Alberto Cazzaniga

This paper introduces SCOUT, a synthetic whole-genome sequencing resource for benchmarking cancer genomic analysis, revealing performance issues in complex tumours and highlighting the importance of evolutionary ground truth.

DPCstruct classifies AlphaFold2-predicted protein structures into domain clusters, recovering known folds and identifying novel structural motifs.

This paper improves AlphaFold2's ability to predict alternative protein conformations by integrating clustering strategies and evolutionary signals, enabling rational mutation design.

The paper improves prediction of protein stability changes from single mutations using MSA-based language models and a curated dataset, achieving better performance than existing methods.

This paper annotates protein families in the Unified Human Gastrointestinal Proteome using DPCfam clustering, improving family coverage over Pfam.

The paper introduces adversarial perturbations to protein sequences to disrupt 3D structure predictions, highlighting inconsistencies between mutation biological impact and model responses.

The paper analyzes the geometric properties of hidden representations in protein language models and identifies three distinct phases that reveal a general strategy for reconstructing missing data.