May 2012
1
Beneficiary 1 : Istituto di Ricerche Farmacologiche Mario Negri
Beneficiary 2 : Istituto Superiore di Sanità
Beneficiary 3 : Federazione Nazionale dell’Industria Chimica
Beneficiary 4 : Politecnico di Milano, Dipartimento di Elettronica e
Informazione
Beneficiary 5 :KnowledgeMiner Software Frank Lemke
May 2012
2
Promotion of non testingmethods (NTM) fortheiruse in the
REACH contextlinkingscientists, regulators and industries
toevaluate and validateexisting NTM
fortheirapplicationaccordingto REACH needs
May 2012
3
REACH &INNOVATION
REACHpromotesInnovation(Article 1)
AllInfo should be used
QSAR
May 2012
ismentioned
4
=f( )
May 2012
5
Survey of current methods for the compliance to the REACH
legislation
Identification of the criteria for the non-testing methods for the REACH
legislation
Identification of suitable experimental databases/data sets for the
ecotoxicological, toxicological and environmental endpoints for
REACH
List of (Q)SAR models for the ecotoxicological, toxicological and
environmental endpoints for REACH, and their review
Validation of non-testing methods (incl. read-across)
Identification of boundaries for best use of models (applicability
domain) and of the assessment factors
May 2012
6
Architecture for integration of different non-testing methods for best
performances and coverage of applicability
Communication and dissemination initiatives
Web portal
Project management
Monitoring
Audit
After-LIFE Communication plan
May 2012
7
Endpointsproducedfor
REACH byLaboratories
0
10
20
% Endpoints for REACH
30 40 50 60 70 80
90
100
Biolab S.p.A.
Biotecnologie S.r.l.
Laboratories GLP
Chelab
Chemservice
FAR.CO.S. S.r.l.
Lata
Physical-Chemical %
Toxicological %
Eco-Toxicological %
Renolab S.r.l
RTC
Stazione Sperimentale Combustibili
Istituto Certificazione Qualità
May 2012
8
 Global Predictions on the 1544 compounds (CPDB+Leadscope) of the seven
programs examined
May 2012
9
1.00
BCF – Best Models based on R2 and Accuracy
0.90
Ordered by R2
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
R2
May 2012
10
May 2012
11
IN SELECTED CASES NON-TESTING METHODS CAN BE USED:
BEST MODELS: mutagenicity, BCF;
CRITICAL MODELS: daphnia, carcinogenicity;
Important to consider the applicability domain;
Important the correct evaluation of the results (expert);
USE of more than one model preferable.
May 2012
12