A5033190613- Protein Structure and Dynamics
- Enzyme Structure and Function
- Machine Learning in Bioinformatics
- Computational Drug Discovery Methods
- RNA and protein synthesis mechanisms
- Fungal and yeast genetics research
University of Basel
2022-2024
SIB Swiss Institute of Bioinformatics
2022-2024
Abstract Motivation Language models are routinely used for text classification and generative tasks. Recently, the same architectures were applied to protein sequences, unlocking powerful new approaches in bioinformatics field. Protein language (pLMs) generate high-dimensional embeddings on a per-residue level encode “semantic meaning” of each individual amino acid context full sequence. These representations have been as starting point downstream learning tasks and, more recently,...
Abstract Reliably scoring and ranking candidate models of protein complexes assigning their oligomeric state from the structure crystal lattice represent outstanding challenges. A community‐wide effort was launched to tackle these The latest resources on interfaces were exploited derive a benchmark dataset consisting 1677 homodimer structures, including balanced mix physiological non‐physiological complexes. in selected bury similar or larger interface area than counterparts, making it more...
Abstract Methods to systematically monitor protein complex dynamics are needed. We introduce serial ultrafiltration combined with limited proteolysis-coupled mass spectrometry (FLiP–MS), a structural proteomics workflow that generates library of peptide markers specific changes in PPIs by probing differences protease susceptibility between complex-bound and monomeric forms proteins. The includes mapping protein-binding interfaces reporting on accompany PPI changes. Integrating the marker...
Abstract Language models are now routinely used for text classification and generative tasks. Recently, the same architectures were applied to protein sequences, unlocking powerful tools in bioinformatics field. Protein language (pLMs) generate high dimensional embeddings on a per-residue level encode “semantic meaning” of each individual amino acid context full sequence. Multiple works use these representations as starting point downstream learning tasks and, more recently, identifying...
Reliably scoring and ranking candidate models of protein complexes assigning their oligomeric state from the structure crystal lattice represent outstanding challenges. A community-wide effort was launched to tackle these The latest resources on interfaces were exploited derive a benchmark dataset consisting 1677 homodimer structures, including balanced mix physiological non-physiological complexes. in selected bury similar or larger interface area than counterparts, making it more difficult...