DPROT is a bi-layer cascade Support Vector Machine (SVM). This method involves the use of bi-layer. The first layer SVM classifiers was trained and optimized with different individual protein sequence features like amino acid composition, dipeptide composition (occurrences of the possible pairs of i and i+1 amino acid residues), higher order dipeptide composition (pairs of i and i+2 residues) and PSSM (Position Specific Scoring Matrix) composition and secondary structure composition. Cascade module based on these features encapsulates comprehensive information that contributes to effective prediction.
            A five-fold cross-validation technique was used for the evaluation of various prediction strategies in the current work. The results from the first layer were cascaded to the second layer SVM classifier to train and generate the final classifier.
            The cascade SVM classifier was able to achieve 80.5, 96.8, 94.6 and 0.77, corresponding to sensitivity, specificity, accuracy and MCC respectively.




Residues most prominent in disordered proteins

If you are using this webserver, please cite:

Sethi D, Garg A and Raghava, G. P. S. (2008) DPROT: Prediction of
Disordered Proteins using Evolutionary Information. Amino Acids.35:599-60