DRHP-PseRA: Detecting remote homology proteins using profile-based pseudo protein sequence and rank aggregation

Junjie Chen, Ren Long, Xiao Long Wang, Bin Liu*, Kuo Chen Chou

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

89 Citations (Scopus)

Abstract

Protein remote homology detection is an important task in computational proteomics. Some computational methods have been proposed, which detect remote homology proteins based on different features and algorithms. As noted in previous studies, their predictive results are complementary to each other. Therefore, it is intriguing to explore whether these methods can be combined into one package so as to further enhance the performance power and application convenience. In view of this, we introduced a protein representation called profile-based pseudo protein sequence to extract the evolutionary information from the relevant profiles. Based on the concept of pseudo proteins, a new predictor, called "dRHP-PseRA", was developed by combining four state-of-the-art predictors (PSI-BLAST, HHblits, Hmmer, and Coma) via the rank aggregation approach. Cross-validation tests on a SCOP benchmark dataset have demonstrated that the new predictor has remarkably outperformed any of the existing methods for the same purpose on ROC50 scores. Accordingly, it is anticipated that dRHP-PseRA holds very high potential to become a useful high throughput tool for detecting remote homology proteins. For the convenience of most experimental scientists, a web-server for dRHP-PseRA has been established at http://bioinformatics.hitsz.edu.cn/dRHP-PseRA/.

Original languageEnglish
Article number32333
JournalScientific Reports
Volume6
DOIs
Publication statusPublished - 1 Sept 2016
Externally publishedYes

Fingerprint

Dive into the research topics of 'DRHP-PseRA: Detecting remote homology proteins using profile-based pseudo protein sequence and rank aggregation'. Together they form a unique fingerprint.

Cite this