Processing SRA data.

[TCLamnidis]

Example entry: https://www.ebi.ac.uk/ena/browser/view/PRJNA883375

When downloading data from the ENA that was originally uploaded to the SRA, it seems 3 FastQ files are made available.

I had assumed these are {merged, R1, R2}, like with ENA-uploaded data, but further inspections suggests otherwise. SRR21678398.fastq.gz contained only ~15M reads out of the total ~580M. The rest are apparently in SRR21678398_{1,2}.fastq.gz, which are presumably paired.

Need to look deeper into it, but it seems like I will need to pick up all three files for processing with eager in cases where the data was uploaded to the SRA. Or perhaps skip the third file, as it might be unpaired reads?

[TCLamnidis]

Read counts in the fastqs:

File	read count
SRR21678398_1.fastq.gz	569985643
SRR21678398_2.fastq.gz	569985643
SRR21678398.fastq.gz	14311741

It seems the read counts in the _{1,2} files matches, so they are indeed paired reads.