Automating Batch Printing of electropherograms from ABI files-How to do it?

David Mathog mathog at seqaxp.bio.caltech.edu
Tue Aug 29 14:52:51 EST 2000


In article <s9a12adb.037 at EM.AGR.CA>, chapmanp at EM.AGR.CA ("Patrick Chapman") writes:
>We have a 377 sequencer supported by a Macintosh running MacOS 8.1 using =
>ABIPrism Sequence Analysis software 3.4.
>
>Presently we use the Sequence Analysis software to print electropherograms =
>to a network printer.
>My main goal is to capture to Postscript files to reduce demands on the =
>Mac and to be able to modify the files prior to printing
>

Our method is currently:

1.  sequencing facility generates chromatograms and transfers them to VMS machine
    via FTP.
2.  VMS machine runs a script which does:
  a.  eliminate bad files (some may not contain trace data!)
  b.  run PHRED on each trace to call bases and qual scores
  c.  run ABIPRINTOUT on the trace and phred output to generate trace 
        printouts more or less equivalent to what the ABI software makes,
        but with qual values shown.
  d.  print/delete the resulting postscript files through a color printer

PHRED was easy to port to VMS and in any case, the way we use it in this 
application is completely compatible with the Unix versions.

ABIPRINTOUT is entirely home grown and has been evolving for 5 years. It is
at this time none too portable since it makes implicit use of byte order
information, is written in Fortran 77 and the graphics go through the GCG
graphics library.  (I estimate about 8 hours work to port it to some
other platform and graphics library - contact me by email if you want to
try.) Anyway, the output from: 

 $ abiprintout/infile=CRB_S577N1_CB4_H03_15.ab1/phd=CRB_S577N1_CB4_H03_15.PHD -
   /pnt=1300/font=0/note="mathog"/page=1/default

can be found at:

  http://seqaxp.bio.caltech.edu/pub/PICKUP/example_abiprintout.ps

The weak part in the chain at present is the HP5Color printer, which is 
quite slow - it can take a couple of hours for a day's sequencing runs to
print out. 

Anyway, the load on the Macs & NT machines at the sequencing facility is 
negligible - they only have to FTP the data over and then they're
completely out of the loop.  The load on the VMS machine is pretty light as
well.  All the processing takes just a couple of minutes, runs at a low
priority, and the print jobs in the queue don't represent any kind of CPU
load.  (At peak usage the temporary postscript files use about 70 Mb of
disk space.) 

Regards,

David Mathog
mathog at seqaxp.bio.caltech.edu
Manager, sequence analysis facility, biology division, Caltech 







More information about the Bio-soft mailing list