blast with unmapped human reads/ fastq

[priyatamapandey]

Hi Gabriel, I am trying to blast with unmapped fastq files of STAR aligner. I found you on GitHub. BLAST takes only fasta files. So, my question is, do I need to do assembly of these unmapped paired fastq files? Or I can just merge or cat the paired end fasta files using interleave fastq ?

Thank you, Priya

[GabrielHoffman]

There are two options for quantifying viruses from the unmapped reads

1) Use trinity to assemble the unmapped reads into contigs. These contigs are in FASTA format so you can BLAST them against your virus database. 2) Use STAR to create an index from the virus database. The align your unmapped reads to the virus database with STAR

The code here uses (1), but it is a few years old. We have used (2) more recently, and it should be faster

[priyatamapandey]

Hi Gabriel, Thank so much for your quick reply. The reason I want to do BLAST is the because I want to check my unmapped reads against viral, bacterial and fungal genomes.

Thank you