[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

MIME types and Applied Biosystems chromatograms.



Dear Software Communauty support,

I am a member of the Debian-Med project, and would like to enquire about
two related questions:

 1) The definition of a MIME type for .ab1 files.
 2) The auto-detection of ABIF chromatograms.

In the Debian-Med project [1], our goal is to provide and integrate free
software for medical practice and life science research. We are
currently working on associating file types to the applications that can
open them, so that our users can have a desktop environment in which a
double-click will work without post-installation configuration.

[1] http://www.debian.org/devel/debian-med

The EMBOSS software suite provides `abiview' [2], a program to display
chromatograms in ABIF format. On desktop systems complying to the
FreeDesktop standards, applications are associated to MIME types [3],
and files are associated to MIME types independently [4].

[2] http://emboss.sourceforge.net/apps/release/5.0/emboss/apps/abiview.html
[3] http://standards.freedesktop.org/desktop-entry-spec/latest/
[4] http://standards.freedesktop.org/shared-mime-info-spec/latest/

ABIF-formatted chromatograms have not officially defined mime types.
Some labs use application/abi1 [5] or application/x-dna [6], but the
first violates the standards [7], and the second is using a very generic
name. According to the section 2.1.2 of the standard, we would like to
use a type like `application/vnd.appliedbiosystems.ab1' or something
similar. We would like to have your approval before using such MIME
types. Also, if you find the idea interesting, it may be useful to
properly register them.

[5] http://gatc.arl.arizona.edu/resources/faqs/sequencing/view.php
[6] http://bioinformatics.unc.edu/software/sequencher/seq_ABI_files_mac.htm
[7] http://www.ietf.org/rfc/rfc2048.txt

The second half of the file—application association is the
auto-detection of the file MIME type. This is done by two ways: from the
suffix and from the contents. For the suffix, we are lucky that .ab1 is
not used by other applications. For the contents however, we did not
manage to unequivocally detect chromatograms in ABIF format [8], because we
did not manage to find a simple way to differentiate them from the
ABIF-formatted fragment analysis files (.fsa) produced by your
instruments.

[8] http://www.appliedbiosystems.com/support/software_community/ABIF_File_Format.pdf

To detect the ABIF format, we detect the character chain `ABIF??tdir'
(ignoring the characters where the question marks are) at the head of
the file. Can you indicate us if there is a specific chain that
distinguishes the chromatograms from the fragments analysis files, at a
given offset in the ABIF-formatted files ?


This email is copied to our mailing-list, as a continuation of our
discussion on this subject [9]. But do not hesitate to answer only in
private if you prefer.

[9] http://lists.debian.org/debian-med/2008/01/msg00135.html


Best regards,

-- 
Charles Plessy
Debian-Med packaging team
http://www.debian.org/devel/debian-med
Wakō, Saitama, Japan


Reply to: