Tapan Chowdhury

The paper develops a graph neural network that combines ligand molecular graphs with protein language model embeddings to predict GPCR ligand activity, showing improved performance over baseline models.

The paper identifies AD-associated miRNA-mRNA regulatory signatures and potential biomarkers using transcriptomic data from ADNI and GSE48552.

This paper develops a Dynamic Bayesian Network model to analyze multi-fault behavior in signaling pathways and prioritize drug interventions based on efficiency and burden scores.

NetPolicy-RL improves drug prioritization by integrating network biology with reinforcement learning to optimize ranking quality in pharmacogenomic screens.