As alternative approaches to fulfill data gaps in chemical safety assessment, in silico methods, such as read-across and quantitative structure activity relationship (QSAR), have generated much attention. Such approaches have been used in ECHA (European Chemicals Agency) submissions of chemical dossiers for REACH (Registration, Evaluation, Authorisation and Restriction of Chemicals). Searching and identifying suitable structural analogs for the chemical of interest is prerequisite for read-across. Although several methodologies and programs are available for analog searching, there are limited practical strategies or guidance for conducting analog search. This work compared several popular analog search programs using 3 different fragrance materials (as indicated in Fig. 1) reported in a published paper by Blackburn et.al.1 As shown in Fig. 1, for each material, 8 different analog search methods from 5 different programs were conducted. The top 100 analogs with Tanimoto similarity2 more than 0.5 were compared with the suitable analogs reported in Blackburn et al.'s paper.1 As shown in Fig. 1, Pipeline Pilot with FCFP4 fingerprint outperforms other methods in all 3 cases. Most of the methods have higher enrichment rate for lauryl alcohol, and have lower recovery rate for 2, 6-Dimethyl-4-heptanone. Based on this and our experience, Pipeline Pilot (FCFP4/1024 bits) is primarily used to identify suitable analogs.  Fig. 1.  The enrichment rate of suitable analogs identified for 3 fragrance materials using 8 different search methods from 5 different programs.  Top 100 analogs with a Tanimoto score more than 0.5 were compared with the analogs reported by Blackburn et.al.1 1. Blackburn, K.; Bjerke, D.; Daston, G.; Felter, S.; Mahony, C.; Naciff, J.; Robison, S.; Wu, S., Case studies to test: A framework for using structural, reactivity, metabolic and physicochemical similarity to evaluate the suitability of analogs for SAR-based toxicological assessments. Regulatory Toxicology and Pharmacology 2011,  60 (1), 120-135.2. Rogers, D. J.; Tanimoto, T. T., A Computer Program for Classifying Plants. Science 1960,  132 (3434), 1115-1118.

Benchmark study: Structural similarity search methods for identifying read-across analogs

248th National Meeting (2014)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.