[mira_talk] Re: Mira results much worse than newbler
- From: Robin Kramer <kodream@xxxxxxxxx>
- To: mira_talk@xxxxxxxxxxxxx
- Date: Fri, 15 Jul 2011 17:16:58 -0600
Maybe this is the only one but I had a bear of a time getting this one to
work, and I expected that it wasn't a bug but just how sff_extract worked.
The SRA sample SRR054580 demonstrates it though. Since it was in the SRA I
couldn't contact the sequence provider, evidently using minleft clip does
exactly that. However going back and trying it on other sequences it does
indeed seem to do exactly the same as sffinfo when it works.
/home/assembly/sw/linux-x86_64/sff_extract/0.2.8/bin/sff_extract
SRR054580.sff
Working on 'SRR054580.sff':
Converting 'SRR054580.sff' ... done.
Converted 906300 reads into 906300 sequences.
********************************************************************************
WARNING: weird sequences in file SRR054580.sff
After applying left clips, 513937 sequences (=57%) start with these bases:
A
This does not look sane.
Countermeasures you *probably* must take:
1) Make your sequence provider aware of that problem and ask whether this
can be
corrected in the SFF.
2) If you decide that this is not normal and your sequence provider does not
react, use the --min_left_clip of sff_extract.
(Probably '--min_left_clip=6' but you should cross-check that)
********************************************************************************
On Fri, Jul 15, 2011 at 3:26 PM, Bastien Chevreux <bach@xxxxxxxxxxxx> wrote:
> On Jul 15, 2011, at 22:05 , Robin Kramer wrote:
> > sff_extract must have been broken when I used it then, because all it did
> was mark the first 4 bases as clipped.
>
> Jose actually made a nice job with the extraction code there ... that part
> of the program is essentially unchanged since quite a number of years and
> has served me (and others) well. Nor do I recall any bug report in that
> area.
>
> B.
>
>
> --
> 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:
