Developed by V. Suresh and S. Parthasarathy*
Dept. of Bioinformatics, Bharathidasan University, Tiruchirappalli 620 024, India
* For correspondence

The SVM-PB-Pred[1] is a Protein Block (PB)[2] prediction server based on Support Vector Machine (SVM) approach. This method uses position specific scoring matrix (PSSM) profiles and secondary structures for the prediction of PB.SVM-PB-Pred has four steps to proceed the PB prediction. First three steps for upload the three input files namely, Amino acid sequence, predicted Secondary Structure and PSSM profile. In step 4, user needs to select the SVM models for PB prediction. Based on the length of the query sequence and the selection of SVM models available in SVM-PB-Pred server, real time taken for prediction of PB may vary.

Click here to use the SVM-PB-Pred server. - To predict the PB sequence.

Click here to use the PredictFold-PB server.- To predict the possible fold of an unknown sequence using PB.[3]

Click here to get the supplementary data used in methodlogy. - To extract different datasets used in methodology.[4]

Click here for sample input files. - Amino acid sequence, Secondary structures, PSSM profiles.


[1]  Suresh,V and Parthasarathy,S. (2014) SVM-PB-Pred: SVM Based Protein Block Prediction Method using Sequence Profiles and Secondary Structures. Protein and Peptide Letters, 21(8), 736-742 .

[2] de Brevern,A.G., Etchebest,C and Hazout,S. (2000) Bayesian probabilistic approach for predicting backbone structures in terms of protein blocks. Proteins, 41, 271-287.

[3]  Suresh,V., Ganesan,K and Parthasarathy,S. (2013) A Protein Block based Fold Recognition Method for the Annotation of Twilight Zone Sequences. Protein and Peptide Letters, 20(3), 249-254 .

[4]  Suresh,V., Ganesan,K and Parthasarathy,S. (2012) PDB-2-PB: A curated online Protein Block Sequence Database. J. Appl. Cryst., 45, 127 -129.

Please, send your comments, suggestions to:

© 2011 - S. Parthasarathy, Bharathidasan University, India