CRF: detection of CRISPR arrays using random forest

Abstract

CRISPRs (clustered regularly interspaced short palindromic repeats) are particular repeat sequences found in wide range of bacteria and archaea genomes. Several tools are available for detecting CRISPR arrays in the genomes of both domains. Here we developed a new web-based CRISPR detection tool named CRF (CRISPR Finder by Random Forest). Different from other CRISPR detection tools, a random forest classifier was used in CRF to filter out invalid CRISPR arrays from all putative candidates and accordingly enhanced detection accuracy. In CRF, particularly, triplet elements that combine both sequence content and structure information were extracted from CRISPR repeats for classifier training. The classifier achieved high accuracy and sensitivity. Moreover, CRF offers a highly interactive web interface for robust data visualization that is not available among other CRISPR detection tools. After detection, the query sequence, CRISPR array architecture, and the sequences and secondary structures of CRISPR repeats and spacers can be visualized for visual examination and validation. CRF is freely available at http://bioinfolab.miamioh.edu/crf/home.php.

CRISPRCasFinder, an update of CRISRFinder, includes a portable version, enhanced performance and integrates search for Cas proteins

“…Several programs have been developed to identify CRISPR arrays in genomic sequences, the most frequently cited being CRISPRFinder ( 7 ), CRT ( 8 ) and PILER-CR ( 9 ). Additional programs such as CRISPRDetect ( 10 ), CRISPRdigger ( 11 ) and CRF ( 12 ) are also available. Three programs have been proposed for CRISPR array strand prediction based on the characteristics of the CRISPR repeat and the leader: CRISPRDirection using CRISPRDetect ( 10 , 13 ), CRISPRstrand ( 14 , 15 ) and CRISPRleader ( 16 ).…”

MetaCRAST: reference-guided extraction of CRISPR spacers from unassembled metagenomes

“…It searches each read for DR sequences matching query DRs specified by the user. These DRs can be selected from CRISPR arrays detected with genomic CRISPR detection tools such as PILER-CR ( Edgar, 2007 ), CRF ( Wang & Liang, 2017 ), or CRISPRFinder ( Grissa, Vergnaud & Pourcel, 2007b ) in fully assembled microbial genomes or assembled metagenomic contigs. The steps of the MetaCRAST pipeline are outlined in Fig.…”

In silico Method in CRISPR/Cas System: An Expedite and Powerful Booster

“…For instance, if users want to compare the CRISPR arrays from different species, CRISPRcompar ( 27 ) comprising CRISPRcomparison and CRISPRtionary and basically derived from CRISPRFinder must be the best choice. Besides, CRF ( 28 ) based on CRT added random forest algorithm to make an extra filtration for invalid CRISPR arrays, but this learning-based tool may partially lose the ability to discover new CRISPRs. Beyond that, three representative tools are designed for CRISPR strand prediction using the characteristics of leader and repeat that include CRISPRstrand ( 29 ), CRISPRleader ( 31 ), and CRISPRDirection ( 37 ).…”