Is it worth crystallizing phosphatases in the age of AI-based structural modeling?
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Recent advances in deep learning have dramatically expanded the scope of protein structure prediction by enabling accurate modeling not only of individual proteins but also of their interactions with ligands, nucleic acids, and small molecules.This development raises a fundamental question in structural biology: is experimental crystallization still necessary, particularly for functionally conserved enzyme families such as phosphatases?In this work, we model a representative set of phosphatases using AlphaFold 3 and analyze the structural features in detail for one selected enzyme.We compare the AI-generated model with its experimentally determined crystal structure to examine differences in side-chain positioning, and active site conformation.
Fields of Study & Subjects
Protein Structure and Dynamics
Enzyme Structure and Function
Machine Learning in Bioinformatics
Publication details
- Journal / source
- Acta Crystallographica Section A Foundations and Advances
- Volume
- 81
- Issue
- a1
- Pages
- e276–e276
- Publication date
- 30 March 2026
- Language
- EN