In vitro pharmacology data contains vital information on the probability of binding to a receptor and species selectivity issues, which can have a considerable impact on drug hunting project progress. Data from the literature and public databases are variable due to factors like assay conditions and diversity of structural types. Both structural alerts and potential chemical modifications are extracted from the data using Matched Molecular Pair (MMP) and fragment frequency analysis. The volume and variety of data requires the development of specific methods to deliver knowledge that can be used to make decisions in a relevant timeframe. This process can be automated to yield a robust database of knowledge. Results from the analysis of all viable datasets from ChEMBL, including more than 10 high priority drug safety assays will be discussed. This knowledge has the potential benefit of reducing avoidable late stage safety failures and the methodology equally can aid the generation potent compounds by potency prediction and chemical space exploration.

Automated extraction of pharmacophores and toxophores from large-scale <i>in vitro</i> pharmacology data

253rd National Meeting (2017)

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

Automated extraction of pharmacophores and toxophores from large-scale in vitro pharmacology data