CSM-lig: a web server for assessing and comparing protein-small molecule affinities

Area of Science:

• Computational chemistry
• Structural biology
• Drug discovery

Background:

• Accurate prediction of protein-ligand binding affinity is essential for drug development.
• Scoring functions are vital for analyzing molecular docking results, despite limitations in predictive power.

Purpose of the Study:

• To develop and present CSM-Lig, a web server for predicting protein-small molecule binding affinities.
• To incorporate both shape and chemical complementarity using graph-based structural signatures.

Main Methods:

• CSM-Lig was trained and evaluated using data from various PDBbind database releases.
• The method utilizes graph-based structural signatures to capture protein and small-molecule complementarity.
• Performance was assessed through 10-fold cross-validation and blind tests.

Main Results:

• CSM-Lig achieved a correlation of up to 0.86 in 10-fold cross-validation.
• Blind tests demonstrated a correlation of 0.80, matching or exceeding other widely used methods.
• The web server enables rapid, automated prediction and assessment of binding affinities and interactions.

Conclusions:

• CSM-Lig is a valuable tool for evaluating molecular docking poses and understanding the impact of mutations on protein-ligand affinity.
• The server aids in unraveling key factors driving protein-compound recognition.
• It offers a robust method for predicting binding affinities in drug discovery and biological studies.