MPLBind: Predicting the Effect of Binding Site Point Mutations on Protein-Ligand Binding Affinity Using Protein Large...

Area of Science:

• Biochemistry
• Computational Biology
• Pharmacology

Background:

• Protein mutations, particularly in binding sites, can alter protein-ligand binding affinity, influencing individual drug responses.
• Accurate prediction of mutation effects on binding affinity is challenging but essential for personalized medicine.

Purpose of the Study:

• To develop and validate a novel computational method, MPLBind, for predicting the impact of mutations on protein-ligand binding affinity.
• To enhance the accuracy of predicting drug response variations attributed to genetic differences.

Main Methods:

• MPLBind integrates ligand descriptors, fingerprints, local mutation environment changes, and large protein language model features.
• Large protein language model features capture contextual, evolutionary, conservation, and functional sequence information.
• A fusion strategy combines ligand and mutation features for improved predictive power.

Main Results:

• MPLBind demonstrates superior performance compared to existing baseline models in predicting protein-ligand binding affinity.
• The method accurately predicts the effect of mutations on protein-ligand binding affinity.
• Integration of large protein language models significantly boosts prediction performance.

Conclusions:

• MPLBind offers a robust and accurate approach for predicting mutation effects on protein-ligand binding affinity.
• The study highlights the utility of large protein language models in computational drug discovery and response prediction.
• MPLBind has the potential to advance the understanding of interindividual differences in drug efficacy.