Historical QSAR Documents

Comparison of the COMMPS Priority Setting Scheme with Total and Partial Ranking Algorithms (May 2007)

In support of the Water Framework Directive, the JRC compared the results obtained with Combined Monitoring-based and Modelling-based Priority Setting (COMMPS) scheme with the results obtained with four ranking methods, including three types of total ranking techniques (Desirability, Utility and Dominance function) and a partial ranking method (Hasse Diagram Technique, HDT).

Download report

Evaluation of (Q)SARs for the Prediction of Eye/Skin Irritation/Corrosion Potential (Oct. 2005 and Feb. 2006)

The German Federal Institute for Risk Assessment (BfR) has developed a QSAR decision support system containing several rulebases for predicting the presence or absence of acute local effects to the skin and the eye. The JRC carried out an assessment of the set of physicochemical exclusion rules for predicting the absence of eye irritation and/or corrosion (I. Tsakovska, T. Netzeva, A. Worth; publication: EUR 21897 EN, Feb. 2006).

With sponsorship from the JRC, a report has been produced by the National Institute of Public Health and Environment (RIVM), NL, on the assessment of one of the set of physicochemical exclusion rules for predicting the absence of skin irritation and/or corrosion (Emiel Rorije, Etje Hulzebos, Oct. 2005).

Download report (Eye)
Download report (Skin)

Review of Non-Commercial (Q)SAR Models for Acute Fish Toxicity (Feb. 2006)

The JRC funded a study to review and evaluate non-proprietary models for acute toxicity to fish. The report of the study, written by Drs Brooke and Crookes (Building Research Establishment, Watford, UK) provides detailed information on a wide range of literature-based models.

Download report (part A)
Download report (part B)

Review of Non-Commercial (Q)SAR Models for Estrogen Receptor and Androgen Receptor Binding (Dec. 2005)

The JRC funded a study to review and evaluate non-proprietary models for ER and AR binding activity. The report of the study, written by Dr Benfenati and colleagues (Mario Negri Institute, Milan), provides detailed information on a wide range of literature-based models. A total of 158 models were evaluated.

Download report

A Series of Papers on Applicability Domain Methods (Feb. 2005)

The Joint Research Centre sponsored a series of papers to review methods for assessing the applicability domains of SARs and QSARs. This work was done by Dr Joanna Jaworska (Procter & Gamble, Belgium) in collaboration with Dr Tom Aldenberg (RIVM, NL) and Dr Nina Nikolova (Bulgarian Academy of Sciences, Bulgaria).

Download overview of the four papers
Review of methods for QSAR applicability domain estimation by the training set (Paper 1)
An approach to determining applicability domain for QSAR group contribution models: an analysis of SRC KOWWIN (Paper 2)
Joint applicability domain and predictive uncertainty in QSAR Regression (Paper 3)
Discussion on the difference between SAR and QSAR (Paper 4)

Evaluation of Different Statistical Approaches for the Validation of QSARs (Nov. 2004)

The validation of QSARs has been a subject of intense debate within the academic regulatory and regulated communities. The questions discussed refer to: a) the validation principles that should be followed; b) the methods and approaches that can be used to apply the principles; c) the criteria for establishing scientific validity of models; and d) the flexibility and pragmatism that should be applied in the "real-world" use of QSARs for regulatory purposes (P. Gramatica, University of Insubria, Varese, Italy).

Download paper

Overview of Structure-Activity Relationships for Environmental Endpoints (1995)

The report presents an overview of QSARs for sorption, degradation, and ecotoxicity. All these parameters play an important role in the evaluation of risks of chemicals in the environment.

Part 1: General outline and procedure
Part 2: Descriptions of selected models

Predictions of the Aquatic Toxicity of High-Production-Volume Chemicals (1993)

The predictions are based on: a) a scheme to classify chemicals into four different classes; b) a set of QSAR equations for different endpoints of ecotoxicity; c) extrapolation methods for calculation of safe level at the ecosystem level; d) an equilibrium partitioning model (J. Bol, H.J.M. Verhaar, C.J. Van Leeuwen and J.L.M. Hermens).

Part A: Introduction and Methodology

Part B: Predictions

The US/EC Joint Project on Evaluation of (Q)SARs (1989)

In 1989, the OECD organised a workshop on Notification Schemes for New Chemicals, during which it was recommended that the (Q)SARs in use by the US EPA should be evaluated in terms of their predictive power. It was recommended that this should be achieved by applying (Q)SARs to chemicals for which test data had been obtained in the context of the EU New Chemicals notification scheme, so that QSAR predictions could be compared with independent experimental data. To follow up this recommendation, the US EPA and the European Commission undertook a joint project in which predictions for 144 substances were made by using the US EPA (Q)SARs for a wide variety of physicochemical properties, ecotoxicological effects and human health effects. The study identified promising areas for the wider application of (Q)SARs, such as biodegradation and acute toxicity to fish and Daphnia, as well as areas where further work was needed, such as the development of (Q)SARs for human health effects. The outcome of the US/EC project was published as OECD Environment Monograph no. 88, and original scanned documents on the outcome of the study can be downloaded from the following website:

Link to US EPA website