RiPPMiner-Genome: A Web Resource for Automated Prediction of Crosslinked Chemical Structures of RiPPs by Genome Mining

被引：22

作者：

Agrawal, Priyesh ^{[1
]}

Amir, Sana ^{[1
]}

Deepak ^{[1
]}

Barua, Drishtee ^{[1
]}

Mohanty, Debasisa ^{[1
]}

机构：

[1] Natl Inst Immunol, Aruna Asaf Ali Marg, New Delhi 110067, India

来源：

JOURNAL OF MOLECULAR BIOLOGY | 2021年 / 433卷 / 11期

关键词：

RiPP; cross-link; genome mining; machine learning; chemical structure prediction; PEPTIDE NATURAL-PRODUCTS; DISCOVERY; INSIGHTS;

D O I：

10.1016/j.jmb.2021.166887

中图分类号：

Q5 [生物化学]; Q7 [分子生物学];

学科分类号：

071010 ; 081704 ;

摘要：

RiPPMiner-Genome is a unique bioinformatics resource for identifying Biosynthetic Gene Clusters (BGC) for RiPPs (Ribosomally Synthesized and Post-translationally Modified Peptides) and automated prediction of crosslinked chemical structures of RiPPs starting from genomic sequences. It is a major update of the RiPPMiner webserver, which used only peptide sequence of RiPP precursors as input for predicting RiPP class and crosslinked chemical structures. Other major improvements are, machine learning (ML) based identification of correct RiPP precursor peptide from among multiple small ORFs (Open Reading Frames) in a BGC, prediction of the cleavage site and cross-links in thiopeptides and identification of non-crosslinked modified residues in lanthipeptides. It has been benchmarked on a dataset of 204 experimentally characterized RiPP BGCs. RiPPMiner-Genome also facilitates visualization of the RiPP BGCs and depiction of the chemical structure of crosslinked RiPP. It also has an interface for searching characterized RiPPs, similar to the predicted core peptide sequence or crosslinked chemical structure. (C) 2021 Elsevier Ltd. All rights reserved.

引用

页数：10