Viruses and CRISPR · What you can do here
Find viruses and prophages in a genome or metagenome
Identify viral contigs and integrated proviruses in assembled genomes or metagenomes using geNomad, then determine their completeness and contamination with CheckV.
What this involves
Phage identification is a sequence of distinct questions that must be addressed in order. Determining whether a contig is viral and checking its completeness must happen before attempting gene annotation, lifestyle prediction, or taxonomic classification. Answering these questions out of order leads to confident false conclusions.
The process starts with geNomad, which scans your assembly contigs to identify viral sequences, score them alongside marker evidence, assign broad taxonomy, and locate provirus coordinates in host sequence. Following identification, CheckV assesses completeness, contamination, and assigns an MIUViG quality tier to each sequence.
The platform's agent executes these programs and manages their reference databases—including geNomad's ~840 MB database and CheckV's ~1.7 GB database—on a dedicated machine. The agent reads the results directly, passing trimmed provirus sequences on to subsequent steps without requiring local installation or manual data transfers.
How it goes
- 1Identify viral contigs and integrated provirusesThe agent runs geNomad on your FASTA sequence to detect viral contigs and integrated proviruses, outputting viral scores, supporting marker genes, broad taxonomy, and proviral coordinates.
- 2Assess completeness and contaminationCandidate viral sequences are processed with CheckV to calculate completeness, detect host contamination, and assign each contig to an MIUViG quality tier.
- 3Trim provirus coordinates for downstream analysisCheckV trims flanking host sequence from integrated proviruses identified by geNomad, providing clean viral coordinates necessary for honest downstream annotation.
Start it here
Ask in plain language. The agent works on a Linux machine of its own — it installs and runs what the job needs, keeps every file, and reads the results back with the numbers explained. Nothing to set up, and a long run continues after you close the page.
What to watch for
- A virus score from geNomad is a ranking rather than a probability, so report the supporting marker gene evidence beside the score instead of treating it as a likelihood.
- geNomad detects all viral contigs and plasmids together, meaning you must check the taxonomy column before categorising a candidate as a bacteriophage.
- Completeness is calculated relative to reference genomes; genuinely novel phages with no close relatives score low because of absent comparators rather than missing sequence.
- Check the completeness_method field before citing a percentage, and report the MIUViG quality tier as the primary summary of sequence quality.
- Prophage boundaries identified by prediction algorithms are inherently approximate unless definitive host attachment sites have been located.
What it runs
Programs
- geNomadRuns as a job on its own machine· BSD-4-Clause (per the bioconda recipe; the repository's own LICENSE is not SPDX-tagged)Identifies plasmids and viruses in sequencing data, predicts proviruses, and assigns viral taxonomy.
- CheckVRuns as a job on its own machine· BSD-3-Clause-LBNLA command-line pipeline for assessing the completeness, contamination, and quality of viral genomes from metagenomes.
nf-core pipelines
Questions
Do I need to install anything to find viruses and prophages in a genome or metagenome?
No. The research agent does this work on a Linux machine of its own — it runs geNomad, CheckV there, keeps the files, and reads the results back to you. Long runs continue if you close the page.
What sequence inputs can I provide for this analysis?
You can supply an NCBI nucleotide accession such as NC_001416.1, a direct URL to a sequence file, or a path to a FASTA file. If you have an NCBI assembly accession such as GCF_..., you must fetch the genomic FASTA first because assembly accessions do not contain sequence directly.
Why must CheckV be run before functional annotation or lifestyle prediction?
Downstream tools depend on knowing if a sequence is complete; for instance, lifestyle predictors infer virulence from the absence of lysogeny domains, which returns false confidence on incomplete fragments. CheckV also trims host flanking regions from proviruses, which prevents host genes from being annotated as viral.
Can I use this approach on large multi-sample virome datasets?
Yes, although when processing large virome datasets with multiple samples, running the indexed nf-core/phageannotator workflow through the agent is preferable to manually chaining single-sample tools.
How are integrated prophages distinguished from free viral contigs?
geNomad detects proviruses integrated within host chromosomal contigs and outputs their specific start and end coordinates, whereas whole viral contigs are reported without host flanking sequence.
Written from the method the research agent follows for this work and the documentation of the tools listed above. Statements citing a figure that does not appear there are removed before publication. Corollary is a non-commercial research demonstration; results are not for clinical use.