MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641108140 A) filed by Pragati Engineering College on September 09, 2026, for Machine Learning Driven Platform For Virtual Screening And Drug Repurposing.
Inventors include Dr. Avinash Gudimetla; Mr. A. Phani Bhaskar; Mr. Bomma Bharath Kumar; and Mrs. Kada Tulasi.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: The present invention pertains to a computer-implemented machine learning platform for virtual screening and drug repurposing by leveraging artificial intelligence, bioinformatics, cheminformatics, and structure-based drug discovery in order to facilitate the identification of potential compounds. Traditional drug discovery is a costly, time-consuming, and resource-intensive procedure which has a high failure rate during preclinical and clinical phases. Drug repurposing is a fast-track alternative for discovering novel uses for existing drugs; nevertheless, computational techniques employed for drug repurposing are mostly isolated prediction approaches. The present invention overcomes the mentioned shortcomings through its computer-implemented automatic computational system using machine learning algorithms in combination with virtual screening and molecular modeling techniques for enhancing prediction and decreasing the lead identification time. The platform collects data about chemical compounds and their biological targets from public and/or proprietary databases, and then processes the data for molecular preprocessing, descriptor generation, and feature engineering for building up the predictive datasets. Machine learning models are used to predict interactions between the compound and the target and prioritize candidate molecules according to their biological activities. The identified compounds are later evaluated through high throughput virtual screening, molecular docking, pharmacokinetic and toxicity analysis, molecular dynamics simulations, and binding free energy estimation to determine their therapeutic potential. Scoring based on an integration of the predictive ability of the compound through machine learning, docking activity, structural integrity, and ADMET analysis is used to prioritize the compounds on the basis of efficacy and safety. The analytical reports which include prioritized drug candidates, interaction profile, pharmacokinetics features, toxicity predictions, molecular dynamics parameters, and binding free energies in order to aid drug discovery process. The invention provides an efficient, robust, and cost-effective platform which reduces reliance on laborious and costly laboratory experiments in the drug discovery process.
Disclaimer: Curated by HT Syndication.
