Damiano Piovesan

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.

This paper reports on the fourth CAFA challenge, showing improvements in Gene Ontology term prediction but highlighting challenges in new evaluation settings.

This work analyzes CAFA5 submissions, showing improved accuracy in protein function prediction and highlights the role of protein language models and 3D structures in performance gains.

emb2dis predicts protein disorder using a deep learning model combining ResNets and dilated convolutions, achieving top performance on benchmark datasets.

GeomeTRe calculates geometric descriptors for tandem repeat proteins, improving their classification by analyzing structural symmetry and variations in spatial arrangements.

PredIDR accurately predicts protein intrinsic disorder regions using a deep convolutional neural network, achieving performance comparable to top state-of-the-art methods in the CAID assessment.

STRPsearch detects structured tandem repeat proteins by aligning protein structures to a curated database and mapping repeat units efficiently.

The paper evaluates AlphaFold2's ability to predict intrinsically disordered regions and conditionally folded regions using the CAID dataset, finding it highly competitive.

PDB ProtVista is a sequence feature viewer that provides 2D visualization of residue-level annotations and integrates with 3D molecular viewers.

3D-Beacons provides a federated network to unify access to protein structure data from multiple sources, enabling comparative analysis of models.

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.

APICURON credits biocurators by tracking their work across knowledgebases, using gamification and an API for resource integration.

The paper analyzes experimental evidence for long intrinsically disordered protein regions in the PDB, showing increased coverage of intrinsic disorder over the last decade.

The paper evaluates hydroxylation site predictors using independent datasets, finding they perform no better than random, highlighting issues with self-reported performance and generalization.