I'm updating some software that reads the chromatograms that
are emitted by ABI software so that it can handle the new
stuff emitted by the 3100 and 3700 sequencers. I've figured
out what most of the new flags are (at least those that
don't have zero values) but I'm still a little hazy on CpEP
and Rate.
CpEP is an immediate character field with a value is 1 (control A).
Anybody know what it is? My best guess at this point is
that it might correspond to "channels available", which is
also 1 as shown by the ABI software. (Why they stored that
in a character field instead of INT2 or INT4 I have no idea.)
Rate is more complex. Its "FLAG" structure has:
FLAG: Rate
instance 1
datatype 1024
datasize 12
nrecords 1
nbytes 12
value(hex) 48040000
spare(hex) 0
This is the first time I've seen datatype 1024. Since
nbytes > 4 the data must be indirect (won't fit into value,
so value becomes a pointer), but when I dump the area in the
file around that location the meaning of the data isn't
obvious.
Anybody know what these two fields are?
As long as I'm writing - is it just me, or does anybody else
see the offset for the GELP field, as stored by the Mac
based software for the 373XL, as being high by 1? SVER on
these files shows "2.0" and "Version 3.0". (I know that at
least Version 1.0.2 didn't have this problem.) There are a
bunch of indirect datatype=18 data fields in those
chromatograms, but only GELP seems to offset _onto_ the
first character of the string, rather than onto the "count"
character which precedes it. When I run my software on this
set of chromatograms it pulls out:
>GELP 1 acko: 373XL Data Collection:Runs:Run Folder-5/16/00 5.02 PM:<EOT>x<XFE> (etc, more garbage follows)
instead of
>GELP 1 Wacko: 373XL Data Collection:Runs:Run Folder-5/16/00 5.02 PM
But every other text field is fine!
Thanks,
David Mathog
mathog at seqaxp.bio.caltech.edu
Manager, sequence analysis facility, biology division, Caltech
