Performance Analysis of Methods that Predict Transmembrane Regions

CHEN Zhong-Qiang*, LIU Qi, ZHU Yi-Sheng, LI Yi-Xue1
( Department of Biomedical Engineering, Shanghai Jiaotong University, Shanghai 200030, China; 1Bioinformation Center, Shanghai Institutes for Biological Sciences, the Chinese Academy of Sciences, Shanghai 200031, China )

Abstract    The increasing protein sequences from the genome project require the oretical methods to predict transmembrane regions. Evaluation of these prediction methods will not only help biologists to select appropriate methods, but also provide bioinformatian guidance to develop new methods. In this work, the latest membrane protein topology database was used as the test set and selected soluble proteins was used as control.The analyses of performance of published online prediction methods are presented. Results show that HMMTOP (2.0) has the best performance among current methods.
Key words     membrane protein;transmembrane region;secondary structure prediction

*Corresponding author: Tel, 86-21-62932832-155; e-mail, [email protected]