[mira_talk] Re: HiSeq data problem with Mira
- From: "Bastien Chevreux" <bach@xxxxxxxxxxxx>
- To: mira_talk@xxxxxxxxxxxxx
- Date: Thu, 3 Nov 2011 13:07:14 +0100 (MET)
From: Mcnally, Alan<alan.mcnally@xxxxxxxxx>
> I have sent Bastian the first 10,000 lines of my FastQ input
> file.................hopefully he can solve for me...........its getting
> very frustrating
The file you sent me is everything but a valid FASTQ. Here's an excerpt, the
first few lines (I've Xed out the bases):
------------------------ snip --------------------------
@HWI-ST300:133:B0908ABXX:3:1101:1242:2117 1:Y:0:ATCACG
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
@HWI-ST300:133:B0908ABXX:3:1101:1242:2117 2:Y:0:ATCACG
NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN
+
####################################################################################################
+
####################################################################################################
------------------------ snip --------------------------
That file is so plain wrong and not FASTQ, I do not know where to start:
1) the first read (HWI-ST300:133:B0908ABXX:3:1101:1242:2117 1:Y:0:ATCACG) has
no quality
2) the second read (HWI-ST300:133:B0908ABXX:3:1101:1242:2117 2:Y:0:ATCACG) has
two quality lines
May I *strongly* suggest you leave out any script (I have this ominous
shuffleseq.pl in mind you wrote about) and simply cat the first n million lines
of your data sets together? That's how I do it and it works like a charm :-)
Furthermore, those are CASAVA 1.8 data sets where Illumina has changed the read
naming scheme. I.e., you will notice that the first two reads have actually the
very same name. For this atrocity the developers at Illumina should be nailed,
crucified, tarred, feathered, shot into space and subsequently dissected by
slimy green purple aliens with 18 tentacles and a single eyeball (in this
order, and preferably alive during the whole procedure).
Anyway, while the development version of MIRA now knows these things, you need
to rename your reads for all public versions out there. More specifically, you
need to pull the first character of the comment to the read name, separating it
with a slash.
I.e., a line reading
@HWI-ST300:133:B0908ABXX:3:1101:1242:2117 1:Y:0:ATCACG
must be changed to
@HWI-ST300:133:B0908ABXX:3:1101:1242:2117/1 1:Y:0:ATCACG
and a line with
@HWI-ST300:133:B0908ABXX:3:1101:1242:2117 2:Y:0:ATCACG
to
@HWI-ST300:133:B0908ABXX:3:1101:1242:2117/2 2:Y:0:ATCACG
Best,
Bastien
--
You have received this mail because you are subscribed to the mira_talk mailing
list. For information on how to subscribe or unsubscribe, please visit
http://www.chevreux.org/mira_mailinglists.html
Other related posts:
