A new method for assessment of molecular similarity based on original description of chemical structure is discussed. The accuracy of similarity assessment obtained with this method is compared with that of the results of four other approaches. The same evaluation set is used to predict: (a) boiling point of 139 hydrocarbons and (b) mutagenicity of 15 nitrosamines. The results show that the proposed method provides reasonable appraisal for both properties, but prediction of mutagenicity is more accurate in this method as compared to the alternatives.
Journal of Chemical Information and Computer Sciences, 1999, 39 (4), 666-670.

The chapter of the monograph describes the following topics: 8.1 Introduction 8.2 Biological Activity 8.2.1 Dose-Effect Relationships 8.2.2 Experimental Data 8.3 Probabilistic Ligand-Based Virtual Screening Methods 8.3.1 Preparation of Training Sets 8.3.2 Creation of Evaluation Sets 8.3.3 Mathematical Approaches 8.3.4 Evaluation of Prediction Accuracy 8.3.5 Single-Targeted vs. Multi-Targeted Virtual Screening 8.4 PASS Approach 8.4.1 Biological Activities Predicted by PASS 8.4.2 Chemical Structure Description in PASS 8.4.3 SAR Base 8.4.4 Algorithm of Activity Spectrum Estimation 8.4.5 Interpretation of Prediction Results 8.4.6 Selection of the Most Prospective Compounds 8.5 Conclusions
In: Chemoinformatics Approaches to Virtual Screening. Eds. Alexandre Varnek and Alexander Tropsha. Cambridge (UK): RSC Publishing, 2008, p.182-216.