mapping unmapped reads to viral genome using blast+

Question

[Shaway]

Hello everyone,

I encountered small questions when analyzing AD patients' blood DNA NGS data. After mapping sequencing data to hg19 genome, I want mapping the remaining unmapped reads to viral genome. I found viral genome reference on ftp://ftp.ncbi.nlm.nih.gov/refseq/release/viral/, but there are 3 files (viral.1.1.genomic.fna.gz, viral.2.1.genomic.fna.gz, viral.3.1.genomic.fna.gz).

I merged these 3 files and using blast to mapping unmapped reads to the reference. Is it right?

The mapping is so slow(20 cores dealing with 20000000 reads). Does anyone knows any methods to do this faster?

Thanks!!!

[GenoMax]

Blast is not the right tool for doing this. You should switch back to an NGS data aligner.

Answer 1

[Philipp Bayer]

Looks like you're looking for a metagenomics tool - tools like Kraken 2 or Centrifuge are much faster for what you want to do

[Gjain]

Do you have a preference on one tool over another?

[Philipp Bayer]

Not really, no - Kraken2 is so new that I haven't seen any benchmarks yet. Kraken1 was slower but more accurate (IIRC)

[Gjain]

Thanks for the information ...I will try Kraken2

[Shaway]

Thank you very much! I will try these 2 tools.