Ziv Ben-Aharon

Protein language models can capture protein quaternary state

The paper trains a protein language model to predict protein quaternary states from sequences, showing it can distinguish multimers from monomers but performs worse than structure-based methods.

PatchMAN docking: Modeling peptide-protein interactions in the context of the receptor surface

PatchMAN models peptide-protein interactions by aligning structural motifs on the receptor surface, achieving high accuracy in docking without using sequence information.

Harnessing protein folding neural networks for peptide-protein docking

The paper applies AlphaFold2 to model peptide-protein interactions using a poly glycine linker, showing accurate predictions without sequence alignment.

Ensuring scientific reproducibility in bio-macromolecular modeling via extensive, automated benchmarks

The paper presents a framework for ensuring reproducibility in bio-macromolecular modeling through automated benchmarks and a test server, demonstrating its utility for Rosetta and other tools.