Protein-carbohydrate interactions play a significant role in several cellular processes and biological functions of living organisms. The substitution of amino acid residues in protein-carbohydrate complexes alters their binding affinities, affects the functions and some of them may lead to diseases. Understanding the residues, which are important for folding and binding, binding affinity and binding affinity change upon mutation in protein-carbohydrate complexes as well as disease-causing mutations provide deep insights to understand the factors influencing the binding affinity and molecular basis of diseases. We have investigated the dual role of amino acid residues, which are involved in binding with carbohydrates and stability of the complex using structure-based parameters. Further, we have developed a database on binding affinity of protein-carbohydrate complexes, ProCaff (Protein-Carbohydrate complex binding Affinity Database), which contains 3713 entries on dissociation constant (Kd), Gibbs free energy change (ΔG), experimental conditions, sequence, structure and literature information. We used a set of 389 complexes from ProCaff database and related with structure-based features. We found that binding site residues, accessible surface area, interactions between various atoms and energy contributions are important to understand the binding affinity. We developed a multiple regression method, showed an average correlation and mean absolute error (MAE) of 0.73 and 1.15 kcal/mol, respectively, between experimental and predicted binding affinities on a jackknife test. Further, we validated our method using a blind data set of 40 complexes and we obtained a correlation and MAE of 0.88 and 1.02 kcal/mol, respectively. We have developed a web server PCA-Pred, for predicting the binding affinity of protein-carbohydrate complexes. Additionally, to understand the effect of mutations, we used a set of 318 unique mutations to develop a model for predicting binding free energy change upon mutation (ΔΔG) using the sequence-based features and developed multiple regression models, which showed an average correlation of 0.74 and a mean absolute error of 0.70 kcal/mol between experimental and predicted ΔΔG on 10-fold cross-validation. In addition, we have developed a novel database, CarbDisMut, a comprehensive integrated resource for disease-causing mutations with sequence and structural features. It has 1,170,330 disease-associated mutations and 38,636 neutral mutations from 7,187 human carbohydrate-binding proteins. CarbDisMut provides data on four levels such as gene, protein, carbohydrate and diseases. These resources help to relate the binding affinity with the disease-causing mutations and designing therapeutic drugs for diseases.
Publications:
1. Shanmugam, N.R.S., Veluraja, K., & Gromiha, M.M. (2022). PCA-MutPred: Prediction of binding free energy change upon mutation in protein-carbohydrate complexes. J Mol Biol. 434(11), 167526. https://doi.org/10.1016/j.jmb.2022.167526
2. Shanmugam, N.R.S., Blessy, J.J., Veluraja, K., & Gromiha, M.M. (2021). Prediction of protein-carbohydrate complex binding affinity using structural features. Brief Bioinform. 22(4), bbaa319. https://doi.org/10.1093/bib/bbaa319
3. Shanmugam, N.R.S., Blessy, J.J., Veluraja, K., & Gromiha, M.M. (2020). ProCaff: protein-carbohydrate complex binding affinity database. Bioinformatics. 36(11), 3615-3617. https://doi.org/10.1093/bioinformatics/btaa141
4. Shanmugam, N.R.S., Selvin, J.F.A., Veluraja, K., & Gromiha, M.M. (2018). Identification and Analysis of Key Residues Involved in Folding and Binding of Protein-carbohydrate Complexes. Protein Pept Lett, 25(4), 379-89. https://doi.org/10.2174/0929866525666180221122529
5. Shanmugam, N.R.S., Kulandaisamy, A., Veluraja, K., & Gromiha, M.M. (2022). CarbDisMut: Database on neutral and disease-causing mutations of human carbohydrate-binding proteins. (Under revision)