The high-throughput screening (HTS) technologies have been widely used in the current toxicology research. For example, US EPA screened 960 chemicals against hundreds of bioassays in the ToxCast project. However, it is impossible to screen millions of compounds using the same protocol. On the other hand, numerous bioassay data has been deposited to the PubChem databases containing 31 millions compounds. To extract useful information in toxicology studies from public sources (e.g. PubChem), data mining has become a powerful tool to identify the bioentities. In this study, we implement an automated data mining method to explore around 10,000 compounds of environmental interest throughout the PubChem databases for �similar� chemicals. We classify similarity by different searching categories, such as similar structure, active responses in similar bioassays or similar pathway perturbations. This procedure results an environmental database with expanded chemical space and biological information that could be used in further toxicity model development.

Applying novel data mining techniques to create a new chemical database of environmental interest

244th National Meeting (2012)

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