In <8lg8ll$5g8$1 at mercury.hgmp.mrc.ac.uk> Phillip San Miguel <pmiguel at purdue.edu> writes:
> How about the GAP4 package though? I saw some mention of Alf gels in
> their documentation. Maybe they have a good Alf basecaller that can
> assign quality values. Also, there are other commercial packages that
> display Alfx format chromats (Sequencher leaps to mind) but I don't
> think any of them assign quality scores. Does anyone know?
We don't have any base callers - we simply use whatever is in the trace file.
We do support Alf format data files, but I think recently they created a
compressed copy which we cannot read (as we don't know what format it is in,
and intuiting this from a compressed data stream is really hard). Why don't
people publish their file formats? Anyway, Alf should support SCF output.
The Staden Package makeSCF program (freely downloadable as part of io_lib)
does have some _very basic_ normalisation functions designed for tidying up
alf data, but I do not know whether this would aid phred at all.
You may also consider investigating ATQA by Wagner Associates. This assigns
confidence values without recalling the bases. Hence it may be possible to
keep the Alfx base caller but still get the best use of Gap4 by providing it
confidence values. I'm not sure what their training set was though.
James Bonfield (jkb at mrc-lmb.cam.ac.uk) Tel: 01223 402499 Fax: 01223 213556
Medical Research Council - Laboratory of Molecular Biology,
Hills Road, Cambridge, CB2 2QH, England.
Also see Staden Package WWW site at http://www.mrc-lmb.cam.ac.uk/pubseq/