Metaviral SPAdes: assembly of viruses from metagenomic data

Abstract

Motivation Although the set of currently known viruses has been steadily expanding, only a tiny fraction of the Earth’s virome has been sequenced so far. Shotgun metagenomic sequencing provides an opportunity to reveal novel viruses but faces the computational challenge of identifying viral genomes that are often difficult to detect in metagenomic assemblies. Results We describe a MetaviralSPAdes tool for identifying viral genomes in metagenomic assembly graphs that is based on analyzing variations in the coverage depth between viruses and bacterial chromosomes. We benchmarked MetaviralSPAdes on diverse metagenomic datasets, verified our predictions using a set of virus-specific Hidden Markov Models and demonstrated that it improves on the state-of-the-art viral identification pipelines. Availability and implementation MetaviralSPAdes includes ViralAssembly, ViralVerify and ViralComplete modules that are available as standalone packages: https://github.com/ablab/spades/tree/metaviral_publication, https://github.com/ablab/viralVerify/ and https://github.com/ablab/viralComplete/. Supplementary information Supplementary data are available at Bioinformatics online.

Thousands of Previously Unknown Phages Discovered in Whole-community Human Gut Metagenomes

“…The putative phage contigs had a median length of 44.9 kb which is consistent with the recent estimates of the median genome size of dsDNA phages [13]. To exclude any contaminating contigs (e.g., a plasmid harboring an integrated phage), each was assessed with ViralVerify [16] and Seeker [26], two bioinformatic tools trained to discriminate phage genomes from other sequences. These tools classi ed almost all the selected contigs as phages with varying levels of con dence except for several cases that, upon manual examination, were found to represent false negative classi cations by these tools (Additional le 2).…”

“…Contigs were classi ed as phages when exceeding 3 kbp in length and possessing at least one ORF that matched a capsid, portal or large terminase subunit protein pro le below the e-value threshold. The phage classi cations were cross-checked with Seeker [26] and ViralVerify [16]. In cases where both tools classi ed a contig as non-phage, the protein annotations were examined manually, revealing four contigs of ambiguous identity that were discarded.…”

“…Moreover, similar-sized plasmids also assemble into circular contigs [15]. A recently developed computational method aims to address this problem by focusing speci cally on the assembly of circular phage genomes and their automatic discrimination from plasmids based on gene content [16]. The genetic repertoire shared between plasmids and phages, for example, the parABS partitioning system encoded by both Escherichia coli phage P1 and plasmids [17], can obfuscate their automatic annotation-based discrimination and necessitate manual curation.…”

Cenote-Taker 2 Democratizes Virus Discovery and Sequence Annotation

“…Moreover, it is known that some sequences, even in very clean virus-like particle preparations, are not viruses but mobile genetic elements that parasitize viral capsid machinery 46 . Two new tools capable of virus discovery have come out very recently, only after we completed our analysis 47,48 .…”

The role of mobile genetic elements in organic micropollutant degradation during biological wastewater treatment

“…Many bioinformatic tools have been recently developed to overcome this challenge in short-read sequencing. Examples of them are PlasmidFinder, cBar, PLACNET, (meta)plasmidSPAdes, recycler, plasflow, MOB-suite, mlplasmids, plaScope, gplas, plasClass, plasGUN,SCAPP, metaviralSPAdes ( Zhou and Xu, 2010 ; Carattoli et al., 2014 ; Rozov et al., 2017 ; Vielva et al., 2017 ; Arredondo-Alonso et al., 2018 ; Arredondo-Alonso et al., 2020 ; Krawczyk et al., 2018 ; Robertson and Nash, 2018 ; Royer et al., 2018 ; Antipov et al., 2019 , 2020 ; Fang et al., 2020 ; Pellow et al., 2020a , Pellow et al., 2020b ; Pellow et al., 2020 , 2020 ). These tools identify plasmids and viruses in metagenomic samples or bacterial isolates in a variety of ways, ranging from the simpler approach of blasting against a plasmid database (PlasmidFinder, MOB-suite) to the use of more complex deep neural networks (cBar, PlasFlow).…”