Haim J. Wolfson

LocAlign: Local Protein Structural Alignment with Geometric Deep Learning

LocAlign uses geometric deep learning to align protein structures and identify functional motifs, achieving high accuracy for both similar and dissimilar folds.

SegmA: Residue Segmentation of cryo-EM density maps

SegmA is a deep learning model that segments cryo-EM density maps into residue types, achieving 80% accuracy for nucleotides and improved amino acid detection after removing low-confidence regions.

ScanNet: An interpretable geometric deep learning model for structure-based protein binding site prediction

ScanNet predicts protein binding sites using an interpretable geometric deep learning model, achieving accuracy on unseen protein folds and identifying epitopes in SARS-CoV-2 spike protein.