[mira_talk] Re: Filtering out crappy sequence - try my sff2phd
- From: Markiyan Samborskyy <ms587@xxxxxxxxxxxxxxxxxx>
- To: Bastien Chevreux <mira_talk@xxxxxxxxxxxxx>
- Date: Thu, 5 Apr 2012 21:10:37 +0100
Dear Bastien,
Thanks you for reply, hope you'll enjoy it.
BC> Whoa. Nice thing, albeit two questions immediately arise:
BC> - was there a reason not to use / extend sff_extract?
It was born in the October 2008, when I was trying to use the 454 paired
end reads in the phrap assembly (since 20 Jun 2007 - till Oct 2008 I
was using and improoving sff2scf, but had problems implemeting linker
seq detection in the pure c), so tried to make a perl version of it.
It's primary purpose was to integrate the sff data properly in the phredPhrap
projetcs. Then during the evolution: filtering, MID
detection/selection, import by the reads inclusion/exclusion
list, fastq format output were added...
Anyway - I will have a go at sff_extract at some stage...
BC> - why Q+64? Even Illumina saw the light and moved to the standard Q+33
Historical reasons, (since 2007), first chr encoding was Q+64, also
fastX toolkit understands the Q+64 encoding only.
Also some programs don't like the Q+33 encoded character strings,
while they work fine with Q+64 ones.
Anyway - will add Q+33 fastq encoding option to the next version.
Also can add writing of the traceinfo.xml file for mira.
PS:
If anybody is intrested, I have a phredPhrap project integration tool
phredMira, which does basically the job of phredPhrap, but using MIRA.
It converts phd_dir to mira's input (in ../mira_dir)
sff+ab1+scf files ->phd_dir->*.fasta->*.fasta.screen->
mira's *_in.fasta + qual+xml (splitting based on the readstechnology)
...run MIRA.....
generate output and fix the MIRA's ace bugs (read timestamps+etc),
place the resulting ace in the edit_dir.
Now it can be edited by the consed. The edits in the reads if
assembly are saved and will be picked up by the next phredMira run on the
same thing. (just read the latest vesrsions of the *.phd.2,3,4...N
files). So one can actualy FINISH THINGS, without manual
multifasta files creation/merging.
Have Nice Easter!
--
Best regards,
Markiyan mailto:ms587@xxxxxxxxxxxxxxxxxx
--
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:
