Rohan Gorantla

This study evaluates if language models improve combinatorial molecular search for ionizable lipids compared to classical methods under budget constraints, finding no significant improvement.

GRASP is a graph Transformer that learns molecular structure from ZINC20 data and adapts to ChEMBL bioactivity, achieving better performance than existing models on ADMET prediction tasks.

AbAffinity predicts antibody-antigen affinity using a chain-aware sequence model, achieving high correlation on benchmark datasets and maintaining performance without structural data.

The paper shows targeted finetuning can enable co-folding models to learn ligand-induced conformational states not present in training data.

BALM-PPI predicts protein-protein binding affinity from sequence using a fine-tuned protein language model, achieving accuracy comparable to structure-based methods and providing residue-level explanations.

This paper develops an explainable active learning framework for ligand binding affinity prediction, showing that method choices affect outcomes but dataset characteristics set performance limits.

This paper introduces BALM, a deep learning framework for predicting protein-ligand binding affinity using pre-trained language models, showing better performance than traditional methods in few-shot scenarios.

This study benchmarks active learning protocols for predicting ligand binding affinity using machine learning models and evaluates the impact of batch size and noise on performance.

The paper evaluates deep learning methods for predicting protein-ligand binding affinity by analyzing the impact of different encodings for proteins and ligands.