On Mar 16, 2012, at 7:56 AM, Karel Berka wrote:

Dear all,

I have a strange problem with Chimera 1.6.1 Modeller (9.10) interface
- when I tried to prepare a model based on template 3DDQ (CDK2
kinase), it showed that Modeller locally is running: 0%, but it stayed
on this value. When I checked Task list, I have found that although it
is still "running", the result was "progress: finished" with no
protein model produced.

I dwelwed more and I later found that Modeller script was prepared,
but Modeller was unable to start due to the misspelling between
sequence used for the template (probably from SEQRES) and sequence
calculated from template structure. When I deleted residues that
Modeller was unable to find in the structure, I was able to build the
models.

Hi Karel,
I can reproduce this problem, so I will be opening a ticket in the Chimera bug-tracking database with you on the recipient list so you'll know when it gets fixed.  The problems seems limited to 3DDQ since other structures I tried worked fine, though I don't doubt there would be other structures that could produce them problem that share whatever feature of 3DDQ that is causing this.

As for the web homology modelling interface - I was also unable to get
model from Opal server, probably for the same reason.

Well, changes I made on the web server have screwed up the Modeller web service for the last two days (fixed now), so it's not certain that the web server problem is the same -- though I would expect the same failure for 3DDQ.

What to repair in Chimera to get Modeller interface running? because
otherwise it seems like homology modelling is not working at all,
which is not true.

I'll share whatever remedy I come up with via the Trac ticket.

--Eric

                        Eric Pettersen

                        UCSF Computer Graphics Lab

                        http://www.cgl.ucsf.edu