PCA-MutPred: Prediction of Binding Free Energy Change Upon Missense Mutation in Protein-carbohydrate Complexes

[Display omitted] •Related binding free energy change upon mutation (ΔΔG) with sequence and structure-based features using multiple linear regression techniques.•Accessible surface area, hydrophobicity, secondary structures, mutation preference, contact energies are important to understand the binding affinity change upon mutation.•Developed a novel sequence and structure-based machine learning method for predicting the binding free energy change upon mutation in protein-carbohydrate complexes.•PCA-MutPred could be a useful resource for designing protein-carbohydrate complexes with desired affinities and to relate with disease causing mutations. Protein-carbohydrate interactions play an important role in several biological processes. The mutation of amino acid residues in carbohydrate-binding proteins may alter the binding affinity, affect the functions and lead to diseases. Elucidating the factors influencing the binding affinity change (ΔΔG) of protein-carbohydrate complexes upon mutation is a challenging task. In this work, we have collected the experimental data for the binding affinity change of 318 unique mutants and related with sequence and structural features of amino acid residues at the mutant sites. We found that accessible surface area, secondary structure, mutation preference, conservation score, hydrophobicity and contact energies are important to understand the binding affinity change upon mutation. We have developed multiple regression equations for predicting the binding affinity change upon mutation and our method showed an average correlation of 0.74 and a mean absolute error of 0.70 kcal/mol between experimental and predicted ΔΔG on a 10-fold cross-validation. Further, we have validated our method using an independent test data set of 124 (62 unique) mutations, which showed a correlation and MAE of 0.79 and 0.56 kcal/mol, respectively. We have developed a web server PCA-MutPred, Protein-CArbohydrate complex Mutation affinity Predictor, for predicting the change in binding affinity of protein–carbohydrate complexes and it is freely accessible at https://web.iitm.ac.in/bioinfo2/pcamutpred. We suggest that the method could be a useful resource for designing protein-carbohydrate complexes with desired affinities.