PEDS Advance Access published online on August 27, 2008
Protein Engineering Design and Selection, doi:10.1093/protein/gzn039
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Methods |
CSS-Palm 2.0: an updated software for palmitoylation sites prediction
1Hefei National Laboratory for Physical Sciences at Microscale and School of Life Sciences, University of Science & Technology of China, Hefei, Anhui 230027, China
2 To whom correspondence should be addressed. E-mail: yaoxb{at}ustc.edu.cn (Y.X.)/xueyu{at}ustc.edu.cn (X.Y.)
Protein palmitoylation is an essential post-translational lipid modification of proteins, and reversibly orchestrates a variety of cellular processes. Identification of palmitoylated proteins with their sites is the foundation for understanding molecular mechanisms and regulatory roles of palmitoylation. Contrasting to the labor-intensive and time-consuming experimental approaches, in silico prediction of palmitoylation sites has attracted much attention as a popular strategy. In this work, we updated our previous CSS-Palm into version 2.0. An updated clustering and scoring strategy (CSS) algorithm was employed with great improvement. The leave-one-out validation and 4-, 6-, 8- and 10-fold cross-validations were adopted to evaluate the prediction performance of CSS-Palm 2.0. Also, an additional new data set not included in training was used to test the robustness of CSS-Palm 2.0. By comparison, the performance of CSS-Palm was much better than previous tools. As an application, we performed a small-scale annotation of palmitoylated proteins in budding yeast. The online service and local packages of CSS-Palm 2.0 were freely available at: http://bioinformatics.lcd-ustc.org/css_palm.
Keywords: clustering and scoring strategy/CSS-Palm/palmitoylated proteins/palmitoylation
Received March 27, 2008; revised June 16, 2008; accepted June 20, 2008.