Grzegorz M. Popowicz

The paper trains a machine-learning model on native protein structures to predict interaction preferences and energetics without using affinity or mutation labels, achieving performance comparable to existing methods on a benchmark.

This study identifies Sgt1 as an essential co-chaperone that stabilizes Hsp90-client complexes and prevents dissociation by Aha1, enhancing client maturation efficiency.

The paper determines the structure of crTPpart2 from two chromatophore-targeted proteins, revealing a fold similar to γ-glutamyl-cyclotransferase and suggesting its role in protein import.

The paper introduces In-Pocket, a computational method that validates and improves the chemical accuracy of protein-ligand binding interfaces using quantum mechanics and experimental data.

The paper introduces a machine learning model that learns spatial preference maps for drug-receptor interactions to guide drug discovery.

This paper reveals structural and functional differences in PA200 interaction with the immunoproteasome (i20S) compared to the standard proteasome (s20S), showing PA200 enhances i20S activation and regulates its gene expression.

This paper investigates how curcumin interacts with DYRK1A and DYRK2 kinases, showing its potential to treat diabetes and cancer by modulating these proteins.

The paper introduces MISATO, a dataset combining quantum mechanics properties and molecular dynamics simulations of protein-ligand complexes to support structure-based drug discovery.

The paper introduces an algorithm for optimizing protein-ligand complex structures, improving data curation and analysis for drug discovery.

The paper introduces an Energy Decomposition Analysis to understand non-covalent binding in drug-protein complexes, validated against experimental data for PEX14-PEX5 and its inhibitors.

The paper introduces a deep learning model called hotWater to predict water binding sites on protein surfaces using limited-resolution data, showing improved performance over existing methods.