VitaPred: Prediction of Vitamin Interacting Residues in Vitamin Binding Proteins Using Evolutionary Information

VitaPred: Prediction of Vitamin Interacting Residues in Vitamin Binding Proteins Using Evolutionary Information VitaPred is a computational tool developed for predicting vitamin-interacting residues in vitamin-binding proteins from amino acid sequences. Vitamins act as important cofactors in many enzymatic reactions. Identification of vitamin-binding residues can help in understanding enzyme activity, vitamin-protein interactions, and drug target development. VitaPred uses evolutionary information and machine learning models to predict residues interacting with vitamins and vitamin subclasses. Web Server: https://webs.iiitd.edu.in/raghava/vitapred/ Citation Panwar, B., Gupta, S., and Raghava, G. P. S. Prediction of vitamin interacting residues in a vitamin binding protein using evolutionary information. BMC Bioinformatics, 14, 44, 2013. https://doi.org/10.1186/1471-2105-14-44 About the Research Vitamins are essential cofactors involved in many metabolic and enzymatic reactions. Many vitamin-dependent enzymes are important drug targets, especially in diseases such as tuberculosis, malaria, cancer, and other metabolic disorders. Experimental identification of vitamin-binding sites is difficult because many protein structures are unavailable and vitamin-protein interactions can be complex. Therefore, VitaPred was developed to predict vitamin-interacting residues directly from protein sequence information. Data Compilation: Protein-vitamin interaction data were collected from SuperSite and Protein Data Bank structures. Ligand Protein Contact analysis was used to identify residues interacting with vitamins using a 5 Å distance cutoff. Methodology: VitaPred uses machine learning models trained on binary sequence profiles and Position-Specific Scoring Matrix profiles. PSSM-based evolutionary information was found to improve prediction performance compared to binary sequence-based models.