The ORCHESTRA e-book on in silico methods
With the specific aim of improving the dissemination of in silico methods use in toxicology and of knowledge and outcomes produced so far by EU funded projects, in 2009 the European Commision financed a three-year project: ORCHESTRA.
One of the most important outcomes produced by ORCHESTRA is the e-book "Theory, guidance and applications on QSAR and REACH". As stated in the book's foreword:
“With this e-book you will learn about the state of the QSAR art. This book is composed of three parts: The first will address the scientific aspect of modelling. Then we will examine some practical cases. Finally, we will discuss the possible applications of QSAR with particular attention to the REACH legislation.”
The ORCHESTRA e-book main targets are the chemical industry and regulators; with this in mind, it has been organized as follow:
Part A. Theory
1) Istituto di Ricerche Farmacologiche Mario Negri, Via La Masa 19, 20156, Milano, Italy
2) Politecnico di Milano, Piazza L. da Vinci 32, 20133, Milano, Italy
3) Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129, Torino, Italy
The book starts with a theoretical introduction to in silico methods, focusing in particular on (Q)SARs. It gives an overview on how the models are built (e.g. Figure 2), from the choice of the dataset used to build the model and some examples of common modelling algorithms, to the processes of validation of these models. This first part also introduces some advanced modelling approaches such as hybrid models. A brief comparison of in silico approaches versus other approaches is also given.
Part B. Practical cases
1) Istituto di Ricerche Farmacologiche Mario Negri, Via La Masa 19, 20156, Milano, Italy
2) Università degli Studi di Bari “Aldo Moro”, Piazza Umberto I 1, 70121, Bari, Italy
In this section, the five model developed by the CAESAR project are taken as examples, explaining deeply how they have been built, including the decision taken about the dataset to use, how it has been split between training set (molecules used to train the model) and test set (molecules used for the model validation). The performances of the models on training and test set are also reported (e.g. Figure 3).
Apart from the model itself, the endpoint and its relevance within the REACH legislation are reported as well as the application of (Q)SAR models for this particular endpoint within the REACH framework.
Part C. The applications and the possible uses of QSAR models
1) Istituto di Ricerche Farmacologiche Mario Negri, Via La Masa 19, 20156, Milano, Italy
2) Politecnico di Milano, Piazza L. da Vinci 32, 20133, Milano, Italy
3) Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129, Torino, Italy
This section of the e-book explores the possible uses of QSAR for regulatory purposes and has been designed especially for people working in the chemical industry. It gives an overview of the use of in silico methods in REACH, focussing on the requirements under this legislation (e.g. Figure 4).
More information and insights are given in the seven ennexes included:
- Annex 1. QSAR and SAR models: Basic definitions
- Annex 2. An introduction to toxicology
- Annex 3. Rigid and flexible topological indices: variety of the representations of the molecular structure
- Annex 4. A free and open source informatics library for chemistry: Chemistry Development Kit (CDK)
- Annex 5. REACH
- Annex 6. Genetic Algorithms in QSAR for REACH
- Annex 7. The CORAL software: principles, results, perspectives
