ModPred is a sequence-based predictor of potential post-translational modification (PTM) sites in proteins. It consists of 34 ensembles of logistic regression models, trained separately on a combined set of 126,036 non-redundant experimentally verified sites for 23 different modifications, obtained from public databases and an ad-hoc literature search. Areas under the ROC curve (AUCs) were estimated to range from ~60 to 97%, depending on the type of PTM.

Supplementary information, usage instructions and more about the datasets can be found on the help page.

ModPred executable files (command-line and user-interface versions) are available for download:

Installation guide for the command-line version can be downloaded here and for the user-interface version can be downloaded here.

ModPred was developed by Vikas Pejaver (Indiana University, School of Informatics and Computing) and Predrag Radivojac (Indiana University, School of Informatics and Computing)

In citing the ModPred software, please refer to:

Pejaver, V., Hsu, W-L., Xin, F., Dunker, A. K., Uversky, V. N., and Radivojac, P. The structural and functional signatures of proteins that undergo multiple events of post-translational modification Protein Science. 23(8):1077-1093. (2014)

Please direct all comments and suggestions to predrag@indiana.edu or vpejaver@indiana.edu.