Hello,
I am downloading some BGIseq files from SRA (fastq-dump with --split-files) and I cannot identify the paired reads. Two files are downloaded, but it appears ALL reads are labelled 1 to n (across both files) with half having a '/1' and half having a '/2'. So, if there are 50,000 F and 50,000 R reads they will be numbered 1 to 100,000, half (seemingly randomly) with /1 and half with /2.
Further, its not like the first half of the file goes into my X_1.fastq file and the second half into X_2.fastq. As a simplified example I might end up with: File 1 contains the following read numbers: 1-10000, 40001-70000, 90000-100000 and File 2 contains: 10001- 40000, 70001-90000 (in reality the numbers are nowhere near round numbers!)
Both are the same length (same NUMBER of reads). But no apparent pairing info.
Can I assume that the first read in file 1 is the pair of the first read in file 2? This is the head of the two files:
File_1.fastq @80384878/1 TCAATCATTAACTTAAATTTTAGTTGAATTGGTTCCTTAATATGGTATCAGAGCAATGTGACAAAACGATCACATGTTTAAATCTCACCTACCCTCAAAG +SRR7121710.80384878 80384878 length=100 FEFFFFFFFFFFFFFFFEFFFFFFFCFFFFFFFFAFFFF:FFFAFFFFFF@FEFFFFDFGFFEFFFFFFFFFFFGFGFGF%@AFCFFBFF@FDE6=FGD@
File_2.fastq @1/2 AAACCTTAAAAATATGTTGCATATTCTAACATGTCCTTTCACATGAGAGCTTTTTGGAGAATTGTGGTCTCTTCCTCATACATGGTTCACAATATTCCAT +SRR7121710.1 1 length=100 G=FFFEFFF0FDFFFFFEEFFBFFFFFFFFFFCFFFF6FFDFFFF?F@EFGFFFFFA:G?FFFAFGDF?GEF>F*D)D:;F88F7FCG@(D5EFGF>F8F
Or can I assume something is wrong with the initial upload. I've repeated the DL several times FYI and emailed the corresponding author to no avail.
Unless someone else has seen this I'm inclined to not trust that read 80384878/1 is a pair of 1/2 just in case...!
Thanks
