SelenzymeRF: updated enzyme suggestion software for unbalanced biochemical reactions

Ruth Stoney, Erik Hanko, Pablo Carbonell, Rainer Breitling

Research output: Contribution to journalArticlepeer-review

Abstract

Selenzyme is a retrobiosynthesis tool that suggests candidate enzymes for user query reactions. Enzyme suggestions are based on identical reactions, as well as similar reactions, since enzymes are often capable of promiscuous substrate binding. Selenzyme is a user-friendly, widely used web-tool for ranking enzymes based on reaction similarity and additional features, including the phylogenetic distance between the source species of the enzyme and the intended host. While Selenzyme has proved invaluable in assisting with enzyme selection for known reactions, as well as many novel or orphan reactions, weaknesses have been exposed in its ability to rank functionally related enzymes. Within this update, we introduce a new reaction similarity scoring algorithm, which is used in conjunction with the previous similarity calculation, to improve the accuracy of enzyme suggestions based on non-identical similar reactions, across a range of EC reaction classes. This allows enzymes to be suggested for reactions not found within the database, even if the reaction is unbalanced. A database update was also carried out, to ensure that reaction and enzyme knowledge remains current. This update can be accessed at http://selenzymeRF.synbiochem.co.uk/.

Original languageEnglish
Article number5868-5876
Pages (from-to)5868-5876
Number of pages9
JournalComputational and Structural Biotechnology Journal
Volume21
DOIs
Publication statusPublished - 23 Nov 2023

Keywords

  • Retrobiosynthesis
  • enzyme selection
  • reacting site
  • reaction similarity
  • unbalanced reactions
  • RXNMapper
  • Enzyme selection
  • Reaction similarity
  • Reacting site
  • Unbalanced reactions

Fingerprint

Dive into the research topics of 'SelenzymeRF: updated enzyme suggestion software for unbalanced biochemical reactions'. Together they form a unique fingerprint.

Cite this