TargetATP is a full-featured GUI application for targeting protein-ATP binding residues in protein sequences, consisting of two parts of local client and remote server. Once receiving the request from the TargetATP client locally run on the user's computer, the remote server will be activated for calculations, which will extract sequence's multi-view features and then input them to an ensembled support vector machines for prediction. The results will be sent back to the client application in two ways: online real-time feedback and independent email notifications. TargetATP achieves an overall AUC value of 0.895 on a benchmark dataset consisting of 227 ATP-binding protein sequences. TargetATP was programmed with Java and is distributed via Java Web Start technology, having the merits of cross-platform, flexibility, and extensibility to be deployed and used on multiple major operating systems, such as Linux and Windows. TargetATP is freely available at http://www.csbio.sjtu.edu.cn/bioinf/targetATP/ for academic use.

The TargetATP client application has been strictly tested on several different operating systems including Windows 2000/XP/VISTA and GNU/Linux. Note that the JRE (Java Runtime Environment, Version 1.6 or higher) should be installed on your computer in advance to enable the TargetATP client application to work properly. If your computer does not have a JRE, please click the hyperlink JRE 1.6 (WIN) or JRE 1.6 (LINUX) according to the operating system on your computer in the first page of the TargetATP web server to download and then install it.

News from Softpedia:  "TargetATP" has been tested in the Softpedia labs using several industry-leading security solutions and found to be completely clean of adware/spyware components. We are impressed with the quality of your product and encourage you to keep these high standards in the future. We have granted "TargetATP" with the "100% CLEAN" Softpedia award. Details can be found at: http://www.softpedia.com/progClean/targetATP-Clean-216680.html.

We have extracted PSSM+PSS feature for each amino acid in a protein sequence.
Note that the sliding window size is 17, thus the dimensionality of LogisticPSSM feature is 340 and the dimensionality of PSS feature is 51.
Any problem, please contact: Jun Hu via h19891007@126.com or Dr. Yu via njyudj@njust.edu.cn
If you use the feautre we provided, please cite the paper:

Dong-Jun Yu, Jun Hu, Zhen-Min Tang, Hong-Bin Shen, Jian Yang, and Jing-Yu Yang. Improving Protein-ATP Binding Residues Prediction by Boosting SVMs with Random Under-Sampling. Neurocomputing. 2013, 104: 189-199. (DataSet.rar)

visits