Jörg A. Gsponer

The paper identifies a bias in IDR-binding site prediction benchmarks due to differences in disorder propensity between datasets, showing predictors trained on one type of data perform poorly on others.

MoRFchibi 2.0 predicts Molecular Recognition Features in protein sequences, outperforming existing tools and models including AlphaFold and protein language models.

This study shows AR-V7 forms condensates independently of androgens and AR-FL in CRPC, driving oncogenic transcription through both condensate-dependent and -independent mechanisms.

The paper analyzes protein levels and solubility in brain regions of aged mice, revealing age-related changes in proteostasis and inflammation markers.

The paper evaluates 43 methods for predicting intrinsically disordered regions and binding regions in proteins, finding that deep learning methods outperform older physicochemical approaches but binding prediction remains challenging.