Bioperl Blast module / New Releases available

Georg Fuellen vsns-bcd-admcoord@TechFak.Uni-Bielefeld.DE
Mon, 8 Jun 1998 19:36:30 +0000 (GMT)

Below please find two recent announcements.
Also, please allow me to point you to the Webarchive
of our discussion list, to which some feedback to
the Blast module announcement has just been posted, see

Bioperl: Bioperl Blast module now available!

Steve A. Chervitz (
Fri, 5 Jun 1998 18:48:43 -0700 (PDT) 

Hello Bioperl members:

The first official public release of my package is 
ready for downloading and installation:

It includes a fair bit of functionality and a portion of my developing
Bioperl module framework.

Please check it out and let me know what you think.


Steve Chervitz

Bioperl: version 1.009 available

Georg Fuellen (fuellen@dali.Mathematik.Uni-Bielefeld.DE)
Fri, 29 May 1998 14:58:00 +0000 (GMT) version 1.009 has become available at

The following improvements have been made; versions 1.007 and 1.008
were never annnounced publicly. Notably, many more formats are now
supported (via ReadSeq).

Version 1.007 on 15 Mar 1998.
* Added functions that return indices instead of actual sequence data:
var_inds(), invar_inds(), gap_free_inds(), unknown_free_inds(),
special_free_inds() and no_allgap_inds() return the indices of
(in)variable, gap-free, unknown-free, gap+unknown-free and non-gap-only
columns; they are the counterparts of var_sites(), etc.
* Added remove_gaps() to return ungapped (original) sequence data
* Added equal_no_gaps() to check for equality ignoring gaps
* Added equalize_lengths() for padding a sequence bag with gaps
such that all rows have equal length (current procedure is slow)
* The default id is now ``_'' (was: ``No_Id_Given'')
* Internal functions _rowbounds() and _colbounds() now return indices
from the user's perspective, but WITHOUT substracting offset. Instead,
all functions subtract offset after applying _rowbounds()/_colbounds()

Version 1.008 on 13 May 1998.
* Added readseq conversion support, making it possible to read and write
the following formats: MSF, Paup, PIR/Codata, ASN.1. (ASN.1 was not
parsed successfully: readseq seems to be unable to read in its own ASN.1
output). Technically, readseq is now used to parse files that have been
processed as ``raw'' before; now ``raw'' format is recognized using the
expression /^[A-Z_0-9$_GAP_SYMBOL$_UNKN_SYMBOL\s]+$/im, i.e. the file
may only have alphanumerical characters, gap and unknown-symbol, and
whitespace. If commata, etc, are detected, readseq is used for parsing.
Readseq itself seems to be unable to detect ``raw'' format in some cases,
causing weird results.
* Added Clustal support for parsing; still to be tested and documented.

Version 1.009 on 25 May 1998.
* The module can now be ``built'' in the standard way (perl Makefile.PL, etc).

Please allow me to ask one question to the `experts': is there a way to 
convince MakeMaker that the module is not in the top-level directory of
the distribution, but instead in the ``appropriate'' directory, according
to hierarchy ? E.g. MakeMaker does not accept in a directory
``Bio'', and this causes major frustration (and duplicate copies). 


P.S. I'm looking forward to meeting many of you at ISMB'98 !
(( since I'll also attend a symposium on large-scale phylogenies,, I may not arrive before Monday
early afternoon.))

best wishes,
Georg Fuellen,,
Univ. Bielefeld, Research Group in Practical Comp Science, Fax +49 521 106 6411, BioMOO:GeorgF, Fon .... 106 2903