It also identifies the phylogenetically most related phage to the target prophages, as well as annotate the function of proteins throughout the phage genome. Prophage Hunter systematically locates prophage regions within bacterial genomes, and predicts the activity of prophages. Prophage Hunter provides both options, and allows the users to choose freely. While incorporating similarity searches increases prediction accuracy, skipping it might raise the possibility of finding novel phages. We developed Prophage Hunter, a novel integrative tool that employs sequence similarity-based searches within our customized phage parts library and prophage genetic features-based machine learning classification, to score the probability of a prophage being active. Also, like all other database-driven annotation systems, PHASTER only recognizes phages whose genes/proteins are close enough to the record in its database ( 14, 15). The evaluation system of PHASTER, however, simply adds up scores representing the number of nucleotides, total genes, cornerstone genes, and phage-like genes while neglecting common mutational events that prophages experience. Among these tools, only PHASTER considers the completeness of a putative prophage region. PhiSpy identifies prophages by focusing on the characteristics of prophages that exhibit no similarity to sequenced genomes ( 18). VirSorter detects prophages in complete microbial genomes or in fragmented genomic datasets, including incomplete genomes, SAGs, or metagenomic assemblies ( 17). MetaPhinder identifies assembled genomic fragments (i.e. contigs) of phage origin in metagenomic data sets by comparison with a database of whole genome bacteriophage sequences ( 16). PHASTER is a web server for the rapid identification and annotation of prophage sequences within bacterial genomes and plasmids ( 14, 15). VirFinder is the first k-mer based program for identifying prokaryotic viral sequences from metagenomic data ( 13). MARVEL predicts phage sequences in metagenomics bins based on random forest machine learning approach ( 12). Several tools have been developed to predict the existence of prophage sequences from bacterial NGS data (Table (Table1). With advances in synthetic biology, prophages are also considered as potential therapeutics for infectious or chronic diseases caused by bacteria ( 6–11). Prophages can participate in a number of bacterial cellular processes, including antibiotic resistance, stress response, and virulence ( 5). While some prophages are active-they can be induced by stresses like UV or antibiotics, others are defective due to bacterial defence systems or mutational decay ( 4). Prophages are temperate phages integrated in bacterial genomes. This is because nearly half of the sequenced bacteria are lysogens, representing a tremendous and previously under-explored source of prophages ( 3). Advances in next-generation sequencing (NGS) technologies support the easy access, analysis, and identification of temperate phages. The traditional source of phage information mainly depends on searching for phages in nature, which is stochastic, and sometimes difficult as proven for anaerobic bacteria and fastidious bacteria that grow only in specific nutrients and growth conditions ( 1, 2). Compared with over 199 000 bacterial genomes, fewer than 11 000 bacteriophage genomes are deposited in NCBI Genome as of 26 April 2019.
0 Comments
Leave a Reply. |