TBpred Prediction Server Home
Click HERE to get the data set of 852 proteins used to develop TBpred.
TBpred is a prediction server that predicts four subcellular localization (cytoplasmic,integral membrane,secretory and membrane attached by lipid anchor) of mycobacterial proteins.It is SVM based method that exploits defferent features of protein such as amino acid compositin, dipeptide composition and position specific scoring matrix (PSSM).The overall prediction accuracy of these SVM modules are 82.51, 80.39 and 86.62% respectively.Along with SVM other techniques like profile HMM and MEME/MAST motif based studies were also applied.Moreover a hybrid approach combining the pssm based SVM model and the MEME/MAST model has been incorporated.
Aim: This server aims at predicting the subcellular localization of mycobacterial proteins.
Facilities:The users of the server may select one approach at a time out of four approaches (three SVM based and one hybrid approach) to get the prediction results.
Importance of this webserver:
Location of a protein inside a cell gives an insight into its function. So this server may serve as a tool for functional annotation of mycobacterial protein.
The organism specific classifier is better than the generalised one. Hopefully the server can allocate the protein's subcellular localization more correctly.
The un-annotated portion of mycobacterial genomes can be annotated and new potential drug/vaccine targets would be identified.
Click HERE to get the Supplementary Material.
Click HERE to get the genome annotation of Mycobacterium tuberculosis H37Rv on dipeptide composition based model
Click HERE to get the genome annotation of Mycobacterium tuberculosis H37Rv on Position Specific Scoring Matrix (PSSM) composition based model
Click HERE to get the genome annotation of Mycobacterium tuberculosis H37Rv on Hybrid Approach based model
Citation: If you are using this server please cite
Rashid M, Saha S, Raghava GPS (2007):Support Vector Machine-based method for predicting subcellular localization of mycobacterial proteins using evolutionary information and motifs. BMC Bioinformantics, 8:337.