Lorenzo Pantolini

Stoic predicts protein complex stoichiometry using protein language model embeddings and graph neural networks, achieving fast and accurate results for homomeric and heteromeric complexes.

This paper introduces a method for fast de novo protein design using a structure-inspired alphabet and Metropolis sampling to generate sequences without known homologues.

This paper introduces a new 20-letter alphabet, TEA, derived from protein language models, enabling efficient homology searches comparable to structure-based methods without structural data.

This paper introduces an embedding-based alignment method using protein language models to detect structural similarities in the twilight zone, outperforming existing approaches without training.

The paper develops personalized logical models to analyze cancer response to BRAF treatments, showing they correlate with experimental sensitivities using omics data without model training.