Hugo Schweke

The paper addresses volumetric cell segmentation challenges due to limited labelled data by using synthetic 3D objects to generate unlimited training data, achieving accurate segmentations of real biological data.

Foldseek-Multimer rapidly aligns protein complexes by clustering superposition vectors, outperforming the gold standard in speed while maintaining alignment quality.

The paper predicts homo-oligomerization across four proteomes using AlphaFold2, finding that 50% of archaea, 45% of bacteria, and 20% of eukaryotes form homomers, with disease mutations enriched at interfaces.

The paper updates a database of coiled-coil structures from experimental data and AlphaFold2 models, enabling searches and analysis of coiled-coil assemblies and their sequence-structure relationships.

SURFMAP maps protein surface features into 2D projections for visual and quantitative analysis.

The paper develops a theoretical framework to analyze how protein interaction energy landscapes are shaped by both functional and non-functional partners through cross-docking simulations.