The schools will be rotated between participating Institutes and each School will be organized taking into account specific partner research expertise. Each school will be one week duration (apart from Winter School 2, which will be two weeks) and will include renowned specialists from associated partners, outside of the network as well as from industry and authorities. All Summer Schools will allocate 2 days to practical training to the use of the software tools while Winter Schools will have 2 days training to complementary skills, including ½ day course on programming in Perl (scripting language that is very important in daily practical work). Number of participants for each Winter School will be 30 people including 5 external speakers (including the associated partners).
The summer schools will be open to external participants, who will be required to pay for their living costs, travel and a small fee of ca. 100€ to cover administrative costs with organization of the school for the external participants. The number of participants for summer school will be limited to 100 people including 5 external speakers. Below is a detailed description of the schools.
- Summer school 1 - (HMGU, Germany)
Experimental methods to detect environmental pollutants - Theory
-
Sources, Occurrence and fate of chemicals in the environment
Persistent organic chemicals
Modeling fate, toxicity, and risk
Data reporting according to the OECD principles (e.g., use of Harmonized Templates) - Practical training
-
In vitro testing with protozoa
In vitro testing with mammalian cell lines
Ultra-trace analysis of persistent organic pollutants for environment and health
Installation and use of IUCLID database (official database for REACH) for data reporting
Publishing data at the project web site
Visit to the most powerful ultrahigh resolution FT-ICR MS in Europe - Complementary training
-
Clear writing seminar & workshop on the structure and efficiency of well-written English.
Backgrounds and hands-on training with focus on scientific publications. - External personnel
-
Associated partners
- Summer school 2 - (RU, The Netherlands)
PhD-course on Chemical Risk Assessment & REACH legislation in work: first lessons - Background
-
General introduction on REACH, history, objectives, data requirements, role of industry, European Chemicals Agency
Legislative aspects of risk assessment within REACH
Role of competent authorities, data management, coordination aspects. - Practical training
-
Integrated testing strategies (general background)
Design of integrated testing strategies incorporating existing data, models, in vitro and in vivo bioassays
Read across methods
Weight of evidence reasoning. - Complementary training
-
- Training in the backgrounds to the EU legislation with a focus on environmental ‘laws’.
- Negotiation techniques: how to optimize discussion and negotiations techniques as essential tools for representatives of industry and national competent authorities in the final stages of chemical risk assessment within REACH.
- PhD Course
-
Chemical Risk Assessment (40 hours), carried out in the framework of the Dutch Research School SENSE and covering major academic groups in the Netherlands will be offered.
- External personnel
-
Associated partners, PhD course lectors from RU
- Summer School 3 - (UMB, Italy)
Model Validation and Estimation of QSAR models - Theory
-
This school will focus on development and validation of models using both classical and new approaches such as Multiple Linear Regression, Genetic Algorithms, variable selection techniques, ranking methods, consensus analysis, definitions and applications of applicability domain approaches. A part will also be dedicated to the molecular descriptors.
- Practical training
-
AMBIT software training
Use of Dragon, Caiman, DART for multicriteria decision making
Ranking approaches
OECD QSAR-toolbox. - External personnel
-
Associated partners
OECD QSAR toolbox 2 developers (tender winner will be known in autumn) - Winter School 1 - (HSF, Germany)
Development and Application of Multivariate Methods - Theory
-
Focus on development and validation of methods using PLS, linear regression, singular value decomposition
Estimation of data quality
Experimental design
Understanding of the applicability domain
Accuracy of estimations in linear methods (leverage, Wiliams plots)
Chemical categories formation
Read-across
Clustering and unsupervised methods
Development of models with respect to the OECD principles - Practical training
-
Development of models using commercial and freely available open-source software in basic and advanced multivariate statistical analysis
Advanced use of IUCLID database, (official database for REACH)
Clustering approaches & experimental design. - External personnel
-
Associated partners
JRC members and developers of IUCLID database will be invited - Winter school 2 - (INIA, Spain)
In vitro and in silico Toxicology: Biological activity and computational rationalization. - Theory
-
Optimization of structures using quantum-mechanical methods
Conformational analysis
Ab initio computation of properties
Advantages and limitations of in vitro methods for toxicology studies
Relationship between in vitro data and computed properties - Practical training
-
Software to search structures in database
To visualize structures, to perform molecular mechanics modeling and to realize quantum-chemical calculations
Introduction to cell culture methods
Cell toxicity
Measurement of expression of detoxification enzymes. - Complementary training
-
Training in patents and IPR rights (provided by Ascenion GmbH).
- External personnel
-
Associated partners
Dr. Gerrit Schüürmann, from UFZ (Umweltforschungszentrum) in Germany. - Winter School 3 - (LnU, Sweden)
Development and Interpretation of Non-linear models - Theory
-
REACH requirement to model description and validation according to the OECD principles
Modern methods of data analysis using neural network, random forest, support vector machines, Gaussian processes, consensus modeling
The problems of overfitting of non-linear methods
Applicability domain of ensemble models
Data integration. - Practical training:
-
Use of non-linear models in open-source WEKA and R-language
Use of tools developed within the CADASTER/GoBio
Data analysis automatization using workflows - External personnel
-
Lectures of associated partners
Authors of WEKA and R-software will be invited.
During summer schools the training will be performed on notebooks of the participants who will be required to download and to complete exercises before the school (to be controlled by the supervisors). The analysed examples will be based on the real-life problems from the ECO. The materials of the course will be made freely-available on the Web site of the project. It should be mentioned, that Dr. Tetko has considerable experience in similar practical courses within Strasbourg Summer School on Chemoinformatics, school on mathematics http://summerschool.ssa.org.ua and also that he will co-organize Summer School on Chemoinformatics in Kiev in 2009 (sponsored by Germany-Ukraine bilateral grant UKR 08/006). Moreover, the Chemometrics and QSAR Research Group of Prof. Todeschini since several years freely distributes chemometric software tools (MATLAB) and tutorial on molecular descriptors downloadable from the http://michem.disat.unimib.it/chm/ and http://www.moleculardescriptors.eu. Prof. Herradón was also organizer and demonstrator of practical courses (both experimental and computational) at the Centro de Química Orgánica “Lora Tamayo” in the Science Weeks of 2006 and 2007. The expertise of these researches will be essential to organize the training courses during the schools.