Active collaborations between SCYNEXIS and DNDi, TDR, MMV and the TB Alliance have identified both opportunities and challenges in building and deploying a cheminformatics toolkit for knowledge discovery in neglected diseases databases. We describe our use of open-source tools in the development of data visualization, data analysis and data mining workflows. Applications range from straightforward data summaries (e.g., molecular weight ranges, number of H-bond donors, log P ranges, etc.) to pairwise comparisons of multiple databases to the construction of predictive machine learning models. In particular, we describe an approach to continuous virtual screening. This workflow makes use of new data, in real time, to construct, validate and deploy models automatically. The toolkit has the ability to build both categorization and regression models, as the data dictate. PMML is used as a means of communicating these models. We show use cases for these workflows with open-source data for anti-malarial screening collections.

Knowledge discovery in neglected disease databases: Open-source cheminformatics tools for data visualization, analysis and mining

243rd National Meeting (2012)

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