Ruibin Liu

Illuminating the Ligandable Proteome with AI Protein Profiling

This paper introduces AiPP, an AI-based platform for predicting ligandable sites in the human proteome, and demonstrates its application in discovering covalent inhibitors for PTPN6.

KaMLs for Predicting Protein pKa Values and Ionization States: Are Trees All You Need?

This paper develops KaML models using decision trees and graph attention networks to predict protein pKa values and ionization states, outperforming existing methods and introducing a new pKa database.

Force Field Limitations of All-Atom Continuous Constant pH Molecular Dynamics

This study evaluates force field limitations in constant pH molecular dynamics simulations of a mini-protein, showing significant pKa errors due to force field and water model choices.

Machine Learned Classification of Ligand Intrinsic Activities at Human mu-Opioid Receptor

The paper develops machine learning models to classify intrinsic activities of ligands at the human mu-opioid receptor using molecular descriptors and SMILES strings, achieving high accuracy even with limited data.

Machine Learning Models to Interrogate Proteomewide Covalent Ligandabilities Directed at Cysteines

The paper develops ML models to predict covalent ligandable cysteines using a curated database, achieving high accuracy and providing a web tool for drug discovery.