Predicting Drug Properties from Transcriptomic Profiles and Docking Simulations using Machine Learning
Cancer therapy development demands precise, data-rich strategies to identify effective and safe compounds. This project integrates transcriptomic profiles from drug-treated cancer cell lines with computational docking and machine learning to better predict compound activity and safety. By combining RNA-seq data with large-scale molecular docking simulations, the goal is to uncover how small molecules affect gene expression and to build predictive models for drug efficacy, targets, and ADMET properties. The work aims to establish a unified computational pipeline to prioritize promising candidates in oncology research more efficiently.