how to interpret fastq

[juhang_62]

Below are some lines I excerpt from a fastq file. It's generated by illumnia. It's a little different from normal fastq format. How to inpterpret it? I want to find information about sequence type, flowlane, paired-end, multiplexed. I think the numbers tell all the information but I do not know what they stand for.


@SRR307074.65513 HWI-EAS88_0007:6:1:19833:18832 length=49
NTTTCGTGTCGCAATAACAATAAGAAAGAAAGAAAAAGAAAACCAGAGA
+SRR307074.65513 HWI-EAS88_0007:6:1:19833:18832 length=49
#:;99=====EEEEEBEEEEEEEEEBEEEEEEEEEEEEEEEEEEEEEEB
@SRR307074.65514 HWI-EAS88_0007:6:1:19833:10713 length=49
NACAAATGTTAAGTTCTTCAGTTTCCTTCAAAATTGGGTTTTCTTGAGA
+SRR307074.65514 HWI-EAS88_0007:6:1:19833:10713 length=49
#972188===EEEEEBBBBBEEEEEEEEEEEEEEEEEEEEEEEEEEEEE
@SRR307074.65515 HWI-EAS88_0007:6:2:1069:8480 length=49
CCAACCTAGAACCATAATTTANTCTCTTACANNANGAAAAAACAAGACN
+SRR307074.65515 HWI-EAS88_0007:6:2:1069:8480 length=49
GGDEGFGGGDGGFGDFEEGFD#DD=B=;350##1#C>=4>BB#######
@SRR307074.65516 HWI-EAS88_0007:6:2:1070:6870 length=49
GAAATTTTATATGGAACTCATNTATAAGAAANNANGAGATGTAAAGGCN
+SRR307074.65516 HWI-EAS88_0007:6:2:1070:6870 length=49

Thank you!

[fkrueger]

I don't know what you mean with 'it's a little different from normal fastq format', to me it looks pretty 'normal'.

@SRR307074.65513 should be the study number
HWI-EAS88_0007 should be instrument and run number
6:1:19833:18832 should be lane, tile, x-coordinate, y-coordinate (of the sequence cluster)
length=49 ...

Quality values in SRA data are encoded in Sanger format (Phred +33), and I can't see any indications that the data was paired-end or multiplexed.

[mgogol]

If it's paired end, you'll have two different files, one with reads from the first end and one with reads from the second.

If it's multiplexed and has already been split up, the file names should have the index in them. If it's multiplexed and hasn't been split, you can use fastx_barcode_splitter.

http://hannonlab.cshl.edu/fastx_toolkit/commandline.html#fastx_barcode_splitter_usage