What is the best way to get rid of unwanted DNA?

[VarroApis]

Dear seqanswers community,

I am currently planning a project on DNA sequencing of virus (probably using Illumina) and have a technical issue I would like to share with you.

I am aiming at comparing the diversity of a honey bee RNA virus (Deformed Wing Virus, or DWV) in several population of bees. Genomes of viral strains and recombinants are currently available (~10kb).

To do so, we (me and colleagues) are thinking of generating and sequencing cDNA. However, we want to get rid of as much honeybee DNA as possible in the final sequences. So, we are considering two options:
_ Multiple PCRs covering the whole virus genome to increase its proportion in the final yield
_ A few PCRs of the regions (~800-1000bp) where recombination is occuring (based on previous work) to focus on those interesting areas

It seems like the best way to have as much DWV sequences at the end. However, we are fairly new to the field. Is there anything we are missing here, or a better way to conduct our project?

Thanks a lot,

Alex

[gringer]

Would a nucleic acid capture work for you? Presumably if you're sequencing already known viruses, a probe capture targeting a conserved region should work:

https://www.thermofisher.com/nz/en/home/life-science/dna-rna-purification-analysis/napamisc/capture-of-biotinylated-targets/nucleic-acid-capture-assays.html

[nucacidhunter]

As you have reference genome for both virus and host you can prime RT reaction with few well designed and spaced primers to specifically revers transcribe only virus genome (I assume it is RNA). A more advanced approach would be adding partial Illumina adapter sequences to RT primer and also to randomers for second strand synthesis. The remaining adapter sequences can be added with a PCR reaction to amplify virus genome which will result in a library ready for sequencing.

[VarroApis]

Thank you very much for your great tips.

I do not know if a nucleic acid capture would work. I need to look at that into more details...

I quite like the RT reaction with virus primers yes, that looks like a great way to deal with the problem. Especially if it generates a library ready for sequencing.

Thanks a lot!