Ivan V. Anishchenko

An atom-level graph neural network that denoises corrupted structures generates small-molecule and protein-ligand conformational ensembles, and its active-site pre-organization metric enriched for more active designed retroaldolases.

This paper introduces LigandMPNN, a deep learning method for protein sequence design that models non-protein atomic contexts, achieving better performance than existing methods in recovering interacting residues.

This study designs novel serine hydrolases with high catalytic efficiency using computational methods, achieving structures close to design models and distinct from natural enzymes.

This study uses deep learning to predict protein-protein interactions in bacterial pathogens, identifying new complexes involving essential genes and virulence factors, with half of tested predictions validated.

The paper designs nanomolar binders for diverse small molecules using a pseudocycle-based shape complementarity method, enabling direct conversion into sensing systems.

One RoseTTAFold-derived network models protein-DNA and protein-RNA complexes plus RNA tertiary structure end to end, returning confidence estimates alongside each 3D model.

The paper predicts structures of disease-associated proteins without known structures, identifies functional sites, and explains 80% of mutations based on their proximity to these sites or structural impact.

The paper introduces a 3-track network for accurate protein structure and interaction prediction, achieving performance close to DeepMind's CASP14 results and enabling rapid modeling of protein complexes from sequence data.

The paper introduces DeepAccNet, a deep learning framework that improves protein structure refinement by estimating accuracy and guiding Rosetta refinement.

The paper predicts SARS-CoV-2 B-cell epitopes using computational methods and validates them with ELISA, finding two antigens with high sensitivity and specificity for serological diagnosis of COVID-19.

Graph convolutions on atom graphs with distance, orientation and bond edge features predict per-residue and global lDDT of protein models, outperforming VoroMQA and retrained Ornate while using about 100k parameters.

This paper introduces a deep residual network and Rosetta protocol for predicting inter-residue orientations and distances, outperforming existing methods in benchmark tests.