PlasForest: a homology-based random forest classifier for plasmid detection in genomic datasets
Résumé
Plasmids are mobile genetic elements that often carry accessory genes, and are vectors for horizontal transfer between bacterial genomes. The detection of plasmids in large sets of genomes is crucial to analyze their spread and quantify their role in bacteria adaptation and particularly in antibiotic resistance genes propagation. Several bioinformatics methods have been developed to detect plasmids. However, they suffer from low sensitivity ( i.e ., most plasmids remain undetected) or low precision ( i.e ., these methods identify chromosomes as plasmids), and are overall not adapted to identify plasmids in whole genomes that are not fully assembled (contigs and scaffolds). Here, we present PlasForest, a homology-based random forest classifier identifying bacterial plasmid sequences in unassembled genomes. This tool is based on the determination of homologies against a database of plasmid sequences, which allow a random forest classifier to discriminate plasmid contigs. Without knowing the taxonomical origin of the samples, PlasForest identifies contigs as plasmids or chromosomes with an accuracy of 98%. Notably, it can detect 96% of plasmid contigs over 50kb with 3.3% of false positives. PlasForest outperforms other currently available tools on test datasets by being both sensitive and precise. We implemented this tool in a user-friendly pipeline that can identify plasmids in large datasets in a reasonable amount of time.