[chimerax-users] display options for model with 2 chains

Elaine Meng meng at cgl.ucsf.edu
Tue Apr 5 09:28:01 PDT 2022 

Hi Mauricio,
You definitely can show specific model(s) like spaghetti or licorice without changing their secondary structure assignments -- just make the width and cross-section of all protein cartoon the same with "cartoon style".  E.g. like menu: Presets... Licorice/Ovals except that you could apply it to specific model numbers instead of everything.

Example command to do it for #1:

cartoon style #1 arrows false xsection round width 1 thick 1

Smoothing could be turned on/off with a separate command, e.g.:

cartoon #1 smooth 0
cartoon #1 smooth 1

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

> On Apr 5, 2022, at 9:09 AM, Mauricio Losilla via ChimeraX-users <chimerax-users at cgl.ucsf.edu> wrote:
> 
> Hi all,
> 
> Thank you so much for your input. You are all correct, the issue was with the beta-strand smoothening-- I don't think a new feature is required. 
> 
> To clarify, ChimeraX thinks the purple ribbon is just a coil because... I told it so, with setattr #1/t res ss_type 0
> and the beta-smoothening of the gray chain explains the issue I was having. 
> 
> I apologize if this was confusing. I did this because, when the 2 chains were displayed with helixes and strands, they looked blended with each other (see "chimerax_blended", attached). I don't like this look. By the way, I believe this cannot be changed by tweaking with cartoon smooth! 
> 
> I am probably showcasing my naivete, but deleting this chain's secondary structure was the only way I was able to show the protein only as a coil (without helixes and strands). Is there a better way to do this? 
> 
> After your feedback, I used the transparency, cartoon, and cartoon style commands to arrange a very satisfactory display ("new_result", attached).
> 
> Thank you all very much!
> 
> Regards,
> Mau
> 
> 
> Mauricio Losilla, PhD
> Integrative Biology; and Ecology, Evolution, & Behavior
> Michigan State University
> he, him, his
> 
> 
> On Tue, Apr 5, 2022 at 4:52 AM JAMES MICHAEL S1JJRUdFUiA= <jmkrieger at cnb.csic.es> wrote:
> Thanks Elaine. That's great. That's even better than pymol then where  
> it's a global setting.
> 
> Elaine Meng <meng at cgl.ucsf.edu> escribió:
> 
> > Hi James,
> > Yes, ChimeraX has an option to turn off the beta-strand smoothing  
> > for specific residues/chains, or the whole structure.  Example:
> >
> > open 1www
> > cartoon /Y smooth 0
> > cartoon smooth 0
> > cartoon smooth 1
> >
> > ...where the above shows doing it for chain Y only, then the whole  
> > structure, then changes back to the default smoothing of 1.0.  You  
> > can also use intermediate values in the 0-1 range.  Details in  
> > "cartoon" help:
> > <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/commands/cartoon.html__;!!HXCxUKc!h1lhfqX3hUVl90-jAZnkZtp9opCBKFVH6f0VoDOYDEkLnLgRyJZ6PUl8I0xTisc$ >
> >
> > Best,
> > Elaine
> > -----
> > Elaine C. Meng, Ph.D.
> > UCSF Chimera(X) team
> > Department of Pharmaceutical Chemistry
> > University of California, San Francisco
> >
> >
> >> On Apr 4, 2022, at 4:19 AM, JAMES MICHAEL S1JJRUdFUiA= via  
> >> ChimeraX-users <chimerax-users at cgl.ucsf.edu> wrote:
> >>
> >> Perhaps, you could turn off the beta strand flattening like in  
> >> pymol. That could be quite a nice feature in chimerax if it isn't  
> >> already there.
> >>
> >> I think it's perfectly understandable and expectable that if you  
> >> show the purple protein as spaghetti rather than a proper cartoon  
> >> with helices and strands then the Calpha atoms follow their direct  
> >> path (with perhaps a little smoothing) and the view does not  
> >> resemble a flattened beta strand representation.
> >>
> >> Best wishes
> >> James
> >>
> >> Tom Goddard via ChimeraX-users <chimerax-users at cgl.ucsf.edu> escribió:
> >>
> >>> Hi Mauricio,
> >>>
> >>>  The weaving of the purple tubular ribbon back and forth across  
> >>> the gray beta-strands is because ChimeraX smooths the beta-strand  
> >>> path so it does not go exactly through the C-alpha atoms.  Because  
> >>> when it goes exactly through the C-alpha atoms it looks like wavy  
> >>> bacon as you showed in your rainbow ribbon image, and most people  
> >>> like the smooth appearance better.  Your purple ribbon is just a  
> >>> tube shape which suggests ChimeraX thinks it is all coil secondary  
> >>> structure with no beta strands, so it does not smooth the it, and  
> >>> it goes on the wavy path through the C-alpha atoms.  I am not sure  
> >>> why ChimeraX thinks your purple model is all coil -- would need  
> >>> the data to know why.
> >>>
> >>>  If I-Tasser made the gray model from the purple and assuming it  
> >>> kept the backbone nearly in the same place, I am not sure what you  
> >>> are trying to show -- maybe just that the backbone did stay in the  
> >>> same place?  At any rate, you could either figure out how to make  
> >>> ChimeraX recognized that the purple template has beta strands so  
> >>> those get smoothed.  Or you could make ChimeraX not smooth the  
> >>> gray predicted structure strands by using the "smooth" option of  
> >>> the cartoon command.
> >>>
> >>>     https://urldefense.com/v3/__https://www.cgl.ucsf.edu/chimerax/docs/user/commands/cartoon.html__;!!HXCxUKc!h1lhfqX3hUVl90-jAZnkZtp9opCBKFVH6f0VoDOYDEkLnLgRyJZ6PUl8AeEsVj0$ 
> >>>
> >>>  Tom
> >>>
> >>>
> >>>> On Apr 1, 2022, at 5:43 PM, Elaine Meng via ChimeraX-users  
> >>>> <chimerax-users at cgl.ucsf.edu> wrote:
> >>>>
> >>>> Oops, typed the wrong number, sorry -- the ticket is #6527
> >>>>
> >>>> Elaine
> >>>>
> >>>>> On Apr 1, 2022, at 5:40 PM, Elaine Meng via ChimeraX-users  
> >>>>> <chimerax-users at cgl.ucsf.edu> wrote:
> >>>>>
> >>>>> For anybody who is interested, I made a feature request ticket #6257.
> >>>>>
> >>>>> However, now that I look carefully at your original image of the  
> >>>>> desired visual outcome (ideal_display.png) I see that it also is  
> >>>>> "threaded" in some places, e.g. where the rainbow ribbon is  
> >>>>> cyan. However, it is a lot less noticeable than in the ChimeraX  
> >>>>> image because the rainbow ribbon is paper thin, what we called  
> >>>>> "flat" ribbon in Chimera. ChimeraX does not have this ribbon  
> >>>>> style option, although you can approximate it by making the  
> >>>>> thickness of the ribbon very small instead of the default 0.4  
> >>>>> Angstroms.
> >>>>>
> >>>>> I added this as a comment on the ticket.
> >>>>> Best,
> >>>>> Elaine
> >>>>>
> >>>>>> On Apr 1, 2022, at 2:20 PM, Mauricio Losilla via ChimeraX-users  
> >>>>>> <chimerax-users at cgl.ucsf.edu> wrote:
> >>>>>>
> >>>>>> I understand. Thank you very much for your assistance!
> >>>>>>
> >>>>>>
> >>>>>> Mauricio Losilla, PhD
> >>>>>> Integrative Biology; and Ecology, Evolution, & Behavior
> >>>>>> Michigan State University
> >>>>>> he, him, his
> >>>>>>
> >>>>>>
> >>>>>> On Fri, Apr 1, 2022 at 10:06 AM Elaine Meng <meng at cgl.ucsf.edu> wrote:
> >>>>>> Hi Mauricio,
> >>>>>> Probably the "real" way to do it is to save two images, one of  
> >>>>>> each chain with the other one hidden, and then use a separate  
> >>>>>> image-editing app like Gimp or Photoshop to composite the two  
> >>>>>> images together.  I wonder if that is what the I-tasser folks  
> >>>>>> did.
> >>>>>>
> >>>>>> The only other idea I had seemed like cheating, to actually  
> >>>>>> Z-translate the purple model closer to the user.  Also it might  
> >>>>>> not look right, even if you change from perspective projection  
> >>>>>> to orthoscopic (command "camera ortho")
> >>>>>>
> >>>>>> I am not a programmer so I would not be the one to figure out  
> >>>>>> an algorithm or how feasible it would be to implement, i.e.  
> >>>>>> I'll create a feature request ticket and put you on the  
> >>>>>> notification list, but can't say anything about how it will be  
> >>>>>> prioritized.
> >>>>>> Best,
> >>>>>> Elaine
> >>>>>> -----
> >>>>>> Elaine C. Meng, Ph.D.
> >>>>>> UCSF Chimera(X) team
> >>>>>> Department of Pharmaceutical Chemistry
> >>>>>> University of California, San Francisco
> >>>>>>
> >>>>>>
> >>>>>>> On Apr 1, 2022, at 8:17 AM, Mauricio Losilla via  
> >>>>>>> ChimeraX-users <chimerax-users at cgl.ucsf.edu> wrote:
> >>>>>>>
> >>>>>>> Hi Elaine,
> >>>>>>>
> >>>>>>> I am sorry, I didn't explain myself well, I will try to do it  
> >>>>>>> here. I was able to use all your tips successfully: I split  
> >>>>>>> the chain in two models, and tweaked the thickness, width, and  
> >>>>>>> transparency at will. It worked great! I am attaching the  
> >>>>>>> result. I think it accomplishes my original goal fairly well:  
> >>>>>>> to visually represent the fit of the gray protein to the  
> >>>>>>> purple template.
> >>>>>>>
> >>>>>>> However, the purple chain still looks sewed into the gray  
> >>>>>>> chain. In the image, this is better seen in the horizontal  
> >>>>>>> beta sheets at the center. The ideal display I was after is  
> >>>>>>> that of the image with the rainbow-colored chain and the  
> >>>>>>> purple template I sent in my original email. There, the purple  
> >>>>>>> chain always looks "on top" of the rainbow chain. That is what  
> >>>>>>> I meant yesterday with "the ability to superimpose the view of  
> >>>>>>> one chain".
> >>>>>>>
> >>>>>>> In case it is helpful, the reference image and the PDB file  
> >>>>>>> came from I-TASSER.
> >>>>>>>
> >>>>>>> I am new to ChimeraX (and to 3D shapes of molecules in  
> >>>>>>> general), so I apologize if I am not making much sense.
> >>>>>>>
> >>>>>>> Thank you!
> >>>>>>>
> >>>>>>> <result.png>
> >>>>>>>
> >>>>>>>
> >>>>>>> Mauricio Losilla, PhD
> >>>>>>> Integrative Biology; and Ecology, Evolution, & Behavior
> >>>>>>> Michigan State University
> >>>>>>> he, him, his
> >>>>>>>
> >>>>>>>
> >>>>>>> On Thu, Mar 31, 2022 at 5:11 PM Elaine Meng <meng at cgl.ucsf.edu> wrote:
> >>>>>>> Hi Mauricio,
> >>>>>>> I don't understand "the ability to superimpose the view of one  
> >>>>>>> chain" -- can you explain it in more words?  What is it that  
> >>>>>>> you want, that cannot be done currently?  Maybe it is already  
> >>>>>>> possible but I didn't explain it because I didn't realize that  
> >>>>>>> was what you wanted.
> >>>>>>>
> >>>>>>> If you mean to make "cartoon style" work on just a chain  
> >>>>>>> instead of whole model, it may be unlikely since the settings  
> >>>>>>> it controls are per-model (not per-chain or per-residue).   
> >>>>>>> However, it is already easy to split chains into separate  
> >>>>>>> models with "split."
> >>>>>>>
> >>>>>>> Sorry if I am misunderstanding what you meant, however.
> >>>>>>> Elaine
> >>>>>>> -----
> >>>>>>> Elaine C. Meng, Ph.D.
> >>>>>>> UCSF Chimera(X) team
> >>>>>>> Department of Pharmaceutical Chemistry
> >>>>>>> University of California, San Francisco
> >>>>>>>
> >>>>>>>> On Mar 31, 2022, at 3:42 PM, Mauricio Losilla via  
> >>>>>>>> ChimeraX-users <chimerax-users at cgl.ucsf.edu> wrote:
> >>>>>>>>
> >>>>>>>> Hi Elaine,
> >>>>>>>>
> >>>>>>>> Thank you very much for your prompt reply. Your workarounds  
> >>>>>>>> were very helpful, I was able to tweak thickness and  
> >>>>>>>> transparency, and arrived at a satisfactory display  
> >>>>>>>> composition. Would you consider as a request feature the  
> >>>>>>>> ability to superimpose the view of one chain?
> >>>>>>>>
> >>>>>>>> Thank you!
> >>>>>>>> Mau
> >>>>>>>>
> >>>>>>>> Mauricio Losilla, PhD
> >>>>>>>> Integrative Biology; and Ecology, Evolution, & Behavior
> >>>>>>>> Michigan State University
> >>>>>>>> he, him, his
> >>>>>>>>
> >>>>>>>>
> >>>>>>>> On Thu, Mar 31, 2022 at 11:43 AM Elaine Meng  
> >>>>>>>> <meng at cgl.ucsf.edu> wrote:
> >>>>>>>> Hi Mauricio,
> >>>>>>>> Some ideas are to:
> >>>>>>>>
> >>>>>>>> - make the gray ribbon transparent
> >>>>>>>> and/or
> >>>>>>>> - make the purple one thicker
> >>>>>>>>
> >>>>>>>> I can't tell from your question whether they are in two  
> >>>>>>>> separate models or not.  Even if they're together in one PDB  
> >>>>>>>> file they can be defined as separate models (as is done for  
> >>>>>>>> NMR ensembles, e.g. PDB 1kfp).  If the purple one is model #1  
> >>>>>>>> and the gray one is #2, could be something like:
> >>>>>>>>
> >>>>>>>> cartoon style #1 width 1 thick 1 xsect round
> >>>>>>>> transparency #2 50 targ r
> >>>>>>>>
> >>>>>>>> ...of course, you could use different width/thickness and  
> >>>>>>>> transparency values.
> >>>>>>>>
> >>>>>>>> <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/commands/cartoon.html*style__;Iw!!HXCxUKc!hKsk6lXSWeI6B9WmfXcRKCWrq70659PcqAPq2FBwdZzjDjOH6myV44XapFhvyuc$  
> >>>>>>>> >
> >>>>>>>> <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/commands/transparency.html__;!!HXCxUKc!hKsk6lXSWeI6B9WmfXcRKCWrq70659PcqAPq2FBwdZzjDjOH6myV44Xa-fKcbds$>
> >>>>>>>>
> >>>>>>>> The cartoon style command works on whole models, so you would  
> >>>>>>>> need them to be in separate models for that approach.  If  
> >>>>>>>> your current PDB file does not have them in separate models,  
> >>>>>>>> you could either text-edit it to add MODEL and ENDMDL  
> >>>>>>>> records, or (probably easier) use the "split" command after  
> >>>>>>>> opening it.
> >>>>>>>> <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/commands/split.html__;!!HXCxUKc!hKsk6lXSWeI6B9WmfXcRKCWrq70659PcqAPq2FBwdZzjDjOH6myV44XaI58NOwM$  
> >>>>>>>> >
> >>>>>>>>
> >>>>>>>> I hope this helps,
> >>>>>>>> Elaine
> >>>>>>>> -----
> >>>>>>>> Elaine C. Meng, Ph.D.
> >>>>>>>> UCSF Chimera(X) team
> >>>>>>>> Department of Pharmaceutical Chemistry
> >>>>>>>> University of California, San Francisco
> >>>>>>>>
> >>>>>>>>> On Mar 31, 2022, at 10:02 AM, Mauricio Losilla via  
> >>>>>>>>> ChimeraX-users <chimerax-users at cgl.ucsf.edu> wrote:
> >>>>>>>>>
> >>>>>>>>> Hi,
> >>>>>>>>>
> >>>>>>>>> I am new to ChimeraX, thank you for making this great  
> >>>>>>>>> software available.
> >>>>>>>>>
> >>>>>>>>> I loaded a pdb file that has 2 amino acid chains aligned  
> >>>>>>>>> (chimerax.png). These chains are a model (gray) and template  
> >>>>>>>>> (purple).
> >>>>>>>>>
> >>>>>>>>> As you can see, the two chains look weaved or blended. I  
> >>>>>>>>> would much prefer to display the uninterrupted template  
> >>>>>>>>> around to model, to visually represent their fit. Is there a  
> >>>>>>>>> way to do this?
> >>>>>>>>>
> >>>>>>>>> I am attaching a second image (ideal_display.png, generated  
> >>>>>>>>> elsewhere) with the visual outcome I am after (the purple  
> >>>>>>>>> chain is the template, the rainbow-colored chain is the  
> >>>>>>>>> model).
> >>>>>>>>>
> >>>>>>>>> Thank you
> >>>>>>>>> Mauricio Losilla, PhD
> >>>>>>>>> Integrative Biology; and Ecology, Evolution, & Behavior
> >>>>>>>>> Michigan State University
> >>>>>>>>> he, him, his
> 
> 
> 
> <chimerax_blended.png><new_result.png>_______________________________________________
> ChimeraX-users mailing list
> ChimeraX-users at cgl.ucsf.edu
> Manage subscription:
> https://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users

More information about the ChimeraX-users mailing list