Jordi Pujols

The paper presents a database of structure-based protein aggregation predictions for the human proteome using AlphaFold structures and the A3D tool.

This paper updates A3D to predict and optimize protein solubility using a computational pipeline and RESTful service for handling large protein collections.

The paper introduces Aggrescan3D, a tool for designing protein variants with improved solubility and reduced aggregation using 3D structures, and demonstrates its application on therapeutic proteins.

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.