[mira_talk] Re: caf2phdball
- From: Sven Klages <sir.svencelot@xxxxxxxxxxxxxx>
- To: mira_talk@xxxxxxxxxxxxx
- Date: Tue, 27 Oct 2009 12:45:01 +0100
Hi Lionel,
2009/10/26 Lionel Guy <guy.lionel@xxxxxxxxx>
> Hi Sven,
>
> Sorry for the avalanche of questions. Brace for impact ;)
>
>
> On 25 Oct 2009, at 18:43 , Sven Klages wrote:
>
> Yes,you do need an actual file, as this should be opened when you request
>> it in consed.
>>
>
> I'm a bit confused... what kind of file would you create? A zero-size file?
> Can you have "chromatograms" for 454 reads as well?
>
Yes, and you still need them (as of v19 of consed) to edit single reads or
to change consensus in a certain way.
This is done on-the-fly by sff2scf which reads the sff file and creates a
fake chromatogram (Consed reads the corresponding line in the phd.ball).
>
> I think I remember there is a option in consed option that allows to
> convert sff to phd files, but does it actually create chromatogram files?
> And how could we deal with read editing in mira? Use a converted phd file
> but the original chromatogram?
AFAIK there some prog called sff2scfAndPhd for consed v19, I never used it
:-)
And, well, yes, the original chromatogram may slightly differ from the
edited sequence, but the phd files gets edited (a new version is created).
So it's not a real problem ... that is one more disadvantage of
consed/ACE/phd. Sigh.
>
> You need to have some approximate positions for the chromatogram to be
>> opened correctly.
>> If you leave all pos at 0, all "peaks" in the chromat are squashed
>> together and no (normal) editing is possible.
>>
>
> Makes sense...
>
>
> We do keep our chromatograms in (indexed) tarballs. No filesystem problem,
>> very fast :-)
>>
>
> I would be interested to know how you create them. And then, do you
> directly link to the tarball in the phd file?
We'll put this "project" public soon on sourceforge or code.google. I'll let
you know then.
One phd entry corresponds to one chromatogram. You need to put the logic of
this in an external script, have a look at:
consed.alwaysRunProgramToGetChromats: true
consed.programToRunToGetChromats: /package/consed/bin/findFileInArchive.pl
!consed.programToRunToGetChromatsOf454Reads: sff2scf
consed.uncompressedChromatDirectory: /tmp
>
>
> Do the numbers (15, 19) in the calculation of $peakpos come from empirical
>> data?
>>
>> Yes, I shamelessly copied it from sff2phdball.c (from consed v17 AFAIR).
>>
>
> Well, I guess it should work with consed then ;)
>
> Hope you survived the avalanche... thanks for your help anyway!
>
Hope I could answer parts of the avalanche ;-)
cheers,
Sven
> Cheers,
>
> Lionel
>
>
> --
> 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:
