Karius uses DNA sequencing of everything that's contained in a sample to detect all bacteria and DNA-viruses in the sample. Given the way that costs of sequencing are falling this approach seems to be the future of detecting bacteria based infections.

It would be great if the same approach would also work for RNA-based viruses. For it to work it would of course be necessary that there's enough RNA of the virus in comparison to the other RNA.

How much virus RNA is there in a person who's infected by an RNA virus in comparison to the other RNA?

  • $\begingroup$ You didn't get an answer to your specific question, though @de_novo answered a related one. It's not possible to give an answer to How much virus RNA is there in a person who's infected by an RNA virus because it's too general; RNA viruses are wildly diverse and include viruses that are highly localized and don't enter the blood at all, as well as some that are blood-borne and can reach fairly high concentrations. So the answer to your question would be cover at least a billion-fold range if not more, which isn't helpful. $\endgroup$
    – iayork
    Aug 3, 2018 at 18:33

1 Answer 1


I think your question is:

Can you diagnose someone with an RNA virus by looking for that RNA in their blood?

The use of nucleic acid tests, or NATs in diagnosis has been around for a long time. The specific for profit company you mention is not the first to use them. Almost all NATs work by first amplifying the genetic material with PCR or other methods, so there does not need to be a large proportion of pathogen RNA vs total RNA in a sample in order to use a NAT.

NATs, rather than culture or antigen/antibody based tests, provide the definitive diagnosis for many viruses, RNA viruses included. As the CDC describes, NATs (among other tests) are also used to screen donated blood for RNA viruses (HCV, HIV, and West Nile).

Incidentally, the for profit company you mentioned does NOT detect all bacteria and DNA-viruses in the sample as you said. It screens for a large number of specific pathogens in their

proprietary micro-organism reference database

  • $\begingroup$ If there's a new bacteria in the blood of patients then if enough of the bacteria is in blood samples that are given to Karius, Karius automatically gets the DNA of those bacteria through their process. Snippets of all DNA in the blood are picked up by their process even when the software analysis might not match all the snippets to organisms. NAT's in the sense that Wikipedia describes the term you actually have to know what organism you are looking for to choose the right primer. More importantly it's price increases linearly if you are looking for more organisms. $\endgroup$
    – Christian
    Jun 26, 2018 at 16:53
  • $\begingroup$ @Christian you still have to know what you're looking for in the Karius test. It matches the non human DNA in a sample to its "proprietary micro-organism reference database." If it's not in the database, it doesn't get identified. $\endgroup$
    – De Novo
    Jun 26, 2018 at 16:56
  • $\begingroup$ @Christian moreover, you don't have to have a large proportion of a nucleic acid for this same type of test to work. Karius's methods are a change in scale from standard of care, not type. If you read the studies from their group, they are not using a novel amplification or detection technique. $\endgroup$
    – De Novo
    Jun 26, 2018 at 16:59
  • $\begingroup$ There test is based on next generation DNA sequencing. That's a technology that recently got a lot of cheaper. The Wikipedia page on NAT speaks about probe-target binding and not about next generation sequencing. $\endgroup$
    – Christian
    Jun 26, 2018 at 17:22

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .