TMFoldWeb: A web server for predicting transmembrane protein fold class

Dániel Kozma, Gábor E. Tusnády

Research output: Contribution to journalArticle

5 Citations (Scopus)


Background: Here we present TMFoldWeb, the web server implementation of TMFoldRec, a transmembrane protein fold recognition algorithm. TMFoldRec uses statistical potentials and utilizes topology filtering and a gapless threading algorithm. It ranks template structures and selects the most likely candidates and estimates the reliability of the obtained lowest energy model. The statistical potential was developed in a maximum likelihood framework on a representative set of the PDBTM database. According to the benchmark test the performance of TMFoldRec is about 77% in correctly predicting fold class for a given transmembrane protein sequence. Results: An intuitive web interface has been developed for the recently published TMFoldRec algorithm. The query sequence goes through a pipeline of topology prediction and a systematic sequence to structure alignment (threading). Resulting templates are ordered by energy and reliability values and are colored according to their significance level. Besides the graphical interface, a programmatic access is available as well, via a direct interface for developers or for submitting genome-wide data sets. Conclusions: The TMFoldWeb web server is unique and currently the only web server that is able to predict the fold class of transmembrane proteins while assigning reliability scores for the prediction. This method is prepared for genome-wide analysis with its easy-to-use interface, informative result page and programmatic access. Considering the info-communication evolution in the last few years, the developed web server, as well as the molecule viewer, is responsive and fully compatible with the prevalent tablets and mobile devices. Reviewers: This article was reviewed by Dr. Michael Gromiha, Dr. Sandor Pongor and Dr. Frank Eisenhaber with Wing-Cheong Wong.

Original languageEnglish
Article number54
JournalBiology Direct
Issue number1
Publication statusPublished - Sep 17 2015


  • Fold recognition
  • Ready for large-scale analysis
  • Statistical potential
  • Transmembrane protein

ASJC Scopus subject areas

  • Immunology
  • Ecology, Evolution, Behavior and Systematics
  • Modelling and Simulation
  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)
  • Applied Mathematics

Fingerprint Dive into the research topics of 'TMFoldWeb: A web server for predicting transmembrane protein fold class'. Together they form a unique fingerprint.

  • Cite this