Hi Elaine OK. Thank you for the help. I have one more question. So I'm building a C-terminal dimer model of human Apo A-I using 3R2P as a template model (a C-terminal truncated x-ray crystal structure of the protein). I loaded the biological assembly (a dimer) in ChimeraX and used the Sequence viewer to get the template-query alignment. But when I try to run Modeller, it gives me the following message: "Modeller cannot handle templates with multi-character chain IDs". Am I missing something here? I would be very grateful for your help. Thanks Tiglath ________________________________ From: Elaine Meng <meng@cgl.ucsf.edu> Sent: Sunday, October 25, 2020 10:13 AM To: Moradkhan, Tiglath A <tmoradkhan@csus.edu> Cc: chimerax-users@cgl.ucsf.edu <chimerax-users@cgl.ucsf.edu> Subject: Re: [chimerax-users] Using ChimeraX Modeller interface Hi Tiglath, This is mostly a question about Blast, since you didn't even get to the Modeller part yet. (1) Using Blast. To specify a sequence as Blast query you have to first open the sequence, and then use the blastprotein command (not the Blast dialog) to specify the sequence. We would like to improve the Blast dialog so that you can input a sequence (ticket #3014) but it is not yet possible, so you have to use the command instead. User Guide for blastprotein command: <http://rbvi.ucsf.edu/chimerax/docs/user/commands/blastprotein.html> ... in that page, if you click "sequence-spec" link in the usage line, it goes to another page explaining how to specify the sequence: <http://rbvi.ucsf.edu/chimerax/docs/user/commands/seqspec.html> To open the sequence, you could open a fasta file, OR you could fetch it automatically using Uniprot ID. You could even open a multiple sequence alignment and specify an individual sequence from it as explained in the link above. The following example commands fetch a sequence from Uniprot and then run blast on it: open uniprot:p00812 .... then there is a sequence window with title bar "Seqview [ID: p00812]" so that you know that the sequence ID to use in the "blastprotein" command is p00812 blastprotein p00812 (2) Looking at Blast results. You must choose a dimer structure as the template if you want to model your sequence as a dimer. Unfortunately that is not obvious from the Blast results. You may want to also go to the RCSB PDB website and look at the information for those specific structures there to see if they are dimers. I probably wouldn't open the structure from the Blast dialog, but instead if RCSB PDB website shows that the biological unit for that structure is a dimer, I would then open the biological unit directly with the "open" command, for example: open pdbe_bio:1wva maxAssemblies 1 (in my example it is a trimer) <http://rbvi.ucsf.edu/chimerax/docs/user/commands/open.html#fetch> Even if you don't use the Blast dialog to get the structure, you should use it to get the sequence alignment between the template and the query, by using the checkbox on the left of the specific hit and the "Show in sequence viewer" button near the top of the dialog. (3) When you finally have a dimer template structure open, then you get to the Modeller part (menu: Tools... Sequence... Modeller Comparative). It has an option "Make multichain model from multichain template" that is already turned on by default, to make your dimer model from the dimer template. <http://rbvi.ucsf.edu/chimerax/docs/user/tools/modeller.html> I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco

On Oct 24, 2020, at 4:34 PM, Moradkhan, Tiglath A <tmoradkhan@csus.edu> wrote:

Hi My name is Tiglath and I'm trying to use ChimerX Modeller interface to build a dimer model of a protein I'm currently studying. I first tried to run a BLASP search to find templates but it seems that in ChimeraX, this works differently than in Chimera. In the user guide it says that if the query is a sequence whose structure hasn't been determined, then it's not necessary to specify particular chains. However, when I try running the BLAST search it doesn't do anything. It seems that it is necessary to specify a chain number. But how is this possible if I don't know the template for the target sequence? Am I missing something? Also, I'm a little bit unsure how the Modeller interface works for building dimers.

I would be very glad for your help. Thank you Tiglath

Hi Elaine OK. That makes sense. Thank you so much! Thanks Tiglath ________________________________ From: Elaine Meng <meng@cgl.ucsf.edu> Sent: Monday, October 26, 2020 3:29 PM To: Moradkhan, Tiglath A <tmoradkhan@csus.edu> Cc: chimerax-users <chimerax-users@cgl.ucsf.edu> Subject: change chain ID Hi Tiglath, That message means exactly what it says: Modeller cannot handle templates with multi-character chain IDs. I see that when you use open pdbe_bio:3r2p ... then you get one chain A and one chain AA (two characters, = multi-character, so can't be used as Modeller template). This is shown in the Log when you open the structure. Not sure why PDBe chose to name the second chain as AA, but you can change it to a single character instead, for example to B with "setattr": setattr /AA chain chain_id B Sometimes (but not in this case) you might need two commands to change the chain ID instead of one, as explained in this recent post: <https://www.rbvi.ucsf.edu/pipermail/chimerax-users/2020-October/001565.html> I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco

On Oct 26, 2020, at 3:02 PM, Moradkhan, Tiglath A <tmoradkhan@csus.edu> wrote:

Hi Elaine OK. Thank you for the help. I have one more question. So I'm building a C-terminal dimer model of human Apo A-I using 3R2P as a template model (a C-terminal truncated x-ray crystal structure of the protein). I loaded the biological assembly (a dimer) in ChimeraX and used the Sequence viewer to get the template-query alignment. But when I try to run Modeller, it gives me the following message: "Modeller cannot handle templates with multi-character chain IDs". Am I missing something here? I would be very grateful for your help. Thanks Tiglath

Hi Ben, Yes I am aware that recent versions of Modeller can take mmCIF files as input, which allows for multi-character chain IDs. When I started writing the Modeller support in ChimeraX, ChimeraX had no ability to save mmCIF files, so PDB files were used. On my very long to-do list is revamping the Modeller support to use mmCIF files to get around the single-chain-ID limitation, but it isn't very high on the priority list since the rename-chain workaround exists. I also have little idea how much effort is involved, since I don't know how well ChimeraX's mmCIF output will correspond to what to what Modeller wants. For instance, the PDB code had to be revised to allow non-standard residues to be written as ATOM records instead of HETATM records if they were part of the sequence being modeled. Things like that are probably lurking around if the code switches to mmCIF. --Eric Eric Pettersen UCSF Computer Graphics Lab

On Oct 26, 2020, at 5:38 PM, Ben Webb <ben@salilab.org> wrote:

I don't want to muddy the waters too much here, but Modeller itself shouldn't have any issues with templates with multi-character chain IDs. But perhaps ChimeraX is giving Modeller PDB files (in which case you'd be limited by the PDB format to 1 character)? Modeller *should* be able to take mmCIF files as templates, which wouldn't be limited.

Ben

On 10/26/20 3:29 PM, Elaine Meng wrote:

...

Hi Tiglath, That message means exactly what it says: Modeller cannot handle templates with multi-character chain IDs. I see that when you use open pdbe_bio:3r2p ... then you get one chain A and one chain AA (two characters, = multi-character, so can't be used as Modeller template). This is shown in the Log when you open the structure. Not sure why PDBe chose to name the second chain as AA, but you can change it to a single character instead, for example to B with "setattr": setattr /AA chain chain_id B Sometimes (but not in this case) you might need two commands to change the chain ID instead of one, as explained in this recent post: <https://www.rbvi.ucsf.edu/pipermail/chimerax-users/2020-October/001565.html> I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco

...

On Oct 26, 2020, at 3:02 PM, Moradkhan, Tiglath A <tmoradkhan@csus.edu> wrote:

Hi Elaine OK. Thank you for the help. I have one more question. So I'm building a C-terminal dimer model of human Apo A-I using 3R2P as a template model (a C-terminal truncated x-ray crystal structure of the protein). I loaded the biological assembly (a dimer) in ChimeraX and used the Sequence viewer to get the template-query alignment. But when I try to run Modeller, it gives me the following message: "Modeller cannot handle templates with multi-character chain IDs". Am I missing something here? I would be very grateful for your help. Thanks Tiglath

---

ChimeraX-users mailing list ChimeraX-users@cgl.ucsf.edu Manage subscription: https://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users

-- ben@salilab.org https://salilab.org/~ben/ "It is a capital mistake to theorize before one has data." - Sir Arthur Conan Doyle _______________________________________________ ChimeraX-users mailing list ChimeraX-users@cgl.ucsf.edu Manage subscription: https://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users