Joel Lapin

The paper evaluates peptide property predictors using SHAP to explain model predictions and finds strong correlations with known amino acid indices.

This paper introduces Pairwise Attention, a modified transformer encoder for de novo peptide sequencing from MS2 spectra, improving precision by 12.7% on a nine-species benchmark.

The paper develops a deep learning model that integrates multiple fragmentation techniques into standard LC-MS workflows, improving proteome coverage and identification rates.

Prosit-XL improves cross-linked peptide identification by predicting fragment intensities, enhancing protein-protein interaction discovery and interactome mapping.