mapping hundreds of RADseq fastqs to ref genome.

[wbsimey]

I would like to map hundreds of fastq.gz single-end RADseq files to a single reference genome.

But, bwa mem is making a map file (.sam) for every fastq file. So I end up with hundreds of individual maps. I want a single map with all fastq files mapped to the reference.
I have tried hundreds of failed commands, here is one failed example:
bwa mem -t 16 ../reference.Arrow.fasta ../fastqs/*.F.fq.gz > RADs_mapped.sam

Any ideas what I am doing wrong? Or what is wrong with my expectation of a single .sam file with all sequences mapped to my reference?

[cmbetts]

It's treating each of your fastq files as an independent sample, and making a sam file for each, which is what most people generally want to do. If you don't care about keeping them separate, you can always merge the fastq files with cat prior to alignment or merge the many individual results using samtools. If you care about keeping track of which alignment came from which fastq, it's probably going to be easier to keep them separate and deal with there being many files programatically

[wbsimey]

thank you cmbetts!

I am running this bash script, which seems to be working and producing many .bam files

for f in $(ls /Projects/RADseq/fastqs/*.gz)
do
        bwa mem -t 16 reference.Arrow.fasta $f |
        samtools view -b |
        samtools sort --threads 8 > $f.bam
done

[GenoMax]

Any reason you are not using an established pipeline like STACKS.

[wbsimey]

Any reason you are not using an established pipeline like STACKS.

I am using Stacks, they recommend using bwa for reference based analyses. But, I could not get their recommended pipeline to work, yet.