The prefetch utility downloads .sra files into the dbGaP project directory.
After downloading, the SRA Toolkit provides other tools to convert this format into bioinformatics formats, including fast(q/a) and SAM.
Two files are downloaded for each sample: an sra file and sra.vbdcache file. Example:
-rw-r--r-- 1 cczysz cczysz 3172424864 Oct 5 2014 SRR1090791.sra
-rw-rw-r-- 1 cczysz cczysz 144198176 Aug 30 18:56 SRR1090791.sra.vdbcache
While a file is downloading, a .lock file exists.
-rw-rw-r-- 1 cczysz cczysz 0 Aug 30 19:18 SRR1490246.sra.vdbcache.lock
The file name of these files can be given as an argument to the file conversion utilities:
./fastq-dump SRR1090791
Get all SRR* ids in the current directory and group into groups of 8.
For each group of 8, test to make sure none of the ids are being downloaded by checking for .lock file.
Extract each file into two gzipped fastq files (paired-end):
fastq-dump -I --split-files --gzip <SRR>
This outputs two files named <SRR>_1.fastq.gz | <SRR>_2.fastq.gz
Rename each file to the GTEx ID using the manifest file.
grep SRR658319 manifest_40651_08-30-2016_s.csv
reads/SRP012682/SRS389106/SRX221542/SRR658319/SRR658319.sra, No, SRR658319, SRS389106, GTEX-VJWN-0726-SM-3GIJ8, 915071, SAMN01887103
mv SRR658319.fastq.1.gz GTEX_VJWN_0726_SM_3GIJ8_R1.fastq.gz