Viral RNA enrichment vs. filtering reads out

Question

[stefan.dvorezky]

I see a lot of advertisement for enrichment kits, e.g. Twist Biosciences hybrid preps, which should allow to "fish" for viral DNA in vitro. Now there is another method exploited by a paper that is based on metagenomic sequencing and filtering for viral reads in silico. When I clearly see the advantage of second method in not having to buy an enrichment kit, what could speak for enrichment in vitro with a kit?

[GenoMax]

You should post this question over at SeqAnswers.com or Biology stackexchange. This is not really a bioinformatics question and more suitable for other fora noted.

Answer 1

[Mensur Dlakic]

If you are only interested in viruses, and you do this without enrichment, you will be throwing away more than 95% of your sequence data as viral DNA fraction will be small. On the other hand, it may be an overkill to sequence viral genomes using a full lane.

While the choice ultimately depends on the viral size and how badly one wants to get a complete viral genome, I would readjust my goals if I were you, and sequence without enrichment. That way I would get some information about the host, potentially about its CRISPR defense system and CRISPR arrays.

[stefan.dvorezky]

Well, we are talking about human samples here, so the host is known. So what are the advantages of viral enrichment in this case, and why people do it?

[GenoMax]

For the same reason as poly-A selection or other rRNA depletion methods in RNAseq studies. If you don't do this 95+% of the data you will sequence will be of no use to you (e.g. assuming you have no interest in human data or rRNA in example above). That is literally money down the drain.