MegaMiner: A Tool for Lead Identification Through Text Mining Using Chemoinformatics Tools and Cloud Computing...

Area of Science:

• Computational chemistry
• Bioinformatics
• Drug discovery

Background:

• High-throughput omics technologies generate vast datasets, necessitating efficient virtual screening methods.
• Automating virtual screening is crucial for managing and interpreting complex biological data.
• Textual chemical data representations often lack standardization, leading to ambiguity.

Purpose of the Study:

• To develop an automated virtual screening portal, MegaMiner, to streamline lead molecule identification.
• To address ambiguities in textual chemical data representation using a novel dictionary-building approach.
• To demonstrate MegaMiner's effectiveness through a case study on antimalarial drug discovery.

Main Methods:

• Cloud computing platform utilized for MegaMiner portal development.
• Combination of statistical models, chemical dictionary, and regular expressions for disease-specific dictionary construction.
• F-score analysis and molecular docking used for validation.

Main Results:

• MegaMiner successfully retrieved relevant PubMed records, protein classes, and drug classes for 'malaria'.
• The system identified 8000 scaffolds, processed and filtered to suggest potential antimalarial molecules.
• MegaMiner demonstrated superior performance compared to existing text mining search engines.
• Docking studies validated the proposed virtual molecules against relevant protein targets.

Conclusions:

• MegaMiner offers an automated and robust solution for virtual screening and lead molecule discovery.
• The developed text mining approach effectively handles ambiguous chemical data for disease-specific applications.
• MegaMiner serves as a valuable tool for uncovering hidden relationships and building biological/chemical ontologies.