- ChimeraX-users - RBVI Mailing Lists

Questions about focal length of camera in ChimeraX
by Bin Shi 24 Mar '21

24 Mar '21

Thanks very much for your software so that I can specify the camera position and then take a snapshot. However, I can't find the information on the focal length of the virtual camera. I am using mono (default) – monoscopic with perspective (apparent size decreasing with distance from the viewer), and I didn't see any details about the camera intrinsics. Regards, Bin ----------------------------------------------------------------------------------------------- Department of Materials Science and Engineering, University of Toronto Email: binmse.shi(a)mail.utoronto.ca

3 4

Re: [chimerax-users] VR Support
by Tom Goddard 24 Mar '21

24 Mar '21

Hi Gökhan, Your Nvidia RTX 3090 should give the best performance possible with current GPUs. There was a shortage of high-end graphics card in 2020 and supply continues to far outpace demand with popular cards priced 2 to 3 times their suggested retail price. For instance your RTX 3090 has a suggested price of $1500 but currently is priced over $3000. This may not improve until 2022 according to news reports. This may be the biggest problem in getting a new molecular visualization computer setup for VR. https://www.theverge.com/2021/3/23/22345891/nvidia-amd-rtx-gpus-price-scalp… <https://www.theverge.com/2021/3/23/22345891/nvidia-amd-rtx-gpus-price-scalp…> I have not tried the HP Reverb G2 virtual reality headset. I know people were excited about the 2160 by 2160 pixels per eye resolution -- for comparison a Valve Index headset has 1440 by 1600 pixels. But reports said the hand-controller tracking with the Reverb G2 was below current VR standards. ChimeraX and any SteamVR application will work with any hand controllers. I have used ChimeraX VR with other Windows Mixed Reality controllers with a Samsung Odyssey headset with no problems. I've also used it with various other controllers with no problem: Vive Pro, Vive, Oculus Rift S, Oculus Rift, Oculus Quest. ChimeraX won't take advantage of the many extra inputs on Valve Index hand controllers. Nanome is a VR molecular visualization application funded by venture capital -- so they of course will need to charge plenty of money once that VC money runs out. I expect they will mostly be aiming their features at pharmaceutical company needs. I don't know if Nanome has any atomic model refinement capabilities. Tom > On Mar 23, 2021, at 12:11 AM, <tolung.bio(a)gmail.com> <tolung.bio(a)gmail.com> wrote: > > Hi Tom, > > Thanks for the prompt response. > > My responses are below. > > Cheers, > > Gökhan > > > > From: Tom Goddard <goddard(a)sonic.net <mailto:goddard@sonic.net>> > Sent: Tuesday, 23 March 2021 5:49 PM > To: tolung.bio(a)gmail.com <mailto:tolung.bio@gmail.com> > Cc: Tristan Croll <tic20(a)cam.ac.uk <mailto:tic20@cam.ac.uk>>; ChimeraX Users Help <chimerax-users(a)cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu>>; Simon Brown <simonb(a)uow.edu.au <mailto:simonb@uow.edu.au>>; James Bouwer <jbouwer(a)uow.edu.au <mailto:jbouwer@uow.edu.au>> > Subject: Re: [chimerax-users] VR Support > > Hi Gökhan, > > ISOLDE works in VR in a minimal way -- you can tug on atoms. Here's a little video showing that in ChimeraX 1.1. > > https://www.rbvi.ucsf.edu/chimerax/data/isolde-vr-mar2021/isolde_vr.mp4 <https://www.rbvi.ucsf.edu/chimerax/data/isolde-vr-mar2021/isolde_vr.mp4> > > I turned off depth cue and clipping with commands "light depthcue off" and "clip off" otherwise it clips to a slab which keeps moving as I move my head. I click a VR button on the ISOLDE panel tug atoms mouse mode icon to assign the VR button. I wouldn't say this makes ISOLDE usable in VR. For instance I don't have a way to move the roaming contour mesh region which normally stays in the middle of the screen. But it is enough to play around a bit. I was using a GTX 1080 Ti graphics card. Both the VR and ISOLDE molecular dynamics are running on the graphics card so this puts heavy demand on the card. For simulating just a region around a residue it was reasonably smooth. > > Thanks for the tips to get it to work with ISOLDE. I have a 3090 to play with (which recently replaced my trusty 1080) that should have significantly better performance than the 1080 Ti. Let’s see how it goes… > > I would like to make ISOLDE work well in VR. I think it is the perfect VR application where you really benefit from the best 3D perception possible and the 6-degree-of-freedom hand controllers for moving atoms around is much better suited than a mouse. > > ABSOLUTELY, and I cannot wait for that day that I’ve been waiting for since I started working on structures, to come… > > VR resolution has not been a major issue for use in molecular visualization. We have been using Vive Pro headsets that have adequate resolution for a few years now. A Valve Index headset is probably the best choice for molecular visualization right now. People also use Quest 2 with link cable. > > I have an HP Reverb G2 that recently replated my original HTC Vive. I am quite disappointed with it so far, and was wondering whether I should’ve gone for an Index. However, Index is not sold in Australia (and therefore would have required having one shipped from the US, which means no warranty here and expensive shipping). That was one of the reasons for going for the G2, and I have already used that warranty, since the first unit they sent died after minutes of first use! > > Do you think ChimeraX+ISOLDE should work with the WMR controllers that come with the G2, or is the current code specific to the Vive controllers? > > The big problem is it would take many developer years of effort to have decent user interfaces in VR. The furthest along on this for molecular visualization is Nanome -- you should check out their molecular VR. We continued to develop ChimeraX VR including multi-person VR. > > I have used Nanome’s one of the earliest versions, but it looks like they’re locking more and more of its features for free users based on my most recent try. ☹ Does it work as well as ChimeraX+ISOLDE in refining models? > > Tom > > > > >> On Mar 22, 2021, at 9:18 PM, <tolung.bio(a)gmail.com <mailto:tolung.bio@gmail.com>> <tolung.bio(a)gmail.com <mailto:tolung.bio@gmail.com>> wrote: >> >> Hello Tristan and the Chimera team, >> >> I have been using ChimeraX in VR: fantastic! >> >> However, ISOLDE was not working in VR, unfortunately. >> >> Is there a fix/workaround currently available to get ISOLDE running in VR? >> >> Is there ongoing work on making ISOLDE VR compatible? >> >> Also: 1) are there any advantages of using a Quadro card over a consumer-line card for ChimeraX VR? and 2) is there a recommended minimum system spec for getting high-enough frame rates in VR with large models? We are planning to put together a new system soon for molecular visualization and modelling, hence the questions. >> >> Now that there are VR headsets with high-enough resolution (such as the HP Reverb G2) that allow reading smaller text and finer details in VR, I am hoping that the molecular visualization and modelling software will start providing built-in easy-to-use support for these devices. I applaud ChimeraX team for leading in this space (no pun intended 😊). >> >> Best, >> >> Gökhan >> >> _______________________________________________ >> 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://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users> >

2 2

Re: [chimerax-users] movie in ChimeraX
by Elaine Meng 24 Mar '21

24 Mar '21

> On Mar 23, 2021, at 5:04 AM, Ruchika Bajaj <phd.ruchika(a)gmail.com> wrote: > > Hi Elaine > I am trying to figure out two things. Could you please help me with this? I would appreciate your help. > 1. I have two structures. The first one doesn't have a ligand and the second one has two ligands. I am trying to make a movie with these steps. 1) first structure, 2) two ligands are coming or flying in the picture, 3) then the 2nd structure with two ligands. However, when I am trying to make the morph between the first and second structure, it doesn't show any ligand. I did check the following link, where it discusses that both structures should have ligands to include it in the morph. So, my question is how should I include the ligand ? Could you please explain me with command usage? Thanks !! > <http://rbvi.ucsf.edu/chimerax/docs/user/commands/morph.html#pairing> > 2. I heard that frames/sec rate could be increased to smoothen steps in movies. I checked the maxFrameRate command. However, I don't know the usage. Could you please help me with that ? > Thanks a lot for your help !! Your prompt response would help me to proceed further. > Regards > Ruchika Hi Ruchika, Please send questions to chimerax-users(a)cgl.ucsf.edu (if you are using ChimeraX) or chimera-users(a)cgl.ucsf.edu (if using Chimera) rather than my individual address, or else they may not get answered. (1) as explained here <https://rbvi.ucsf.edu/chimerax/docs/user/commands/morph.html#pairing> morph only contains the atoms that are the same in both start and end structures. You can show ligands but because they are not included in the morph model, you have to display the ligand(s) from the original model (or an additional copy of the original model) along with the protein in the morph model. Moving them to different positions is even more complicated than showing/hiding. There is a Chimera tutorial that covers this, and what you would do with ChimeraX is similar except that ChimeraX "view" is used to save and restore positions instead of the Chimera commands "savepos" and "reset": <https://www.rbvi.ucsf.edu/chimera/data/tutorials/movies08/moviemaking.html> (2) if you mean you want more steps in the morph (more gradual change) then you would use the "frames" option in the morph command. Probably that is what you want. <https://rbvi.ucsf.edu/chimerax/docs/user/commands/morph.html#top> However, if you really mean you wanted movie framerate, then it is the "framerate" option of the "movie encode" command. <https://rbvi.ucsf.edu/chimerax/docs/user/commands/movie.html#encode> I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco

2 3

VR Support
by tolung.bio＠gmail.com 22 Mar '21

22 Mar '21

Hello Tristan and the Chimera team, I have been using ChimeraX in VR: fantastic! However, ISOLDE was not working in VR, unfortunately. Is there a fix/workaround currently available to get ISOLDE running in VR? Is there ongoing work on making ISOLDE VR compatible? Also: 1) are there any advantages of using a Quadro card over a consumer-line card for ChimeraX VR? and 2) is there a recommended minimum system spec for getting high-enough frame rates in VR with large models? We are planning to put together a new system soon for molecular visualization and modelling, hence the questions. Now that there are VR headsets with high-enough resolution (such as the HP Reverb G2) that allow reading smaller text and finer details in VR, I am hoping that the molecular visualization and modelling software will start providing built-in easy-to-use support for these devices. I applaud ChimeraX team for leading in this space (no pun intended 😊). Best, Gökhan

2 1

Move model to specific coordinates of reference model
by Vorländer,Matthias Kopano 19 Mar '21

19 Mar '21

Hi there, I am trying to place a model (volume) at specific coordinates of a second volume (the first model is a map and the second model is a tomogram). I have managed to place a sphere at the desired position, but I can’t figure out how do this with a volume file. Example for placing the sphere: shape sphere radius 100 center 5243.80,3103.88,1086.01 coordinateSystem #1 How can I achieve the same but placing a volume at the desired location? Thanks a lot, Best, Matthias

2 2

building peptides in Chimera/ChimeraX
by Elaine Meng 18 Mar '21

18 Mar '21

Hello Ishita, Your question was forwarded. I'm CC'ing chimerax-users(a)cgl.ucsf.edu (which is the recommended address for asking ChimeraX questions, for future reference). Peptide-building in ChimeraX (and Chimera) does not predict the structure, it simply uses whatever angles you choose in the Build Structure tool. So the reason it looks like a helix is that you told it to use the helix phi,psi angles. ChimeraX build structure: <https://rbvi.ucsf.edu/chimerax/docs/user/tools/buildstructure.html#start> In the paper figure, it says that the indolicin structure is PDB 1g89 ... so if you want it to look like that, just open 1g89. In the legend it says structures generated by Chimera, but apparently they just mean that images of the structures (from the PDB) were generated by Chimera. 1g89 is an NMR ensemble of 16 structures, so they apparently just chose one of the structures to show and hid the others. I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco > -------- Forwarded Message -------- > Subject: Request regarding ChimeraX (From UC Davis) > Date: Wed, 17 Mar 2021 02:04:17 +0000 > From: Ishita M Shah <imshah(a)ucdavis.edu> > > Dear Dr. Ferrin, > I’m a Project Scientist at UC Davis and the newest user of ChimeraX. > I have a few questions regarding peptide secondary structure generation using ChimeraX. Would you please connect me to someone who could spare a few minutes to walk me through? > > Specifically, when I look up the secondary structure of Indolicidin (generated using Chimera) _ image attached from publication (it shows an extended structure), whereas, using Tools -> Structure editing -> Build Structure -> Start Structure -> Peptide/ Input sequence, the image I get is completely different (attached .bmp file). I really have basic requirements of prediction of secondary structures of peptides (and was using Indolicidin as an control) prior to exploring more. > > I apologize if this information regarding the use of ChimeraX for this purpose is readily available and that I could just be unaware of it. Any advice you can offer is highly appreciated. > > Best, > ishita. > > Ishita M. Shah, PhD > Associate Project Scientist > Associate Director, Microbial Programs > Foods for Health Institute > Department of Food Science and Technology > University of California Davis > 1221 Robert Mondavi Institute, South > Davis, CA 95616 > Phone: (530) 754-7962 > > https://ffhi.ucdavis.edu/people/ishita-shah > https://www.linkedin.com/in/ishita-m-shah-ph-d-a6a9664/ >

2 1

Re: [chimerax-users] Global Diffusion Tensor Models
by Tom Goddard 17 Mar '21

17 Mar '21

Hi Arun, I see from the image you sent that the diffusion ellipsoid script I sent you gave the wrong ellipsoid orientation -- it was using the inverse of the rotation. Here is the correct ChimeraX script which I have also put on the ChimeraX Recipes web site: https://rbvi.github.io/chimerax-recipes/ellipsoid/ellipsoid.html <https://rbvi.github.io/chimerax-recipes/ellipsoid/ellipsoid.html> Tom > On Mar 17, 2021, at 12:23 PM, Gupta, Arun <Arun.Gupta(a)warwick.ac.uk> wrote: > > Dear Tom, > > I think this is what you referring w.r.t to orientation. I have also overlaid the ROTDIF generated output " Tensor axes " on the protein PDB in the attached pic. Although, the ellipsoid shape model representation doesn't looks great. > > Best Regards > > Arun >

1 0

How to produce a multiple sequence alignment
by Luca Pellegrini 17 Mar '21

17 Mar '21

Hello, I have superposed several similar structures with Matchmaker, and now would like to obtain a structure-based multiple sequence alignment. In Chimera, Matchmaker has a nice button ‘After superposition, compute structure-based multiple sequence alignment’. ChimeraX produces all pairwise alignments with the reference structure, but how do I obtain a single alignment with all my sequences? Best wishes, Luca Luca Pellegrini, PhD Department of Biochemistry University of Cambridge Cambridge CB2 1GA UK

2 1

Global Diffusion Tensor Models
by Gupta, Arun 16 Mar '21

16 Mar '21

Dear ChimeraX team, I am interested in generating Global diffusion tensor geometry models such as Prolate, Ellipsoid, Oblate or Sphere with overlaid protein PDB or structure of my interest. (As shown in attached pic). I am using diffusion tensor prediction module provided in ROTDIF software (http://gandalf.umd.edu/FushmanLab/pdsw.html) to generate an ellipsoid diffusion tensor of a protein molecule of my interest. Is it possible that I can use ROTDIF generated output (Which is in Pymol output format, please find attached ) to open in ChimeraX with my protein PDB, to render such representative figure ? Thanks!!! Arun Arun Gupta PhD, MRSC Research Fellow Challis & Lewandowski Group Chemistry Department University of Warwick Gibbet Hill Road Coventry CV4 7AL United Kingdom

3 2

Blast Protein not working
by Moradkhan, Tiglath A 16 Mar '21

16 Mar '21

Hi I found what the problem is. I forgot to blast my sequence. But when I run the blastprotein tool I get the following error: ValueError: not enough values to unpack (expected 2, got 1) File "C:\Program Files\ChimeraX 1.1\bin\lib\site-packages\chimerax\blastprotein\pdbinfo.py", line 206, in fetch_info entry, auth_cid = pcid.split('_') See log for complete Python traceback. I would be very grateful for your help. Thank you Tiglath

2 1