VitaPred is a SVM (Support Vector Machines) based prediction method for the vitamin-interacting residues in protein sequences. Vitamins play vital role in the cell and involves in the various enzymatic reaction and functions as a cofactor. VitaPred predicts that which amino acid residues in the protein sequence will interact with the vitamins. An example of protein-vitamin binding given in the following diagram, where pyridoxal 5'-phosphate (PLP, red color) interacts with specific residues (yellow color) of a protein (cyan color).
VitaPred contains four different SVM-based prediction modules, which were developed using evolutionary (PSSM) and Binary information. There was need to develop different prediciton modules for vitamins and their sub-types because their protein binding patterns are different from each other. These difference are present in the following Two Sample Logos (TSL) of four different sub-type of vitamins.
(i) TSL of vitamin-interacting residues (VIRs)
(ii) TSL of vitamin A interacting residues (VAIRs)
(iii) TSL of vitamin B interacting residues (VBIRs)
(iv) TSL of vitamin B(6)-derived pyridoxal 5'-phosphate interacting residues (PLPIRs)
Datasets: We have provided all the datasets, which were used in the development and evaluation of VitaPred, please download datasets from here: VitaPred datasets