Thank you very much Elaine !

Actually I've tried to follow this but modeller did not work in that case.

Briefly, to be clear I need to model only one chain (A) in my complex but considering the presence of other chains in order to omit any sterical clashes between newly built residues and other chains.

So having all complex loaded in chimera-X window I loaded sequence for the chain A

seq chain /A

and then as it was mentioned associated it to the whole complex

sequence associate #1

Then I run modeller from the GUI which indicated a problem in the alignment file

Please check your alignment file header to be sure you correctly specified
the starting and ending residue numbers and chains. The alignment sequence
must match that from the atom file exactly.

Another possibility is that some residues in the atom file are missing,
perhaps because they could not be resolved experimentally. (Note that Modeller
reads only the ATOM and HETATM records in PDB, NOT the SEQRES records.)
In this case, simply replace the section of your alignment corresponding
to these missing residues with gaps.
read_te_288W> Protein not accepted: 1 8hti_1

Il giorno lun 26 ago 2024 alle ore 23:42 Elaine Meng <meng@cgl.ucsf.edu> ha scritto:

Hello,
How to use the Modeller tools in ChimeraX for multichain complexes is explained in the help, see the second gray box:
<https://rbvi.ucsf.edu/chimerax/docs/user/tools/modelloops.html>

To summarize, you need a separate sequence window for each chain, with the template for that chain associated with that sequence in that window.

The command would be rather complicated, so I won't try to show it here. Instead you should try using the Model Loops GUI to do this (in menu under Tools... Structure Prediction). Using the GUI tool will show the corresponding command(s) in the Log.

I hope this helps,
Elaine
-----
Elaine C. Meng, Ph.D.
UCSF Chimera(X) team
Resource for Biocomputing, Visualization, and Informatics
Department of Pharmaceutical Chemistry
University of California, San Francisco

> On Aug 26, 2024, at 2:17 PM, Enrico Martinez via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:
>
> Dear Chimera-X users!
>
> I need to predict missed loop fragments in a part of a multi-chain complex corresponded to GPCR bound to heteromeric G-protein (pdb id 8HTI). In this structure the chain A (a subunit of the G-protein) contain missed loop fragments. So the following commands were executed in command line to reconstruct the entire structure of this subunit:
>
> open 8hti
> seq chain /A
> ui tool show "Model Loops"
> modeller refine 1/A:1:internal-missing numModels 5 fast false adjacentFlexible 1 protocol standard
>
> which very slowly produces 5 models of the chain A.
>
> The problem was that the modeller considered the chain A in isolation from the rest of the structure and predicted loops overlapped with other chains of the G protein. Would it be possible to reconstruct the missing fragments in the chain A considering also the presence of other chains to build the full complex ?
>
> Many thanks in advance
>
> Enrico