IUBio GIL .. BIOSCI/Bionet News .. Biosequences .. Software .. FTP

Staden Package 2001.0 beta release

jkb at mrc-lmb.cam.ac.uk jkb at mrc-lmb.cam.ac.uk
Thu Aug 16 03:33:21 EST 2001


In <cupton-1508011339460001 at hector.bioc.uvic.ca> cupton at uvic.ca (Chris Upton) writes:
> Hi,  
> Don't take offense :-) (I've used the Staden package for years and have
> loved it!)  but is it safe to have my grad student move her precious
> sequencing project onto this beta version?

No offense - it's a good question.

We spent several months (literally) hammering away at bug testing and bug
fixing. During that process we identified several bugs which were in the old
release too, although clearly they didn't impact on many people as we haven't
received these in bug reports. There are quite a lot of known bugs which we
haven't yet fixed in this beta release, but these are basically cosmetic
things (and once again quite a few are in the 2000.0 release anyway). All
serious bugs were fixed.

That said, we can never test the software ourselves as well as the combined
efforts of our users, so once it's been "used" for a while (and we've worked
through the remainder of the buglist) we will make the official release. There 
have been a few low-level, potentially dangerous, changes. This includes
several speed ups (directed assembly is MUCH faster - useful when importing
phrap data, for example) and a quicker method of reopening databases. This
stores the memory copy of the "database free tree" at the end of the .aux
file, so that subsequent database opens do not need to recompute this. Time
stamps and CRCs are used to ensure that any database crash or similar failure
will invalidate the cached tree, which then gives the "rebuilding free tree"
(paraphrased) message. Both of these points have been tested of course, but if 
there are bugs there then they have the potential for database corruptions.

My advice would be for people to take regular database backups, accompanied by 
using 'copy_db' (the same as Copy Database in garbage collection mode) to
ensure data integrity. However that advice applies equally well to any version
of the program. You can never have enough backups!

James
--
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/




More information about the Staden mailing list

Send comments to us at archive@iubioarchive.bio.net