Jeffrey Skolnick

The paper investigates if structural mechanisms of protein pocket engagement can predict drug-drug interactions without clinical data, finding high precision in identifying known interactions.

The paper maps human metabolite-protein interactions using LIGMAP and shows improved pathway coverage and prediction when combined with STRING and BioLiP2.

The paper analyzes protein-ligand interactions to show that ancient metabolites form a globally connected network across protein families, while non-ancient metabolites expand this network without disrupting its connectivity.

AlphaInterp analyzes AlphaFold 3's internal mechanisms to show how evolutionary data influences protein structure prediction, revealing dependencies on comparative context and phylogenetic diversity.

This study compares addictive and non-addictive drugs by analyzing their molecular targets and pathways in brain circuits, revealing distinct engagement patterns linked to addiction propensity.

The paper integrates three computational methods to predict E. coli protein-protein interactions and functional networks, identifying over 20K high-confidence interactions.

PHENOCAUZ links symptoms to their molecular causes using Mendelian disease biology, predicting symptom-causing proteins and enabling drug safety and therapeutic target discovery.

GlueFinder identifies molecular glue binding sites near protein interfaces, enabling rational design of degraders that recruit diverse E3 ligases to therapeutic targets.

The paper evaluates AlphaFold and ESMFold by introducing adversarial mutations and finds structural invariance and unreliable confidence metrics, suggesting reliance on templates rather than biophysical principles.

MENDELSEEK predicts Mendelian genes using machine learning, achieving high accuracy and identifying novel candidates with precision over 0.7.

AF3Complex improves structural predictions of protein complexes by building on AlphaFold 3 with new modifications, outperforming AlphaFold 3 and achieving top results in CASP16.

The paper uses deep learning to predict protein-protein interactions in E. coli's outer membrane biogenesis pathway, revealing new insights into protein complex structures and functions.

CDPred predicts inter-chain residue-residue distances in protein complexes using 2D attention-based deep learning, outperforming existing methods on homodimer and heterodimer datasets.

The paper adapts AlphaFold2 models to predict structures of multimeric protein complexes and introduces metrics for predicting direct protein-protein interactions.

This paper introduces SAdLSA, a deep-learning algorithm for protein sequence comparison that generates fold distograms and detects structural relationships without explicit fold training.