Yes
Yes
Yes
Yes
Yes
Yes
Table 6.3 Modern sample information. Only samples that were successfully
amplified in 4 fragments were sequenced. Samples that exhibited high coverage
above 90% were retained for phylogenetic analysis.
Location
Site
Cook Islands
Pukapuka
Atiu
Aitutaki
Society Islands
Tetiaroa
Modern sample
code
Puk001
Puk002
Puk003
Puk004
Puk005
Puk006
Puk007
Atiu1A
Atiu2A
Atiu3A
Rino1
Rama24
Rama3
Rama14
Tet27.6
Tet60.6
Tet13.6
Tet4.6
Tet11.6
Tet9.6
Tet24.6
Tet3.6
Tet12.6
Tet74.6
Tet5.6
Tet10.6
Tet6.6
Tet15.6
Tet29.6
Tet25.6
Tet14.6
Tet28.6
Tet65.6
Tet78.6
Tet7.6
Tet64.6
Tet18.6
86
Yes
Coverage >
90%
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Sequenced
Huahine
Tahiti
Raiatea
Tet2.6
Tet75.6
Tet31.6
Tet21.6
Tet16.6
Tet26.6
Tet1.6
Tet23.6
Tet57.6
Tet8.6
Tet22.6
Tet30.6
Tet17.6
Hua23
Hua19
Hua40
Hua43
Hua13
Hua41
Hua18
Hua29
Hua24
Hua47
Hua12
Hua36
Hua44
Hua11
Hua30
Hua39
Hua26
Hua42
Hua28
Hua27
Hua32
Tah5
Tah1
Tah2
Tah3
Tah4
Tah8
Tah6
Raia28
Raia16
Raia21
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
87
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Raia47
Raia6
Raia23
Raia25
Raia45
Raia22
Raia3
Raia26
Raia29
Raia8
Raia46
Raia27
Raia17
Samoa
Apia,
Upolu
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Apia001
Apia002
Togitogigh4
88
Yes
Yes
Yes
Yes
7 Reference List
Alberts, J. R., & Galef, B. G. (1973). Olfactory cues and movement: Stimuli mediating
intraspecific aggression in the wild Norway rat. Journal of Comparative and
Physiological Psychology, 85(2), 233-242. doi: 10.1037/h0035050.
Allen, M. (1984). Elders, Chiefs, and Big Men: Authority legitimation and political
evolution
in
Melanesia.
American
Ethnologist,
11(1),
20-41.
doi:
10.1525/ae.1984.11.1.02a00020.
Allen, M., & Steadman, D. (1990). Excavations at the Ureia site, Aitutaki, Cook Islands:
preliminary results. Archaeology in Oceania, 25(1), 24-37.
Allen, M. S. (2014). Marquesan colonisation chronologies and post-colonisation
hypothesis. Journal of Pacific Archaeology, 5(2).
Allen, M. S., & Johnson, K. T. (1997). Tracking ancient patterns of interaction: Recent
geochemical studies in the southern Cook Islands. In M. Weisler (Ed.),
Prehistoric Long-distance Interaction in Oceania: an interdisciplinary approach
(Vol. 21, pp. 111-113): New Zealand Archaeological Association Monograph.
Allen, M. S., & Wallace, R. (2007). New evidence from the East Polynesian gateway:
Substantive and methodological results from Aitutaki, southern Cook Islands.
Radiocarbon, 49(3), 1163-1180.
Anderson, A., & Sinoto, Y. H. (2002). New Radiocarbon Ages of Colonization Sites in
East Polynesia. Asian Perspectives, 41(2), 242-257.
Bandelt, H. J., Forster, P., & Röhl, A. (1999). Median-joining networks for inferring
intraspecific phylogenies. Molecular Biology and Evolution, 16(1), 37-48.
Barnes, S. S., Matisoo-Smith, E., & Hunt, T. L. (2006). Ancient DNA of the Pacific rat
(Rattus exulans) from Rapa Nui (Easter Island). Journal of Archaeological
Science, 33(11), 1536-1540. doi: 10.1016/j.jas.2006.02.006.
Barnett, S. A. (1955). Competition among Wild Rats. Nature, 175(4446), 126. doi:
10.1038/175126b0.
Bellwood, P. S. (1970). Dispersal centers in East Polynesia, with special reference to the
Society and Marquesas Islands. In R. C. Green & M. Kelly (Eds.), Studies in
89
Oceanic culture history. Bernice Pauahi Bishop Museum.
Bellwood, P. S. (1978). The Polynesians: prehistory of an island people. London:
Thames and Hudson.
Best, S. (1988). Tokelau archaeology: a preliminary report of an initial survey and
excavations. Bulletin of the Indo-Pacific Prehistory Association, 8, 104-118.
Best, S., Sheppard, P., Green, R., & Parker, R. (1992). Necromancing the stone:
Archaeologists and adzes in Samoa. The Journal of the Polynesian Society,
101(1), 45-85.
Biggs, B. (1972). Implications of Linguistic Subgroup with Special Reference to
Polynesia. Pacific Anthropological Records, 13, 143-152.
Bollongino, R., Tresset, A., & Vigne, J. D. (2008). Environment and excavation: Pre-lab
impacts on ancient DNA analyses. Comptes Rendus Palevol, 7(2), 91-98.
Boocock,
J.
(2014).
Ancient
DNA
Pipeline.
from
GitHub
Repository,
https://github.com/theboocock/ancient_dna_pipeline
Briggs, A. W., Stenzel, U., Johnson, P. L., Green, R. E., Kelso, J., Prüfer, K., . . . Lachmann,
M. (2007). Patterns of damage in genomic DNA sequences from a Neandertal.
Proceedings of the National Academy of Sciences, 104(37), 14616-14621.
Briggs, A. W., Stenzel, U., Meyer, M., Krause, J., Kircher, M., & Pääbo, S. (2010).
Removal of deaminated cytosines and detection of in vivo methylation in
ancient DNA. Nucleic Acids Research, 38(6), e87.
Brotherton, P., Endicott, P., Sanchez, J. J., Beaumont, M., Barnett, R., Austin, J., &
Cooper, A. (2007). Novel high-resolution characterization of ancient DNA
reveals C> U-type base modification events as the sole cause of post mortem
miscoding lesions. Nucleic Acids Research, 35(17), 5717-5728.
Buck, P. H. (1944). Arts and crafts of the Cook Islands. Honolulu, Hawaii: Bernice P.
Bishop Museum.
Burley, D., Edinborough, K., Weisler, M., & Zhao, J.-X. (2015). Bayesian modeling and
chronological precision for Polynesian settlement of Tonga. PLoS One, 10(3),
e0120795. doi: 10.1371/journal.pone.0120795.
Burley, D. V., Nelson, D. E., & Shutler, R. (1999). A radiocarbon chronology for the
Eastern Lapita frontier in Tonga. Archaeology in Oceania, 34(2), 59-70. doi:
10.1002/j.1834-4453.1999.tb00429.x.
90
Burney, D., & Kikuchi, W. (2006). A Millennium of Human Activity at Makauwahi Cave,
Human Ecology, 34(2), 219-247. doi: 10.1007/s10745006-9015-3.
Burney, D. A., James, H. F., Burney, L. P., Olson, S. L., Kikuchi, W., Wagner, W. L., . . .
Nishek, R. (2001). Fossil evidence for a diverse biota from Kaua'i and its
transformation since human arrival. Ecological Monographs, 71(4), 615-641.
Burrows, E. G. (1938). Western Polynesia: a study of cultural differentiation (Vol. 7).
Goteburg: Etnologiska Studier. (Reprinted 1970 with same pagination by
University Bookshop, Dunedin, NZ).
Clark, G. R., Reepmeyer, C., Melekiola, N., Woodhead, J., Dickinson, W. R., &
Martinsson-Wallin, H. (2014). Stone tools from the ancient Tongan state reveal
prehistoric interaction centers in the Central Pacific. Proceedings of the
National Academy of Sciences, 111(29), 10491-10496.
Clark, J. T., Quintus, S., Weisler, M., St Pierre, E., Nothdurft, L., & Feng, Y. (2016).
Refining the chronology for West Polynesian colonization: New data from the
Samoan archipelago. Journal of Archaeological Science: Reports, 6, 266-274.
doi: 10.1016/j.jasrep.2016.02.011.
artefact transfer within and beyond the archipelago. Archaeology in Oceania.
doi: 10.1002/arco.5090.
Connell, J. F., & Allen, J. (2004). Dating the colonization of Sahul (Pleistocene Australia
New Guinea): a review of recent research. Journal of Archaeological Science,
31(6), 835-853. doi: 10.1016/j.jas.2003.11.005.
Conte, E., & Kirch, P. V. (2004). Archaeological investigations in the Mangareva Islands
(Gambier Archipelago), French Polynesia. Berkeley, Calif.: Archaeological
Research Facility, University of California, Berkeley.
Dabney, J., Meyer, M., & Pääbo, S. (2013). Ancient DNA damage. Cold Spring Harbor
Perspectives in Biology, 5(7), a012567.
Darriba, D., Taboada, G. L., Doallo, R., & Posada, D. (2012). jModelTest 2: more models,
new heuristics and parallel computing. Nature Methods, 9(8), 772-772.
Davidson, J. (1976). The Prehistory of Western Polynesia. In J. Garanger (Ed.), La
Prehistorie Océanienne (pp. 27-51). Nice: Union Internationale des Sciences
91
Prehistoriques et Protohistoriques.
Denham, T., Ramsey, C. B., & Specht, J. (2012). Dating the appearance of Lapita pottery
in the Bismarck Archipelago and its dispersal to Remote Oceania. Archaeology
in Oceania, 47(1), 39-46. doi: 10.1002/j.1834-4453.2012.tb00113.x.
Derrington, I. M., Butler, T. Z., Collins, M. D., Manrao, E., Pavlenok, M., Niederweis, M.,
& Gundlach, J. H. (2010). Nanopore DNA sequencing with MspA. Proceedings of
the National Academy of Sciences of the United States of America, 107(37),
16060. doi: 10.1073/pnas.1001831107.
DeSalle, R., Gatesy, J., Wheeler, W., & Grimaldi, D. (1992). DNA Sequences from a
Fossil Termite in Oligo-Miocene Amber and Their Phylogenetic Implications.
American Journal of Physical Anthropology, 87, 291.
Dickinson, W. R., & Green, R. C. (1998). Geoarchaeological context of Holocene
subsidence at the ferry berth Lapita site, Mulifanua, Upolu, Samoa.
Geoarchaeology, 13(3), 239-263.
Duggan, Ana T., Evans, B., Friedlaender, Françoise r., Friedlaender, Jonathan s., Koki,
G., Merriwether, D. a., . . . Stoneking, M. (2014). Maternal History of Oceania
from Complete mtDNA Genomes: Contrasting Ancient Diversity with Recent
Homogenization Due to the Austronesian Expansion. The American Journal of
Human Genetics, 94(5), 721-733. doi: 10.1016/j.ajhg.2014.03.014.
Dumont d'Urville, J. (1832). Sur les îles du Grand Océan. Bulletin de la Société de
Géographie, 17, 1-21.
Dye, T. S., & Dickinson, W. R. (1996). Sources of sand tempers in prehistoric Tongan
pottery. Geoarchaeology, 11(2), 141-164.
Edgar, R. C. (2004). MUSCLE: multiple sequence alignment with high accuracy and high
throughput. Nucleic Acids Research, 32(5), 1792-1797.
Edward, M. G., David, E. G., Michael, T. C., Charles, J. S., Mary, D., David, H., & Gerard,
Z. (1990). Chloroplast DNA sequence from a Miocene Magnolia species. Nature,
344(6267), 656. doi: 10.1038/344656a0.
Elbert, S. H. (1982). Lexical diffusion in Polynesia and the Marquesan-Hawaiian
relationship. The Journal of the Polynesian Society, 91(4), 499-517.
Emory, K. P. (1946). Eastern Polynesia: its cultural relationships. (PhD), Yale University.
Emory, K. P., & Sinoto, Y. H. (1965). Preliminary report on the archaeological
92
investigations in Polynesia : field work in the Society and Tuamotu Islands,
French Polynesia, and American Samoa in 1962, 1963, 1964. . Honolulu:
Department of Anthropology, Bernice P. Bishop Museum.
Felsenstein, J. (1981). Evolutionary trees from DNA sequences: a maximum likelihood
approach. Journal of Molecular Evolution, 17(6), 368-376.
Flannery, T. F. (1995). Mammals of the south-west Pacific & Moluccan Islands: Cornell
University Press.
Friedlaender, J. S., Gentz, F., Green, K., & Merriwether, D. A. (2002). A Cautionary Tale
on Ancient Migration Detection: Mitochondrial DNA Variation in Santa Cruz
Islands, Solomon Islands. Human Biology, 74(3), 453-471.
Green, R. (1966). Linguistic subgrouping within Polynesia: the implications for
prehistoric settlement. The Journal of the Polynesian Society, 75(1), 6-38.
Green, R. C. (1981). Location of the Polynesian homeland: a continuing problem. In J.
Hollyman & A. Pawley (Eds.), Studies in Pacific languages and cultures in honour
of Bruce Biggs (pp. 133-158). Auckland: Linguistic Society of New Zealand.
Green, R. C. (1991). Near and remote Oceania
history. In A. Pawley (Ed.), Man and a half: essays in Pacific anthropology and
ethnobiology in honour of Ralph Bulmer (pp. 491-502). Auckland.
Green, R. C., Jones, M., & Sheppard, P. (2008). The reconstructed environment and
Archaeology
in
Oceania,
43(2),
49-61.
doi:
10.1002/j.1834-
4453.2008.tb00030.x.
Green, R. E., Briggs, A. W., Krause, J., Prüfer, K., Burbano, H. A., Siebauer, M., . . .
Pääbo, S. (2009). The Neandertal genome and ancient DNA authenticity. EMBO
Journal, 28(17), 2494-2502. doi: 10.1038/emboj.2009.222.
Greig, K., Boocock, J., Prost, S., Horsburgh, K. A., Jacomb, C., Walter, R., & MatisooSmith, E. (2015). Complete Mitochondrial Genomes of New Zealand's First
Dogs. PLoS One, 10(10), e0138536. doi: 10.1371/journal.pone.0138536.
Groube, L., Chappell, J., Muke, J., & Price, D. (1986). A 40,000 year-old human
occupation site at Huon Peninsula, Papua New Guinea. Nature, 324(6096), 453.
doi: 10.1038/324453a0.
Groube, L. M. (1971). Tonga, Lapita pottery, and Polynesian origins. The Journal of the
93
Polynesian Society, 80(3), 278-316.
Guindon, S., Dufayard, J.-F., Lefort, V., Anisimova, M., Hordijk, W., & Gascuel, O.
(2010). New algorithms and methods to estimate maximum-likelihood
phylogenies: assessing the performance of PhyML 3.0. Systematic Biology,
59(3), 307-321.
Guindon, S., & Gascuel, O. (2003). A simple, fast, and accurate algorithm to estimate
large phylogenies by maximum likelihood. Systematic Biology, 52(5), 696-704.
Haddon, A. C. (1975). Canoes of Oceania. Honolulu, Hawaii: Bishop Museum Press.
Hage, P., & Marck, J. (2003). Matrilineality and the Melanesian Origin of Polynesian Y
Chromosomes. Current Anthropology, 44(S5), S121-S127. doi: 10.1086/379272.
Hagelberg, E., Quevedo, S., Turbon, D., & Clegg, J. (1994). DNA from ancient Easter
Islanders. Nature, 369(6475), 25-26.
Harlow, R. (1979). Regional variation in Maori. New Zealand Journal of Archaeology, 1,
123-138.
Hasegawa, M., Kishino, H., & Yano, T. A. (1985). Dating of the human-ape splitting by a
molecular clock of mitochondrial DNA. Journal of Molecular Evolution, 22(2),
160-174.
Heyn, P., Stenzel, U., Briggs, A. W., Kircher, M., Hofreiter, M., & Meyer, M. (2010).
Road blocks on paleogenomes- polymerase extension profiling reveals the
frequency of blocking lesions in ancient DNA. Nucleic Acids Research, 38(16),
e161. doi: 10.1093/nar/gkq572.
Hill, A., Gentile, B., Bonnardot, J., Roux, J., Weatherall, D., & Clegg, J. (1987). Polynesian
origins and affinities: globin gene variants in eastern Polynesia. American
Journal of Human Genetics, 40(5), 453.
Hofreiter, M., Serre, D., Poinar, H. N., Kuch, M., & Pääbo, S. (2001). Ancient DNA.
Nature Reviews Genetics, 2(5), 353-359.
Huelsenbeck, J. P., & Ronquist, F. (2001). MRBAYES: Bayesian inference of phylogenetic
trees. Bioinformatics, 17(8), 754-755.
Hunt, T. L. (1979). Alternative Perspectives for Early East Polynesian prehistory. Paper
presented at the New Zealand Archaeological Association Biennial Conference,
Auckland.
Hunt, T. L., & Lipo, C. P. (2006). Late colonization of Easter Island. Science, 311(5767),
94
1603.
Hunt, T. L., & Lipo, C. P. (2012). Ecological Catastrophe and Collapse: The Myth of
'Ecocide' on Rapa Nui (Easter Island). PERC Research Paper (12/3).
Irwin, G. (1981). How Lapita lost its pots: the question of continuity in the colonisation
of Polynesia. The Journal of the Polynesian Society, 90(4), 481-494.
Irwin, G. (1992). The Prehistoric Exploration and Colonization of the Pacific. Cambridge:
Cambridge University Press.
Irwin, G. (2008). Pacific Seascapes, Canoe Performance, and a Review of Lapita
Voyaging with Regard to Theories of Migration. Asian Perspectives, 47(1), 1227.
Jennings, J. D. (1979). The Prehistory of Polynesia. Cambridge, Mass.: Harvard
University Press.
Jónsson, H., Ginolhac, A., Schubert, M., Johnson, P. L. F., & Orlando, L. (2013).
mapDamage2.0: Fast approximate Bayesian estimates of ancient DNA damage
parameters.
Bioinformatics,
29(13),
1682-1684.
doi:
10.1093/bioinformatics/btt193
Jordan, F. M., Gray, R. D., Greenhill, S. J., & Mace, R. (2009). Matrilocal residence is
ancestral in Austronesian societies. Proceedings of the Royal Society of London
B: Biological Sciences, 276(1664), 1957. doi: 10.1098/rspb.2009.0088
Kahn, J. G. (2012). Coastal Occupation at the GSIslands. Journal of Pacific Archaeology, 3(2), 52-61.
Kahn, J. G., Dotte-Sarout, E., Molle, G., & Conte, E. (2015). Mid-to late prehistoric
landscape change, settlement histories, and agricultural practices on Maupiti,
Society Islands (Central Eastern Polynesia). The Journal of Island and Coastal
Archaeology, 10(3), 363-391.
Kayser, M. (2010). The Human Genetic History of Oceania: Near and Remote Views of
Dispersal. Current Biology, 20(4), R194-R201. doi: 10.1016/j.cub.2009.12.004
Kayser, M., Brauer, S., Weiss, G., Underhill, P., Roewer, L., Schiefenhövel, W., &
Stoneking, M. (2000). Melanesian origin of Polynesian Y chromosomes. Current
Biology, 10(20), 1237-1246. doi: 10.1016/S0960-9822(00)00734-X.
Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., . . . Duran,
C. (2012). Geneious Basic: an integrated and extendable desktop software
95
platform for the organization and analysis of sequence data. Bioinformatics,
28(12), 1647-1649.
Kemp, B. M., & Smith, D. G. (2005). Use of bleach to eliminate contaminating DNA
from the surface of bones and teeth. Forensic Science International, 154(1), 5361.
Kirch, P., & Ellison, J. (1994). Palaeoenvironmental evidence for human colonization of
remote Oceanic islands. Antiquity, 68(259), 310.
Kirch, P. V. (1986). Rethinking East Polynesian Prehistory. The Journal of the Polynesian
Society, 95(1), 9-40.
Norwegian Archaeological Review, 21(2), 103-117.
Kirch, P. V. (1996). Late Holocene human-induced modifications to a central Polynesian
island ecosystem. Proceedings of the National Academy of Sciences, 93(11),
5296-5300.
Kirch, P. V. (1997). The Lapita Peoples: Ancestors of the Oceanic World. The Peoples of
South-East Asia and the Pacific: Blackwell, Cambridge, MA.
Kirch, P. V. (2000). On the road of the winds : an archaeological history of the Pacific
Islands before European contact. Berkeley, California: University of California
Press.
Kirch, P. V., Steadman, D. W., Butler, V. L., Hather, J., & Weisler, M. I. (1995). Prehistory
and human ecology in Eastern Polynesia: Excavations at Tangatatau
Rockshelter, Mangaia, Cook Islands. Archaeology in Oceania, 30(2), 47-65. doi:
10.1002/j.1834-4453.1995.tb00330.x
Knapp, M., & Hofreiter, M. (2010). Next Generation Sequencing of Ancient DNA:
Requirements,
Strategies
and
Perspectives.
Genes,
1(2),
227.
doi:
10.3390/genes1020227.
Knapp, M., Stiller, M., & Meyer, M. (2012). Generating barcoded libraries for multiplex
high-throughput sequencing. In B. Shapiro & M. Hofreiter (Eds.), Ancient DNA:
Methods and Protocols (pp. 155-170): Humana Press.
Larson, G., Cucchi, T., Fujita, M., Matisoo-Smith, E., Robins, J., Anderson, A., . . .
Dobney, K. (2007). Phylogeny and ancient DNA of Sus provides insights into
neolithic expansion in Island Southeast Asia and Oceania. Proceedings of the
96
National Academy of Sciences of the United States of America, 104, 4834-4839.
Leach, B. F., Davidson, J. M., & Horwood, L. M. (1994). Analaysis of faunal material
from Ngaaitutaki, Tepaopao and Erua, Mangaia, Cook Islands Museum of New
Zealand Te Papa Tongarewa Technical Report (Vol. 1, pp. 43).
Leavesley, M. G., Bird, M. I., Fifield, L. K., Hausladen, P. A., Santos, G. M., & Di Tada, M.
L. (2002). Buang Merabak: Early Evidence for Human Occupation in the
Bismarck Archipelago, Papua New Guinea. Australian Archaeology (54), 55-57.
plotype network
construction. Methods in Ecology and Evolution, 6(9), 1110-1116.
Leonard, J. A., Shanks, O., Hofreiter, M., Kreuz, E., Hodges, L., Ream, W., . . . Fleischer,
R. C. (2007). Animal DNA in PCR reagents plagues ancient DNA research.
Journal
of
Archaeological
Science,
34(9),
1361-1366.
doi:
10.1016/j.jas.2006.10.023
Li, H. (2013). Aligning sequence reads, clone sequences and assembly contigs with
BWA-MEM. arXiv preprint arXiv:1303.3997.
Li, H., & Durbin, R. (2009). Fast and accurate short read alignment with Burrows
Wheeler transform. Bioinformatics, 25(14), 1754-1760.
Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., . . . Durbin, R. (2009).
The sequence alignment/map format and SAMtools. Bioinformatics, 25(16),
2078-2079.
Librado, P., & Rozas, J. (2009). DnaSP v5: a software for comprehensive analysis of DNA
polymorphism data. Bioinformatics, 25(11), 1451-1452.
Lindahl, T. (1993). Instability and decay of the primary structure of DNA. Nature,
362(6422), 709-715.
Lindgreen, S. (2012). AdapterRemoval: easy cleaning of next-generation sequencing
reads. BMC Research Notes, 5(1), 1.
Maricic, T., Whitten, M., & Pääbo, S. (2010). Multiplexed DNA sequence capture of
mitochondrial genomes using PCR products. PLoS One, 5(11), e14004.
Matisoo-Smith, E. (1994). The human colonisation of Polynesia. A novel approach:
genetic analyses of the Polynesian rat (Rattus exulans). The Journal of the
Polynesian Society, 103(1), 75-87.
Matisoo-Smith, E. (2002). Something Old, Something New: Do Genetic Studies of
97
Contemporary Populations Reliably Represent Prehistoric Populations of Pacific
Rattus exulans? Human Biology, 74(3), 489-496.
Matisoo-Smith, E. (2015). Ancient DNA and the human settlement of the Pacific: A
review.
Journal
of
Human
Evolution,
79,
93-104.
doi:
10.1016/j.jhevol.2014.10.017.
Ancient DNA from Polynesian rats: extraction, amplification and sequence from
single small bones. Electrophoresis, 18(9), 1534-1537.
Matisoo-Smith, E., Roberts, R. M., Irwin, G. J., Allen, J. S., Penny, D., & Lambert, D. M.
(1998). Patterns of Prehistoric Human Mobility in Polynesia Indicated by
mtDNA from the Pacific Rat. Proceedings of the National Academy of Sciences
of the United States of America, 95(25), 15145-15150.
Matisoo-Smith, E., & Robins, J. (2008). Mitochondrial DNA evidence for the spread of
Pacific rats through Oceania. Biological Invasions, 11(7), 1521-1527. doi:
10.1007/s10530-008-9404-1.
Matisoo-Smith, E., Robins, J. H., & Green, R. C. (2004). Origins and Dispersals of Pacific
Peoples: Evidence from mtDNA Phylogenies of the Pacific Rat. Proceedings of
the National Academy of Sciences of the United States of America, 101(24),
9167-9172.
Matisoo-Smith, L. (1996). No hea te kiore: MtDNA variation in Rattus exulans: a model
for human colonisation and contact in prehistoric Polynesia. (PhD), The
University of Auckland.
McKenna, A., Hanna, M., Banks, E., Sivachenko, A., Cibulskis, K., Kernytsky, A., . . . Daly,
M. (2010). The Genome Analysis Toolkit: a MapReduce framework for analyzing
next-generation DNA sequencing data. Genome Research, 20(9), 1297-1303.
Melton, T., Peterson, R., Redd, A. J., Saha, N., Sofro, A., Martinson, J., & Stoneking, M.
(1995). Polynesian genetic affinities with Southeast Asian populations as
identified by mtDNA analysis. American Journal of Human Genetics, 57(2), 403.
Milham, P., & Thompson, P. (1976). Relative antiquity of human occupation and
extinct fauna at Madura Cave, southeastern Western Australia. Mankind, 10(3),
175-180.
Mitchell, D., Willerslev, E., & Hansen, A. (2005). Damage and repair of ancient DNA.
98
Mutation Research: Fundamental and Molecular Mechanisms of Mutagenesis,
571(1), 265-276.
Molle, G., & Conte, E. (2011). New Perspectives on the Occupation of Hatuana Dune
Site, Ua Huka, Marquesas Islands. Journal of Pacific Archaeology, 2(2), 103-108.
Nei, M. (1987). Molecular evolutionary genetics. New York: Columbia University Press.
Nielsen-Marsh, C. M., & Hedges, R. E. (2000). Patterns of diagenesis in bone I: the
effects of site environments. Journal of Archaeological Science, 27(12), 11391150.
Noonan, J. P., Hofreiter, M., Smith, D., Priest, J. R., Rohland, N., Rabeder, G., . . . Rubin,
E. M. (2005). Genomic sequencing of Pleistocene cave bears. Science (New
York, N.Y.), 309(5734), 597.
Connell, J. F., & Allen, J. (2004). Dating the colonization of Sahul (Pleistocene Australia
New Guinea): a review of recent research. Journal of Archaeological Science,
31(6), 835-853. doi:10.1016/j.jas.2003.11.005
seafaring. In A. Anderson, J. H. Barrett & K. V. Boyle (Eds.), The global origins
and development of seafaring (pp. 57-68). Cambridge, UK: McDonald Institute
for Archaeological Research, University of Cambridge.
Orlando, L., Ginolhac, A., Raghavan, M., Vilstrup, J., Rasmussen, M., Magnussen, K., . . .
Willerslev, E. (2011). True single-molecule DNA sequencing of a Pleistocene
horse bone. Genome Research, 21(10), 1705-1719.
Oskarsson, M. C. R., Klütsch, C. F. C., Boonyaprakob, U., Wilton, A., Tanabe, Y., &
Savolainen, P. (2012). Mitochondrial DNA data indicate an introduction through
Mainland Southeast Asia for Australian dingoes and Polynesian domestic dogs.
Proceedings of the Royal Society of London B: Biological Sciences, 279(1730),
967-974. doi: 10.1098/rspb.2011.1395.
Overballe-Petersen, S., Orlando, L., & Willerslev, E. (2012). Next-generation sequencing
offers new insights into DNA degradation. Trends in Biotechnology, 30(7), 364368. doi: 10.1016/j.tibtech.2012.03.007.
Pääbo, S. (1989). Ancient DNA: extraction, characterization, molecular cloning, and
enzymatic amplification. Proceedings of the National Academy of Sciences,
86(6), 1939-1943.
99
Pawley, A. (1966). Polynesian languages: A subgrouping based on shared innovations
in morphology. The Journal of the Polynesian Society, 75(1), 39-64.
Pawley, A. (2015). Linguistic Evidence as a Window into the Prehistory of Oceania. In E.
Cochrane & T. L. Hunt (Eds.), The Oxford Handbook of Prehistoric Oceania.
Petchey, F. (2001). Radiocarbon determinations from the Mulifanua Lapita site, Upolu,
western Samoa. Radiocarbon, 43(1), 63-68.
Petchey, F., Addison, D. J., & McAlister, A. (2010). Re-interpreting old dates:
Radiocarbon determinations from the Tokelau Islands (South Pacific). Journal of
Pacific Archaeology, 1(2), 161-167.
Poinar, H., Schwarz, C., & Shapiro, B. (2006). Metagenomics to Paleogenomics: LargeScale Sequencing of Mammoth DNA. Science, 311(5759), 392-394.
Poinar, H. N., & Cooper, A. (2000). Ancient DNA: do it right or not at all. Science, 5482,
1139.
Pruvost, M., Schwarz, R., Correia, V. B., Champlot, S., Braguier, S., Morel, N., . . . Geigl,
E.-M. (2007). Freshly excavated fossil bones are best for amplification of
ancient DNA. Proceedings of the National Academy of Sciences, 104(3), 739744.
Pushkarev, D., Neff, N. F., & Quake, S. R. (2009). Single-molecule sequencing of an
individual
human
genome.
Nature
Biotechnology,
27(9),
847.
doi:
10.1038/nbt.1561.
Quintus, S., Clark, J. T., Day, S. S., & Schwert, D. P. (2016). Landscape Evolution and
Human Settlement Patterns on Ofu Island, Manu'a Group, American Samoa.
Asian Perspectives, 54(2), 208-237.
Ramakrishnan, U., & Hadly, E. A. (2009). Using phylochronology to reveal cryptic
population histories: review and synthesis of 29 ancient DNA studies. Molecular
Ecology, 18, 1310-1330. doi: 10.1111/j.1365-294X.2009.04092.x.
Redd, A. J., Takezaki, N., Sherry, S. T., McGarvey, S. T., Sofro, A., & Stoneking, M.
(1995). Evolutionary history of the COII/tRNALys intergenic 9 base pair deletion
in human mitochondrial DNAs from the Pacific. Molecular Biology and
Evolution, 12(4), 604-615.
Rieth, T. M., Hunt, T. L., Lipo, C., & Wilmshurst, J. M. (2011). The 13th century
Journal of Archaeological Science,
100
38(10), 2740-2749.
Rizzi, E., Lari, M., Gigli, E., De Bellis, G., & Caramelli, D. (2012). Ancient DNA studies:
new perspectives on old samples. Genetics Selection Evolution, 44(1), 1.
Robarts, E. (1974). The Marquesan journal of Edward Robarts, 1797-1824. Canberra:
Australian National University Press.
Robins, J., Matisoo-Smith, E., & Furey, L. (2001). Hit or miss? Factors affecting DNA
preservation in Pacific archaeological material. In M. Jones & P. Sheppard
(Eds.), Australasian Connections and New Directions: Proceedings of the 7th
Australasian Archaeometry Conference (Vol. 5, pp. 295-305).
Robins, J. H., McLenachan, P. A., Phillips, M. J., Craig, L., Ross, H. A., & Matisoo-Smith,
E. (2008). Dating of divergences within the Rattus genus phylogeny using whole
mitochondrial genomes. Molecular Phylogenetics and Evolution, 49(2), 460466.
Robins, J. H., McLenachan, P. A., Phillips, M. J., McComish, B. J., Matisoo-Smith, E., &
Ross, H. A. (2010). Evolutionary relationships and divergence times among the
native rats of Australia. BMC Evolutionary Biology, 10, 375-375. doi:
10.1186/1471-2148-10-375.
Robinson, J. T., Thorvaldsdóttir, H., Winckler, W., Guttman, M., Lander, E. S., Getz, G.,
& Mesirov, J. P. (2011). Integrative genomics viewer. Nature Biotechnology,
29(1), 24-26.
Rohland, N., & Hofreiter, M. (2007). Ancient DNA extraction from bones and teeth.
Nature Protocols, 2(7), 1756-1762.
extraction method for increased sample throughput. Molecular Ecology
Resources, 10(4), 677-683.
Rolett, B. V. (2002). Voyaging and interaction in ancient East Polynesia. Asian
Perspectives, 41(2), 182-194.
Ronquist, F., & Huelsenbeck, J. P. (2003). MrBayes 3: Bayesian phylogenetic inference
under mixed models. Bioinformatics, 19(12), 1572-1574.
Roullier, C., Benoit, L., McKey, D. B., & Lebot, V. (2013). Historical collections reveal
patterns of diffusion of sweet potato in Oceania obscured by modern plant
movements and recombination. Proceedings of the National Academy of
101
Sciences, 110, 2205-2210. doi: 10.1073/pnas.1211049110.
Sarah, B. N., Emily, H. T., Peggy, D. R., Steven, D. F., Abigail, W. B., Choli, L., . . . Jay, S.
(2009). Targeted capture and massively parallel sequencing of 12 human
exomes. Nature, 461(7261), 272. doi: 10.1038/nature08250.
Savolainen, P., Leitner, T., Wilton, A. N., Matisoo-Smith, E., & Lundeberg, J. (2004). A
detailed picture of the origin of the Australian dingo, obtained from the study
of mitochondrial DNA. Proceedings of the National Academy of Sciences of the
United States of America, 101(33), 12387-12390.
Sawyer, S., Krause, J., Guschanski, K., Savolainen, V., & Pääbo, S. (2012). Temporal
patterns of nucleotide misincorporations and DNA fragmentation in ancient
DNA. PLoS One, 7(3), e34131.
Schubert, M., Ermini, L., Der Sarkissian, C., Jónsson, H., Ginolhac, A., Schaefer, R., . . .
McCue, M. (2014). Characterization of ancient and modern genomes by SNP
detection and phylogenomic and metagenomic analysis using PALEOMIX.
Nature Protocol, 9(5), 1056-1082.
Schubert, M., Ginolhac, A., Lindgreen, S., Thompson, J. F., Al-Rasheid, K. A., Willerslev,
E., . . . Orlando, L. (2012). Improving ancient DNA read mapping against modern
reference genomes. BMC Genomics, 13(1), 1.
Sheppard, P. J., Walter, R., & Parker, R. J. (1997). Basalt sourcing and the development
of Cook Island exchange systems. In M. Weisler (Ed.), Prehistoric Long-distance
Interaction in Oceania: an interdisciplinary approach (pp. 85-110): New Zealand
Archaeological Association Monograph.
Simmons, D. (1973). Suggested periods in South Island prehistory. Records of the
Auckland Institute and Museum, 1-58.
Sinoto, Y. H. (1970). An archaeologically based assessment of the Marquesas Islands as
a dispersal center in East Polynesia. Studies in Oceanic culture history, 1, 105132.
Sinoto, Y. H. (1983). An analysis of Polynesian migrations based on the archaeological
assessments. Journal de la Société des Océanistes, 39(76), 57-67.
Slatkin, M., & Racimo, F. (2016). Ancient DNA and human history. Proceedings of the
National Academy of Sciences of the United States of America, 113(23), 6380.
doi:10.1073/pnas.1524306113
102
Smith, C. I., Chamberlain, A. T., Riley, M. S., Stringer, C., & Collins, M. J. (2003). The
thermal history of human fossils and the likelihood of successful DNA
amplification. Journal of Human Evolution, 45(3), 203-217.
Spriggs, M., & Anderson, A. (1993). Late colonization of East Polynesia. Antiquity,
67(255), 200.
Stoneking, M. (2000). Hypervariable Sites in the mtDNA Control Region Are Mutational
Hotspots. The American Journal of Human Genetics, 67(4), 1029-1032. doi:
10.1086/303092.
Storey, A. A., & Matisoo-Smith, E. A. (2014). No evidence against Polynesian dispersal
of chickens to pre-Columbian South America. Proceedings of the National
Academy of Sciences of the United States of America, 111, E3583.
Storey, A. A., Ramírez, J. M., Quiroz, D., Burley, D. V., Addison, D. J., Walter, R., . . .
Matisoo-Smith, E. A. (2007). Radiocarbon and DNA evidence for a preColumbian introduction of Polynesian chickens to Chile. Proceedings of the
National Academy of Sciences, 104, 10335-10339.
Storey, A. A., Spriggs, M., Bedford, S., Hawkins, S. C., Robins, J. H., Huynen, L., &
Matisoo-Smith, E. (2010). Mitochondrial DNA from 3000-year old chickens at
the Teouma site, Vanuatu. Journal of Archaeological Science, 37(10), 24592468. doi: 10.1016/j.jas.2010.05.006.
Summerhayes, G. R., Leavesley, M., Fairbairn, A., Mandui, H., Field, J., Ford, A., &
Fullagar, R. (2010). Human adaptation and plant use in highland New Guinea
49,000 to 44,000 years ago. Science (New York, N.Y.), 330(6000), 78. doi:
10.1126/science.1193130.
Sutton, D. G. (1980). A culture history of the Chatham Islands. The Journal of the
Polynesian Society, 89(1), 67-93.
Tajima, F. (1983). Evolutionary relationship of DNA sequences in finite populations.
Genetics, 105(2), 437.
Thomson, V. A., Lebrasseur, O., Austin, J. J., Hunt, T. L., Burney, D. A., Denham, T., . . .
Cooper, A. (2014). Using ancient DNA to study the origins and dispersal of
ancestral Polynesian chickens across the Pacific. Proceedings of the National
Academy of Sciences of the United States of America, 111(13), 4826. doi:
10.1073/pnas.1320412111.
103
Thorvaldsdóttir, H., Robinson, J. T., & Mesirov, J. P. (2013). Integrative Genomics
Viewer (IGV): high-performance genomics data visualization and exploration.
Briefings in Bioinformatics, 14(2), 178-192.
Trudgill, P. (2004). Linguistic and social typology: The Austronesian migrations and
phoneme
inventories.
Linguistic
Typology,
8(3),
305-320.
doi:
10.1515/lity.2004.8.3.305.
Tsangaras, K., Wales, N., Sicheritz-Pontén, T., Rasmussen, S., Michaux, J., Ishida, Y., . . .
Greenwood, A. D. (2014). Hybridization Capture Using Short PCR Products
Enriches Small Genomes by Capturing Flanking Sequences (CapFlank). PLoS
One, 9(10), e109101.
Van Tilburg, J. A. (1994). Easter Island: archaeology, ecology and culture: British
Museum Press.
Walter, R. (1998). Anai'o: the archaeology of a fourteenth century Polynesian
community in the Cook Islands. Auckland [N.Z.]: New Zealand Archaeological
Association.
Weisler, M. I. (1994). The settlement of marginal Polynesia: New evidence from
Henderson Island. Journal of Field Archaeology, 21(1), 83-102.
Weisler, M. I. (1998). Hard evidence for prehistoric interaction in Polynesia. Current
Anthropology, 39(4), 521-532.
Whistler, W. A. (2009). Plants of the canoe people: an ethnobotanical voyage through
Polynesia
Wickler, S., & Spriggs, M. (1988). Pleistocene human occupation of the Solomon
Islands, Melanesia. Antiquity, 62(237), 703.
Willerslev, E., & Cooper, A. (2005). Ancient DNA. Proceedings of the Royal Society of
London B: Biological Sciences, 272(1558), 3-16.
Wilmshurst, J., & Higham, T. F. G. (2004). Using rat-gnawed seeds to independently
date the arrival of Pacific rats and humans in New Zealand. The Holocene,
14(6), 801-806. doi: 10.1191/0959683604hl760ft.
Wilmshurst, J., Hunt, T., Lipo, C., & Anderson, A. (2011). High-precision radiocarbon
dating shows recent and rapid initial human colonization of East Polynesia.
Proceedings of the National Academy of Sciences of the United States of
America, 108(5), 1815-1820.
104
Wilmshurst, J. M., Anderson, A. J., Higham, T. F. G., & Worthy, T. H. (2008). Dating the
late prehistoric dispersal of Polynesians to New Zealand using the commensal
Pacific rat. Proceedings of the National Academy of Sciences of the United
States of America, 105(22), 7676-7680. doi: 10.1073/pnas.0801507105.
Worthy, T. H., & Bollt, R. (2011). Prehistoric Birds and Bats from the Atiahara Site,
Tubuai, Austral Islands, East Polynesia. Pacific Science, 65(1), 69-85.
Yang, D. Y., & Watt, K. (2005). Contamination controls when preparing archaeological
remains for ancient DNA analysis. Journal of Archaeological Science, 32(3), 331336.
105