Re: [ccp4bb]: guiSIDE

[Anastassis Perrakis <perrakis@nki.nl>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

Hi David -

Thanx for giving  us a chance to clarify the following (RTFM would not have
helped,
we shamely admit that the manual is not F-for-Fine - but we are working on it):

When guiSIDE is used for side chains rebuilding its very picky on the PDB:

It only accepts the 20 standard aminoacids.

It will NOT accept OXT's, ligands, WATs or DUMs. Sad but true.
It i ssupoed to work with multiple chains, but might be a bit unpredictable.

What I would suggest to any users is to use an editor to strip their file from
anything
non-aa, run guiSIDE and then append to the guiSIDE output file what they got rid
of.

Since guiSIDE is currently being rewritten using Clipper and mmdb, I really did
not have
time to make the I/O a bit more sensible. If people find that really annoying
(mail me,
NOT ccp4) I will try and get some time to get it to at least not crash with
non-std aa.
Must be trivial - sorry again. The new guiSIDE will take at least 9-12 more
months to
be really completed, but will be far more powerful - I hope.

Double conformations are NOT implemented but should be easy to code - should I
take your
question as an offer to do the beta-testing ? I can send you a beta-version
(and to anybody else inetrested) in a few weeks, after I back from holiday.

Also, to any guiSIDE users, can you please report any really erroneous
placements
(i.e. Trp chi2  90 deg wrong ?) to me ? I would also appreciate any bad/good
reports for
side chain building in lowish resolution maps (2.7 and below). No time to test
everything.

Tassos