John Karanicolas

The paper identifies small molecules that stabilize and restore function to missense-mutated VHL proteins in kidney cancer cells.

This study shows that p53 variants with R181 mutations retain mitochondrial apoptosis function but lose DNA binding cooperativity, leading to reduced cancer risk.

vScreenML v2.0 improves virtual screening by reducing false positives through a streamlined Python implementation, outperforming existing tools in hit discovery.

PPIscreenML is a classification model that distinguishes interacting protein pairs from decoys using AlphaFold2 structures, outperforming existing methods in identifying ligand-receptor interactions.

This paper designs synthetic peptide substrates for TAM family kinases using phosphoproteomics and AlphaFold2, showing that predicted interactions correlate with biochemical validation.

This paper shows that current deep learning models for predicting kinase inhibitor binding affinities do not generalize well to unseen data, suggesting they rely on data leakage rather than learning molecular interactions.

The paper introduces a machine learning classifier for virtual screening that reduces false positives by using a novel training dataset and shows promising results in identifying active compounds.