April 2021 - ChimeraX-users - RBVI Mailing Lists

Re: [chimerax-users] ChimeraX reading cube files
by Elaine Meng 10 Apr '21

10 Apr '21

> On Apr 10, 2021, at 7:54 AM, Carles Serrat Jurado <carles.serrat-jurado(a)upc.edu> wrote: > > Dear Elaine, > Thank you very much for your help. These instructions will certainly be very important for my work. > > There is still something that is not clear to me. My cube files, produced by the NWChem package, include information about the differential density. This appears by default when I open one of these files with the VESTA visualization software (https://jp-minerals.org/vesta/en/). Here I attach the 10 cube files I sent to you just converted to tif images after opening them with VESTA. As you can see, it is not the position of the atoms what evolves, as you already observed, but the differential charge density. > > I don’t know why ChimeraX does not produce the differential densities. If it would do it, then following your kind indications I could make now a movie to see that evolution (I have thousands of cube files to see a quite long time evolution - 5 femtoseconds, actually :-), which is what I am looking for. > > Again, I sincerely thank you for your answers and indications. I understand that probably ChimeraX is not able to read this information from my cube files. > > Best regards, > Carles > > PS: Please no worries about the data I am sending, it is not private so to say, these are tests I am doing at the moment. > PS2: I subscribed to the ChimeraX mailing list, but I don’t manage yet to answer there and attach files, please be free to share my mails however. Dear Carles, From your first message, I mistakenly thought you only wanted to see the atomic coordinates, so I ignored the density part. The instructions I sent before were how to make an atomic trajectory. If the atoms don't change position, then you can ignore most of those long instructions and just make one atomic model and keep showing it at the same time you show the changing density. For example, just open the attached PDB file, carles1.pdb. By default, ChimeraX does show the density from the cube files... that is the blob you see as soon as you open the file. If you mean that you want to see the densities as a series that changes over time, you just need to use the "open" command "vseries" option to tell ChimeraX that it is a volume series. For example, you previously sent a folder containing 10 files named density_subgs.0000030000.cube .... density_subgs.0000039000.cube (numbering increments of 1000). In ChimeraX, the following command opens the 10 files as a single volume series (time series): open [folder location]/density*cube vseries true Now you have a slider to play the series, or you can use the "vseries" command. Screen shot below. <https://rbvi.ucsf.edu/chimerax/docs/user/commands/vseries.html> More about atomic trajectories and volume series: <https://rbvi.ucsf.edu/chimerax/docs/user/trajectories.html> The blob is probably bigger than what you are used to seeing in Vesta. This is just the contour level. To change contour level, you can move the vertical bar on the volume viewer (density map) histogram that automatically appears when you open the data in ChimeraX. <https://rbvi.ucsf.edu/chimerax/docs/user/tools/volumeviewer.html> Your cube files are signed data, so there are two vertical bars, one for the positive contour level and one for the negative contour level. You can drag either bar, and change the color by then clicking on the small color square above the histogram. In the attached screenshot I made the negative one greenish and the positive one magenta. I only see a very small change in these densities as I play the 10 frames. 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 10, 2021, at 3:10 AM, Elaine Meng <meng(a)cgl.ucsf.edu> wrote: >> >> Hi Carles, >> Attached is the PDB file from step #3 that you can open as a trajectory as described in step #4. >> Best, >> Elaine >> >> <carles-atomic.pdb> >> >> >> >>> On Apr 9, 2021, at 6:07 PM, Elaine Meng <meng(a)cgl.ucsf.edu> wrote: >>> >>> Dear Carles, >>> It took me a couple of days to try that recipe script on your data, sorry about the slow reply. Now I realize there are a few extra steps to convert your data to a trajectory (e.g. with slider to play as movie), so I thought I should share the details with the list: >>> >>> (1) In ChimeraX, I opened your 10 cube files, and then I used the python file in the ChimeraX recipe to create 10 atomic models. >>> <https://rbvi.github.io/chimerax-recipes/gaussian_cube_atoms/gcatoms.html> >>> >>> (2) Then I saved the 10 atomic models as a single PDB file (using ChimeraX menu: File... Save), File type PDB, relative to the first atomic model. Doesn't matter which one you choose as the "relative to" model, however. >>> >>> (3) Then I was going to open that single PDB file as a trajectory. However, there is a bug in the saving that currently prevents it. The bug puts "END" after each atomic structure in the file. So I first needed to text-edit the PDB file to remove all of the "END" lines (but not "ENDMDL"). >>> >>> **I will send you the resulting PDB file in a separate message in case it is private data.** >>> >>> (4) In ChimeraX, I closed all the other models and then opened the new edited multi-model PDB file with the "coordsets" option to make it into a trajectory, for example, command: >>> >>> open carles-atomic.pdb coordsets true >>> >>> However, 10 frames of the trajectory all look exactly the same to me, but far as I know that is a correct conversion of the data from the 10 cube files you sent me. >>> >>> When the bug is fixed (#4487 <https://plato.cgl.ucsf.edu/trac/ChimeraX/ticket/4487>), step #3 would not be necessary. >>> >>> 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 7, 2021, at 9:33 AM, Anthony James Schaefer <tony.schaefer(a)uga.edu> wrote: >>>> >>>> ChimeraX does actually store the atom info in the cube file. Someone asked about this a while ago, and it was mentioned that there's some limitation where the stuff for opening volume data can't load atomic structures or something. I can't find the thread, but Tom Goddard wrote a script to load those atoms into ChimeraX: https://rbvi.github.io/chimerax-recipes/gaussian_cube_atoms/gcatoms.html >>>> >>>>> >>>>>> On Apr 7, 2021, at 7:25 AM, Carles Serrat Jurado <carles.serrat-jurado(a)upc.edu> wrote: >>>>>> >>>>>> Dear Elaine, >>>>>> >>>>>> I am writing because I have problems to plot a cube file with ChimeraX. I would greatly appreciate your assistance. Here I send the only figure I get when I open the cube file with ChimeraX. Also I attach what I would more expect, which is obtained with the software VESTA. >>>>>> >>>>>> My ultimate goal is to make a movie with using different frames. I also send a zipped file with the frames, so that you can see the kind of cube files I am working on. >>>>>> >>>>>> Thank you very much in advance for your support. >>>>>> >>>>>> Best regards, >>>>>> >>>>>> Carles >>>>>> >>>>>> <ChimeraX.png> >>>>>> >>>>>> <VESTA.png> >>>>>> <cube_files.zip> >>>>>> —————————-————————— >>>>>> Prof. Carles Serrat (PhD) >>>>>> Department of Physics >>>>>> UPC - Polytechnic University of Catalonia >>>>>> Colom 11, 08222 - Terrassa >>>>>> Phone: +34 93 739 8137 >>>>>> Mobile: +34 696 738 728 >>>>>> https://donll.upc.edu >>>>>> ——————————————————- >>>>> >>>>> >>>>> _______________________________________________ >>>>> ChimeraX-users mailing list >>>>> ChimeraX-users(a)cgl.ucsf.edu >>>>> Manage subscription: >>>>> https://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users >>>>> >>>> >>>> >>>> _______________________________________________ >>>> ChimeraX-users mailing list >>>> ChimeraX-users(a)cgl.ucsf.edu >>>> Manage subscription: >>>> https://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users >>>> _______________________________________________ >>>> ChimeraX-users mailing list >>>> ChimeraX-users(a)cgl.ucsf.edu >>>> Manage subscription: >>>> https://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users >>> >>> >>> _______________________________________________ >>> ChimeraX-users mailing list >>> ChimeraX-users(a)cgl.ucsf.edu >>> Manage subscription: >>> https://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users >>> >> > > —————————-————————— > Prof. Carles Serrat (PhD) > Department of Physics > UPC - Polytechnic University of Catalonia > Colom 11, 08222 - Terrassa > Phone: +34 93 739 8137 > Mobile: +34 696 738 728 > https://donll.upc.edu > ——————————————————- > > > > > > > > > > > > > > > > >

2 1

ChimeraX reading cube files
by Elaine Meng 09 Apr '21

09 Apr '21

Hi Carles, We recommend using the chimerax-users(a)cgl.ucsf.edu address (CC'd here) for questions about ChimeraX, unless you have attached data that needs to stay private. ChimeraX reads cube format but only as maps (e.g. density maps, electron localization functions, ...) with values on a grid. It does not read atomic coordinates from the cube format. This is a list of the atomic formats ChimeraX can read -- can you output any of those atomic formats instead of cube? (pdb, mol2, mol, sdf) <https://rbvi.ucsf.edu/chimerax/docs/user/commands/open.html#formats> <https://rbvi.ucsf.edu/chimerax/docs/user/commands/open.html#atomic> You can see that "cube" is in the list for volume (map) formats, but not as an atomic format. <https://rbvi.ucsf.edu/chimerax/docs/user/commands/open.html#volume> 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 7, 2021, at 7:25 AM, Carles Serrat Jurado <carles.serrat-jurado(a)upc.edu> wrote: > > Dear Elaine, > > I am writing because I have problems to plot a cube file with ChimeraX. I would greatly appreciate your assistance. Here I send the only figure I get when I open the cube file with ChimeraX. Also I attach what I would more expect, which is obtained with the software VESTA. > > My ultimate goal is to make a movie with using different frames. I also send a zipped file with the frames, so that you can see the kind of cube files I am working on. > > Thank you very much in advance for your support. > > Best regards, > > Carles > > <ChimeraX.png> > > <VESTA.png> > <cube_files.zip> > —————————-————————— > Prof. Carles Serrat (PhD) > Department of Physics > UPC - Polytechnic University of Catalonia > Colom 11, 08222 - Terrassa > Phone: +34 93 739 8137 > Mobile: +34 696 738 728 > https://donll.upc.edu > ——————————————————-

5 5

Surface caps only colored interactively
by Daniel Asarnow 09 Apr '21

09 Apr '21

Hi, I'm trying to generate a figure using several surfaces colored by map values. If I clip a surface, and then color it, the caps have the correct colors. However, when I do the same thing non-interactively, the caps are uncolored. Am I doing something wrong? view front.cut color sample #1.1 map #1.2 palette cyanmaroon range 2.25,7.0 #surface dust #1.1 size 100 save front_cut.png pixelSize 0.1 transparentBackground true Best, -da

2 5

chimerax python module - LigParGen tutorial
by mdigennaro＠uliege.be 07 Apr '21

07 Apr '21

Dear ChimeraX users, I am trying to follow this tutorial to get the proper residues: http://zarbi.chem.yale.edu/ligpargen/gmx_tutorial.html I have found chimera (0.4.7) as pip module, but it's written in python2 and therefore no longer maintained. Is there a chimerax module I can import? Pychimera is also available and could install it with python2, but it also show compatibility problems with python3 Is DockPrep a separate module? I have found this thread on the chimera forum which doesn't really answer the question: http://plato.cgl.ucsf.edu/pipermail/chimera-users/2014-July/010115.html Generally, how do I get unknown/custom residues (from LigParGen in this case) for Gromacs input files. Thank you for your support Marco ------------------- Marco Di Gennaro University of Liege

3 2

Use symmetry information in Isolde
by Sean Connell 06 Apr '21

06 Apr '21

Hello I have a naive question about using Isolde to build a model into a cryoEM map generated with symmetry (ie. D4). Basically I don't understand how to rebuild a chain and have the changes reflected in the symmetry related chains when using a cryoEM map. Is it possible to use, for example, a NCS file (like symmetry_from_map.ncs_spec) generated by phenix in order to get Isolde to build simultaneously the NCS related chains? Or is there another way build symmetry related chains into a symmetrical cryoEM map? Thanks for any tips you have.

2 2

An issue about chimeraX
by 祗俊生 02 Apr '21

02 Apr '21

Dear expert： I used "chimeraX 1.1 for windows 10 64-bit" to make animation. As can be seen in the figure provided in the attachment, the labels of hydrogen bonding distance overlapped when I played the animation. However, when I used "Tools-Structure Analysis-Contacts" to show the labels of "Contacts", these labels did not exhibit the above-mentioned problem when I played the animation. I wonder how to solve the problem of the label overlap regarding hydrogen bonds because I want to make a movie to show all of the hydrogen bonds. Moreover, I want to download "chimeraX 1.0 for windows 10 64-bit", but I can't find the download address of the old version. Many thanks!

2 1

Re: [chimerax-users] Displaying silhouette around specific visual elements
by Tom Goddard 31 Mar '21

31 Mar '21

Hi Gustavo, Glad you noticed the smoothness of the surface. I made it smoother than the default. The surface came from the molmap command which computes a density map from the atomic model. That command uses a density map grid spacing of 1/3 the requested resolution by default, but I specified an extra option that used a grid spacing 1/8 of the resolution: open 2gbp molmap #1 8 gridSpacing 1 Also if you are doing this with a big molecule (100,000 atoms) the density map created may be very large and may be shown subsampled so the surface looks coarser. Make sure to set the "step" to 1 in the Volume Viewer panel so that it uses the full resolution of the computed map instead of step 2 which would use only every other grid point. One more pro tip about that image composition example. It only worked nicely because the atomic structure was completely behind the surface, so I could layer the surface on top. If part of the atomic model was behind and part in front (for example with a cryoEM map surface), then the layering would not work well -- the composite image would have the surface in front everywhere. Elaine's selection highlight approach can handle that situation with correct occlusions but the edging is not as nice as silhouettes. Tom > On Mar 31, 2021, at 1:00 PM, de Miranda Seabra, Gustavo <seabra(a)cop.ufl.edu> wrote: > > Hi Tom, > > This looks beautiful! Just one tangential question: How do you get such a smooth surface? Can you share the exact commands you gave? > > Thanks! > > -- > Gustavo Seabra > Assistant Scientist > Center for Natural Products, Drug Discovery and Development (CNPD3) > Department of Medicinal Chemistry > College of Pharmacy > University of Florida, Gainesville, FL, USA > From: ChimeraX-users <chimerax-users-bounces(a)cgl.ucsf.edu <mailto:chimerax-users-bounces@cgl.ucsf.edu>> on behalf of Tom Goddard <goddard(a)sonic.net <mailto:goddard@sonic.net>> > Sent: Tuesday, March 30, 2021 8:55 PM > To: Andre LB Ambrosio <andre(a)ifsc.usp.br <mailto:andre@ifsc.usp.br>> > Cc: ChimeraX Users Help <chimerax-users(a)cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu>> > Subject: Re: [chimerax-users] Displaying silhouette around specific visual elements > > [External Email] > Hi Andre, > > If you are willing to go to a little trouble you can get silhouettes on an atomic model beneath a transparent surface that has no silhouettes. You first save an image of the atomic model with silhouettes with no surface, then save an image of the transparent surface with no atomic model and no silhouettes, then overlay the second image on the first in a photo editor like Gimp. Below is Elaine's example done this way. The command to save the surface image with transparency for overlaying is "save transpsurf.png transparentBackground true". > > Tom > > Overlay of two images. > > > Atomic model image > > > Transparent surface image > > >> On Mar 30, 2021, at 3:11 PM, Andre LB Ambrosio <andre(a)ifsc.usp.br <mailto:andre@ifsc.usp.br>> wrote: >> >> Dear Elaine, that is excellent! It is the visual effect I was looking for indeed. >> Thank you very much for this and thank you Tom for the very detailed response and for the feature request ticket. >> Andre. >> >> Em ter., 30 de mar. de 2021 às 15:59, Elaine Meng <meng(a)cgl.ucsf.edu <mailto:meng@cgl.ucsf.edu>> escreveu: >> Hi Andre, >> Sometimes you can approximately "fake" this effect by using the selection outline. It is not the same, however, as silhouettes; it will only go around the outside perimeter of the selected item(s). For example, if I have a molecule inside a map surface, I can make the surface transparent, select the atomic structure, and make the selection outline color black etc. as in these commands, resulting in the appearance in the attached image. >> >> open 2gbp >> molmap protein 8 >> trans 70 >> view >> set bgColor white >> select @* >> graphics selection color black >> >> 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 25, 2021, at 8:23 PM, Tom Goddard <goddard(a)sonic.net <mailto:goddard@sonic.net>> wrote: >>> >>> Hi Andre, >>> >>> It is not currently possible to display silhouettes on some models and not others. >>> >>> Several people have wanted this but I think it is complex to achieve. The silhouettes show boundaries where the depth in the entire scene changes abruptly. In order to have per-model silhouettes the ChimeraX GPU shader programs would need to compute an additional depth buffer separately for each model. Also the silhouette edges are not currently drawn at a specific depth. If per-model silhouettes were supported a stick model with silhouettes beneath a transparent surface needs to draw the silhouettes under the transparent surface so knowing the depth of the silhouette edge becomes important. When not using silhouettes the shader programs need to not compute the extra per-model depth buffers otherwise it would slow down rendering even when not being used. Adding these additional complexities to already complex graphics rendering shader programs have so far made this feature out of reach for our one graphics rendering developer (me). >>> >>> I made a feature request for per-model silhouette edges and perhaps in the future we will find a simpler way to implement it that makes it feasible. >>> >>> https://www.rbvi.ucsf.edu/trac/ChimeraX/ticket/4420 <https://urldefense.proofpoint.com/v2/url?u=https-3A__www.rbvi.ucsf.edu_trac…> >>> >>> Tom >>> >>>> On Mar 25, 2021, at 10:09 PM, Andre LB Ambrosio <andre(a)ifsc.usp.br <mailto:andre@ifsc.usp.br>> wrote: >>>> >>>> Dear all, >>>> >>>> I wonder if it's possible to display silhouettes selectively in the same figure; for instance, displaying silhouettes only on structure representation (around sticks, spheres, cartoons and so on), but not around the maps or volumes that surround them. >>>> Many thanks in advance, >>>> >>>> Andre >>>> _ >> >> >> -- >> Andre LB Ambrosio >> Associate Professor, IFSC/USP - Brazil >> www.ifsc.usp.br/alba <https://urldefense.proofpoint.com/v2/url?u=http-3A__www.ifsc.usp.br_alba&d=…> >> _______________________________________________ >> ChimeraX-users mailing list >> ChimeraX-users(a)cgl.ucsf.edu <mailto:ChimeraX-users@cgl.ucsf.edu> >> Manage subscription: >> https://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users <https://urldefense.proofpoint.com/v2/url?u=https-3A__plato.cgl.ucsf.edu_mai…>

2 1

ChimeraX-users April 2021