There are controversial questions about the role of viruses in some types of cancer. There are some viruses "associated with" or "suspected to play a role in" some types of cancer. The major problem is that there is a lack of evidence for the presence of the virus in the majority of the observed cases leading to the conclusion that the virus cannot be considered as the cause of cancer.
So my question is simple, can we certify that there is an absence of a virus in a cell with today's techniques or are we limited with a detection threshold if a small number of copies may be sufficient to trigger cancerogenicity?
There are a few categories of answer to your questions:
[Theoretical Answer] You only need 1 copy of an RNA to detect it, providing the sequencing depth is sufficiently high, the enrichment for RNA doesn't preclude the viral RNA (like oligo-dT priming might), and you are aligning your RNA-seq to a viral genome. If you set up spiked in controls, you could detect it providing you know what you are looking for.
[Practical Answer] A virus that is causing cancer will usually be pretty abundant. A paper by my colleagues Aparna Bhaduri and Kun Qu developed software to detect viruses in RNA-seq (see here for details about how many reads were needed, etc.: http://bioinformatics.oxfordjournals.org/content/early/2012/02/28/bioinformatics.bts100). For practical purposes, I'd say that if you know what you're looking for, you can make a yes/no call with reasonable accuracy on the presence of a virus.
[Other] Even if you detect a million copies of an RNA virus in the cell, it doesn't matter if the virus has an unknown sequence and you aren't aligning to it, or even if you do de-novo sequence assembly, and don't know what the virus does. There are likely many viruses that have an unknown effect and you need to do a Koch Postulate-Style (http://en.wikipedia.org/wiki/Koch's_postulates) set of experiments to see if that was indeed your oncogenic event.
- Badges
- Science method