Cytokine prediction, cytopred

From DrugPedia: A Wikipedia for Drug discovery

CytoPred is a bioinformatics based server that predict Cytokine proteins and further classify them into families and subfamilies with high sensitivity and specificity. A PSI-BLAST + SVM based hybrid approach is adopted to develop the prediction methods. This method CytoPred is capable of distinguishing cytokines from non-cytokines with an accuracy of 98.29% and MCC 0.91. The overall accuracy of family and subfamily prediction is 96.34% and 92.64% respectively and MCC of 0.99 and 0.92 respectively.

This server allows the submission of sequence in any of the standard formats. The user can paste plain sequence in the provided text area. The server also has the facility for uploading the local sequence files.

Amino acid sequences must be entered in the one-letter code. All the non standard characters like [*&^%$@#!()_+~=;'",<>?.\|} are ignored from the sequence. The warning is displayed if the user submit input from both sources or no sequence is submitted.

User scan use only one of aforementioned methods(PSI-BLAST, SVM and Hybrid) for their prediction at a time. However, a user is suggested to provide just a single sequence if the prediction approach selected is PSI-BLAST based. In order to predict for multiple sequences, SVM based approach must be selected. This method could be of great help in order to annotate the proteins and would aid the experimental validation, in turn, saving time and labor. This server is developed at Dr. G.P.S.Raghava's group and is accessible for free at http://www.imtech.res.in/raghava/cytopred .