Design and development of a high performance molecular Informatics toolkit to handle Big Data automatically for the design of novel anti-cancer drugs utilizing machine learning and artificial intelligent systems is described. This platform CCCTK is capable of handling multi-dimensional data related to cancer for the design of novel anti-cancer drugs utilizing open source based in-silico library design and analysis tools. The commercial tools are not able to meet current requirement of integrated approach with multi-dimensional data (numeric, textual, image, compressed, encrypted, both in structured and un-structured formats, RDBMS etc). there is a need to develop open source based tools that can handle large scale data in a distributed and parallel computing environment. The prototype of compute cure could facilitate compile, curate, and integrate chemical genomics data relevant to cancer and develop virtual screening assay platform for target activity prediction of drug /lead candidates. The system currently linked to data for a class of cancer specific biological targets that are annotated in public repositories such as PubChem, ChEMBL for drug repurposing with respect to cancer treatment. The challenging part in this toolkit is to develop virtual library of bioactive molecules from known library of cancer specific drugs and leads. We established ChemScreener platform using java programming toolkit to generate unique scaffolds and functional groups from a library of molecules that can be used to reconstruct the expanded virtual library of molecules (millions) and screen using descriptor space and bioactivity space. The in-house developed chemical, pharmacophore, and toxicophore descriptors and fingerprints are being used to efficiently filter out the potential candidate molecules that can go for drug discovery pipeline The molecular informatics computing platform with user friendly interface would facilitate Quantitative Structure Activity Relationship (QSAR) models for all identified datasets (selected Cancer cellline bioactivity data and other toxicity data) to integrate these models in a virtual screening platform enabling the prediction of putative target and off-target effects for selected drugs. The current CCCTK prototype is capable of handling all chemoinformatics tasks generation of large collection and screening of virtual library of novel molecules from the collection of known drugs and lead molecules for further studies.

CCCTK: High performance molecular informatics toolkit for the design of anti-cancer molecule

256th National Meeting (2018)

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