Authors
Darius Lingė, Daumantas Matulis
Published in
Biophysical reports. Pages 100296. Oct 27, 2026. Epub Oct 27, 2026.
Abstract
Protein-small molecule binding thermodynamics and structural recognition is a significant challenge in molecular biophysics and difficult to model computationally. To apply AI-ML approaches for rational drug design of small molecules, a sufficiently large database is required that contains reliable and well-organized machine-readable experimental raw data. Current databases contain data on highly diverse proteins and small molecules, but even small chemical changes can cause large differences in binding affinities, so-called Activity Cliff. Therefore, we have assembled a database where the proteins, namely, carbonic anhydrases (CA), are similar and the compounds are primary sulfonamides that are structurally closely related. Despite high similarity of compound chemical structures, the affinities span from undetectable to femtomolar. The database currently contains data on 12 catalytically active human CA isozymes, approximately 800 compounds, and 10,000 interactions by four techniques: fluorescent thermal shift assay, isothermal titration calorimetry, surface plasmon resonance, and inhibition of enzymatic activity. The thermodynamic data are complemented with 138 X-ray crystal structures of the same ligands bound to 6 CA isozymes. Here we describe the functionalities of tools for web users of the Protein-Ligand Binding Database for the visualization and analysis of data, available through a collection of web applications presented at https://tools.plbd.org.
PMID:
42946170
Bibliographic data and abstract were imported from PubMed on 28 Oct 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 7
- Comments 0