How to blast a paired end fastq file?

Question

[chillchik29]

Hello everyone, I have a paired end fastq file and I know that BLAST+ in command line, accepts fasta format. But I don't know how does it work for a paired end fastq file (I mean in two different files R1 and R2). Do I have to interleave them and then convert them to fasta? Or should I concatenate them and convert them to fasta? Or should I treat them as separate files and then convert them to fasta?

Thank you, any help would be appreciated.

[Ram]

Why do you wish to BLAST read sequences? Plus, why do you want to BLAST both pairs and all reads?

[chillchik29]

I already selected these group of reads from all the metagenome sequences by aligning them with bbmap, to the phage_nt db. So before assembling them, I wanted to have a first approach of what I had in the sample by aligning them with BLAST and then export it to MEGAN.

[GenoMax]

You could save some effort and only convert one end to fasta (reformat.sh in=R1.fq.gz out=R1.fa) and then BLAST. PE reads are not going to give you much additional information that MEGAN can use.

[chillchik29]

Thank you! I'm going to try that out :)

Answer 1

[chen]

You can use fastp to merge it first, then blast the merged reads

See the intro here: https://github.com/OpenGene/fastp#merge-paired-end-reads

Answer 2

[JC]

Use Magic-Blast

[GenoMax]

While this would work to do the alignments it would produce SAM format output. I am not sure if MEGAN can use that.

[JC]

magic-blast also reports aligment in tabular format, which is similar to blast outfmt 6

[GenoMax]

Ah I missed that. It may need some manipulation (since it is not quite outfmt 6). I guess OP can let us know.