- ChimeraX-users - RBVI Mailing Lists

scalemol extension not working
by André Graça 19 Jan '23

19 Jan '23

Hi everyone, I hope to at least get a reply from Tom, but this may need a bit of troubleshooting (?). I looked into a very old post of yours, Tom, about rescaling an atomic model (see it at the end of my email). I put everything in place but I get an error and I wonder if anything has changed with the PDB format since 2005, or some incompatibility with new chimera versions... If there is actually a way of doing the same with ChimeraX, that would be incredible =) Cheers! André Error and forwarded message comes now: __________________________________________________ AttributeError Exception in Tk callback Function: <bound method MidasUI.processCommand of <Midas.midas_ui.MidasUI instance at 0x00000000079092C8>> (type: <type 'instancemethod'>) Module: <module 'Midas.midas_ui' from 'C:\Program Files\Chimera 1.16\share\Midas\midas_ui.pyc'> (line: 618) Args: (<Tkinter.Event instance at 0x00000000198D8AC8>,) Event type: KeyPress (type num: 2) Traceback (innermost last): File "C:\Program Files\Chimera 1.16\bin\lib\site-packages\Pmw\Pmw_1_3_3\lib\PmwBase.py", line 1747, in __call__ return apply(self.func, args) File "C:\Program Files\Chimera 1.16\share\Midas\midas_ui.py", line 628, in processCommand midas_text.makeCommand(cmdText) File "C:\Program Files\Chimera 1.16\share\Midas\midas_text.py", line 69, in makeCommand f(c, args) File "C:\Program Files\Chimera 1.16\share\scalemol\ChimeraExtension.py", line 23, in scale_molecule x, y, z = a.coord().xyz.data() AttributeError: '_chimera.Point' object has no attribute 'xyz' ================================================ Event contents: char: delta: 0 height: ?? keycode: 13 keysym: Return keysym_num: 65293 num: ?? serial: 14638 state: 8 time: 154527203 type: 2 widget: .121252488L.126915336L.428746824L.428747592L.428747720L width: ?? x: 228 x_root: 288 y: -22 y_root: 941 ______________________________________________________________________ ------- Start of forwarded message ------- Date: Wed, 27 Apr 2005 11:05:55 -0700 (PDT) From: Thomas Goddard <goddard at cgl.ucsf.edu<http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users>> Hi Jinghua, By "virus cage pdb file" I guess you mean you placed atoms at icosahedral positions, and want to show them connected up to show some icosahedral pattern. I think the easiest approach is to open your old session, run a command to change the diameter of the cage, then save a new session. Below is some Python code that provides a "scalemol" Chimera command. Make a directory chimera/share/scalemol in your Chimera distribution and put the code below in file chimera/share/scalemol/ChimeraExtension.py. Then start Chimera, open the cage PDB model, use menu entry Favorites/Command line, and type a command: scalemol 0 1.5 to make model number 0 bigger by a factor of 1.5. If you have more than one model open you can see the model numbers using Favorites / Model Panel, the model number is shown in the left hand column. Tom - ---- scalemol/ChimeraExtension.py code follows: # ----------------------------------------------------------------------------- # def scale_molecule(cmdname, args): from Midas.midas_text import error fields = args.split() try: model_num = int(fields[0]) factor = float(fields[1]) except: error('Syntax error: scalemol <model-number> <factor>') return import chimera mlist = chimera.openModels.list(id = model_num) if len(mlist) == 0: error('scalemol: No model number %d' % model_num) return for model in mlist: for a in model.atoms: x, y, z = a.coord().xyz.data() c = chimera.Coord() c.x, c.y, c.z = (factor*x, factor*y, factor*z) a.setCoord(c) # ----------------------------------------------------------------------------- # import Midas.midas_text Midas.midas_text.addCommand('scalemol', scale_molecule) ------- End of forwarded message -------

2 1

missing or invalid sequences
by Liang, Junyi 19 Jan '23

19 Jan '23

I would like to predicate interaction between two proteins, and tried chimerax version 1.5 after noticing a youtube video https://www.youtube.com/watch?v=6lXeCPuTePs. [https://www.bing.com/th?id=OVP.95Ygyv4AXQTgCS0WB9iXtgHgFo&pid=Api]<https://www.youtube.com/watch?v=6lXeCPuTePs> Running AlphaFold to Predict Protein Complexes from ChimeraX<https://www.youtube.com/watch?v=6lXeCPuTePs> How to predict the structure of a photoreaction center complex of 3 proteins from ChimeraX using AlphaFold-Multimer. Requires a ChimeraX daily build newer than December 2, 2021 or version 1.4. It is not in ChimeraX version 1.3. Web page describing steps https://www.rbvi.ucsf.edu/chimerax/data/alphafold-dec2021/af_multimer.html www.youtube.com I input the two protein sequences in same way as shown in the video and hit predict. But your software says "missing or invalid sequences". How come? Please consider the environment before printing this e-mail Cleveland Clinic is currently ranked as the No. 2 hospital in the country by U.S. News & World Report (2017-2018). Visit us online at http://www.clevelandclinic.org for a complete listing of our services, staff and locations. Confidentiality Note: This message is intended for use only by the individual or entity to which it is addressed and may contain information that is privileged, confidential, and exempt from disclosure under applicable law. If the reader of this message is not the intended recipient or the employee or agent responsible for delivering the message to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please contact the sender immediately and destroy the material in its entirety, whether electronic or hard copy. Thank you.

2 5

Is RR Distance Maps implemented in ChimeraX?
by Michael Bruist 19 Jan '23

19 Jan '23

I use RR Distance Maps in Chimera for my teaching. I want to switch to ChimeraX, but I have not bee able to find this tool. Is it available in ChimeraX? Mike Bruist Michael F. Bruist Professor of Biochemistry Department of Chemistry and Biochemistry St. Joseph’s University 215-596-8530 mbruist(a)sju.edu

2 1

Reasign secondary structure for back-bone trace
by Enrico Martinez 19 Jan '23

19 Jan '23

Dear Chimera-X users! I am working on the visualization of a trajectory containing only back-bone atoms of the protein obtained from the PCA. When I load this multi-pdb ensemble to Chimera-X I can see only back-bone traces without any secondary structure information. I've tried to use dssp command to re-assign the secondary structure but it did not change the visualization state. Would it be possible to do it in some automatic fashion via some built-in command? Many thanks in advance Cheers Enrico

2 4

incomplete output from matchmaker embedded in python script
by Borchert, Milan 19 Jan '23

19 Jan '23

Hello ChimeraX community, I have been trying to use the Chimerax tool to sequentially check one query pdb against multiple (later up to 10000) pdb files. For this I have been working with the python scripting functionality. With the following code I was able to get the resulting data (number of matched atoms and resulting RMSD) saved as a html log file: import os from chimerax.core.commands import run as rc # change to folder with data files os.chdir("/home/ubuntu/Desktop/Chimerax_search/N_crassa_pdb") # gather the names of .pdb files in the folder file_names = [fn for fn in os.listdir(".") if fn.endswith(".pdb")] # give path to query file to be searched against pdb library query = "/home/ubuntu/Desktop/Chimerax_search/query/BDFB_007497.pdb" # open the query file as #1 in ChimeraX rc(session, "open /home/ubuntu/Desktop/Chimerax_search/query/BDFB_007497.pdb") # loop through the files for fn in file_names: # open first target pdb as #2 rc(session, "open " + fn) # match both structures rc(session, "match #1 to #2") # save log with ID matching input filename, close #2 and clear log rc(session, "log save " + fn.split(".")[0] + ".log") rc(session, "close #2") rc(session, "log clear") Instead of trying to gather the relevant data from thousands of rather large html output files I instead would prefer to only get then resulting values from matchmaker. I learned that the command actually has all the values I wanted as return values, so I tried to write a modified script, that saves these values in a variable and collects the values for every target pdb file in a dict: import os from chimerax.core.commands import run as rc # change to folder with data files os.chdir("/home/ubuntu/Desktop/Chimerax_search/N_crassa_pdb") # gather the names of .pdb files in the folder file_names = [fn for fn in os.listdir(".") if fn.endswith(".pdb")] # give path to query file to be searched against pdb library query = "/home/ubuntu/Desktop/Chimerax_search/query/BDFB_007497.pdb" # open the query file as #1 in ChimeraX rc(session, "open /home/ubuntu/Desktop/Chimerax_search/query/BDFB_007497.pdb") # create empty dict to collect matchmaker output output = {} # loop through the files, opening, processing, and closing each in turn for fn in file_names: id = fn.split("/")[-1].split(".")[0] rc(session, "open " + fn) match = rc(session, "match #1 to #2") output.update({id : match}) rc(session, "log save " + fn.split(".")[0] + ".log") rc(session, "close #2") rc(session, "log clear") # display dict print(output) For now I just used 4 target pdbs with my query and only tried printing the result to quickly check if it fits. However the result is a bit baffling to me. I do get the RMSD values, but not the other data: {'NCU01382': [{'full match atoms': <chimerax.atomic.molarray.Atoms object at 0x7f1944492130>, 'full ref atoms': <chimerax.atomic.molarray.Atoms object at 0x7f194448f610>, 'final match atoms': <chimerax.atomic.molarray.Atoms object at 0x7f1946694430>, 'final ref atoms': <chimerax.atomic.molarray.Atoms object at 0x7f1946694550>, 'full RMSD': 63.065617285514556, 'final RMSD': 1.365365439813019, 'transformation matrix': <chimerax.geometry.place.Place object at 0x7f194669c130>, 'aligned ref seq': <chimerax.atomic.molobject.Sequence object at 0x7f19f7b8f8b0>, 'aligned match seq': <chimerax.atomic.molobject.StructureSeq object at 0x7f19f7b8f610>} here is one of the resulting hits, for my first target pdb (NCU01382). As you can see, the RMSD scores are correctly given, but for other values the internal object names and no resulting value are displayed. Any help with figuring this out would be greatly appreciated. Best Wishes Milan Borchert

2 2

select displayed atoms via command line / transfer saved views
by Helge Paternoga 18 Jan '23

18 Jan '23

Dear ChimeraX users, I have to unrelated questions: 1) Is there a way to select displayed atoms via the command line? 2) Is it possible to transfer saved views from one session to another one? Many regards and thank you for reading, Helge

3 3

Clipper's strange behaviour
by André Graça 16 Jan '23

16 Jan '23

Hi everyone! Is there any known issue with Clipper plugin for ChimeraX? After installing clipper and restarting ChimeraX, the software is not responding after I execute the Open command. I expect that it takes some time to open an .mtz file as the map needs to be calculated, however I have waited more than 45 minutes and ChimeraX keeps unresponsive... I am using Chimera 1.5 and I installed the (hopefully compatible) version of Clipper 0.19.0 Did anyone faced the same problem? Let me know if you have a fix or trick ;) Thanks for any feedback in advance, André ____________________ [cid:image001.png@01D9272E.120B2740]

2 2

Movie clipping planes in chimera x
by Greg Martin 13 Jan '23

13 Jan '23

Hi, I’m trying to make a movie where I highlight different parts of a large complex. For one scene, I’m trying to interpolate between a zoomed out view of the whole complex (view p1), and then a zoomed in view of an interface (view p2), which shows a series of interactions. What I want is a smooth transition of both the view and clipping plane from p1 to p2. However, when I give the command: view p1 p2 100 it interpolates the two positions/zooms fine, in 100 frames, but then after the movement adjusts the clipping plane all at once. Is there a way to have the plane adjusted smoothly during transition between p1 and p2? I’ve also tried fly, but this disregards clipping planes altogether. Views are saved with view name p1 view name p2 Thanks in advance, Greg -- Gregory Martin, PhD. Staff Scientist Lab of Andrew Ward Department of Integrative Structural & Computational Biology The Scripps Research Institute La Jolla, CA 92037

3 3

Inquiry: Map sequence conservation onto a protein structure
by Gomez Barillas, Andrea 13 Jan '23

13 Jan '23

Dear ChimeraX developers, I am trying to map sequence conservation onto a protein structure. I am using ChimeraX version (version 1.4 (2022-06-03). For this purpose, I was trying to use a tutorial from 2014 for Chimera for Mapping Sequence Conservation onto Structures (Showing Sequence Conservation in Chimera (ucsf.edu)<https://www.cgl.ucsf.edu/chimera/data/tutorials/systems/outline.html>). It seems that some the tools mentioned in this tutorial are no longer available, or maybe work differently. Specifically, the tutorial's "Case 4" (calculating values outside of Chimera but showing them in Chimera) has you set an attribute, and then render/color the structure based on the attribute. The Tools > Structure Analysis > Define attribute mentioned in the tutorial is no longer available. I was as able to assign values directly to structure residues in a text file and read it in using the command defattr instead. The next step would be to use the tool Render by Attribute for coloring but I cannot find this option anymore. Is there a different command I can use for this step, or a new functionality built into ChimeraX I could use for my purpose of mapping sequence conservation onto the structure? One relevant point is that I already have my sequence identify values calculated from another program, and I want to import those values. Thank you in advance for your help, Andrea Showing Sequence Conservation in Chimera<https://www.cgl.ucsf.edu/chimera/data/tutorials/systems/outline.html> Example 2B Chimera session: ConSurf-1hd2-chimera19.py [show this last!] (result of submitting 1hd2 to the ConSurf server and choosing to show results in Chimera; alignment has 150 sequences including that of the query, 1hd2) . Given a protein structure, the ConSurf server estimates the evolutionary conservation of amino acid positions based on the phylogenetic relations between homologous ... www.cgl.ucsf.edu

2 1

Inquiry: Map sequence conservation onto a protein structure
by Gomez Barillas, Andrea 13 Jan '23

13 Jan '23

Dear ChimeraX developers, I am trying to map sequence conservation onto a protein structure. I am using ChimeraX version (version 1.4 (2022-06-03). For this purpose, I was trying to use a tutorial from 2014 for Chimera for Mapping Sequence Conservation onto Structures (Showing Sequence Conservation in Chimera (ucsf.edu)<https://www.cgl.ucsf.edu/chimera/data/tutorials/systems/outline.html>). It seems that some the tools mentioned in this tutorial are no longer available, or maybe work differently. Specifically, the tutorial's "Case 4" (calculating values outside of Chimera but showing them in Chimera) has you set an attribute, and then render/color the structure based on the attribute. The Tools > Structure Analysis > Define attribute mentioned in the tutorial is no longer available. I was as able to assign values directly to structure residues in a text file and read it in using the command defattr instead. The next step would be to use the tool Render by Attribute for coloring but I cannot find this option anymore. Is there a different command I can use for this step, or a new functionality built into ChimeraX I could use for my purpose of mapping sequence conservation onto the structure? One relevant point is that I already have my sequence identify values calculated from another program, and I want to import those values. Thank you in advance for your help, Andrea

1 0

ChimeraX Log window remains blank
by Scott White 13 Jan '23

13 Jan '23

Hi, Sorry if this is a dumb question, but recently when starting ChimeraX the Log window is not on Display. Opening the Log window (Tools> log) opens it as a separate window, and whether docked or free-floating the Log window remains blank no matter what I do. Even just "open 1nov" on the command line will download and open the PDB entry 1NOV, but no text appears in the Log window. I've tried a fresh install of ChimeraX 1.5 to no avail. I've tried opening an older version of ChimeraX, which despite having worked countless times before, now shows the same behaviour regarding the Log window as my abberant ChimeraX 1.5. I've tried rebooting the computer - no fix. I've tried "log clear" and "log show". It's a Windows 10, university owned PC. I don't have the issue on other computers. Is it a software clash? Thanks in advance for any help or advice. Scott _______________________________________ Dr Scott A. White SFHEA School of Biosciences, Room T716 University of Birmingham, B15 2TT E: s.a.white(a)bham.ac.uk<mailto:s.a.white@bham.ac.uk> T: +44-(0)121-414-7534 Office Hours Generally Tues 10am - 11am, Wed 3 pm - 4 pm Please contact me for an appointment. ______________________________________

2 2

Changing MD trajectory default frame rate
by Visvaldas K. 12 Jan '23

12 Jan '23

Dear all, Is there a way to change frame rate when I hit that red "record" button when playing MD trajectory? I could do "movie record", but then it becomes interactive (I hit triangle "play", and then "movie encode /Users/visvaldas/Desktop/movie.mp4 framerate 5"), so inadvertently can record some undesired frames. Of course I probably can reduce framerate using external programs, but how to that non-interactively in ChimeraX? Best regards, Vis

3 3

How to integrate ChimeraX with phenix
by Suresh K 12 Jan '23

12 Jan '23

Hi, We integrate pymol and coot with phenix but unable to integrate ChimeraX. In internet we found that you using ChimeraX with phenix. Could you help us with way to integrate ChimeraX with phenix? Thanks and regards, Suresh K

2 1

Select complelementary of range
by Pranav Shah 12 Jan '23

12 Jan '23

Hi Team, How do I go about selecting everything that does NOT lie between a given range of residues? For instance I want to select everything that is NOT r.410420. Best, Pranav -- Pranav Shah Postdoctoral Research Fellow. Division of Structural Biology, Wellcome Trust Centre for Human Genetics, University of Oxford, Roosevelt Drive, Oxford OX3 7BN, UK

2 1

Smooth MD trajectory during moovie making
by Enrico Martinez 12 Jan '23

12 Jan '23

Dear Chimera-X users! I am trying to make a movie using MD trajectory (in multi-model pdb file) by means of the Molecular Dynamics Viewer (v 1.2). is it possible to introduce smooth effects on the snapshots transition in the same way like it could be done via V*D ? Many thanks in advance Cheers Enrico

3 10

Trouble displaying missing structure dashed lines in cartoon mode
by Jonathan Jih 11 Jan '23

11 Jan '23

Hello ChimeraX team, I'm having quite a bit of trouble displaying missing structure dashed lines when my models are shown in cartoon mode. This is despite the "missing structure" sub-model box being ticked in the model panel. I recall finding a workaround for this nearly a year ago on ChimeraX 1.3, but I cannot for the life of me remember how I did so (now using ChimeraX 1.5). Interestingly, old sessions that I saved in 1.3 and 1.4 with whatever workaround I applied back then display the missing structure dashed lines properly in 1.5. However, opening the exact same pdb files from those older sessions in 1.5 yields no dashed lines in cartoon mode. Full atom mode appears fine. I also found this thread ( https://www.rbvi.ucsf.edu/pipermail/chimerax-users/2022-February/003273.html) which seems to indicate the issue has been fixed, so not sure if I'm just doing something wrong? My OS is Windows 11 if that matters. Thanks in advance for the help! Jonathan -- Biochemistry, Biophysics, and Structural Biology (BBSB) Molecular Biology Institute, University of California, Los Angeles

2 6

Saving .mrc volume as coordinates
by Joshua Jenkins 10 Jan '23

10 Jan '23

Hi, I've segmented a tomogram using EMAN2 neural network workflow. I then use the segger function to extract regions of interest within the segmented volume, and save these to .mrc. I wondered if there is a way in chimera, or chimeraX, to convert/extract/save the .mrc volume as a simple 3D coordinates file (txt or tbl)? Best wishes, Josh

3 2

Morph movie saving from a script file
by Pranav Shah 09 Jan '23

09 Jan '23

Hi Team, Would there be a way to record a morph between two structures using the command line? Currently, I am clicking the play and record buttons from the GUI in quick succession to achieve this. Best, Pranav -- Pranav Shah Postdoctoral Research Fellow. Division of Structural Biology, Wellcome Trust Centre for Human Genetics, University of Oxford, Roosevelt Drive, Oxford OX3 7BN, UK

3 2

Re: [chimerax-users] [Chimera-users] Stereo glasses
by Jonathan Sheehan 06 Jan '23

06 Jan '23

I'm always on the lookout for new options for 3D stereo display hardware (Chimera or Chimerax). I just saw this Asus system with glasses-free stereo, using eye-tracking cameras and a lenticular screen. It looks to me like a premium OLED laptop targeted to gamers and CAD designers. I'm unfamiliar with Asus Spatial Vision and Acer SpatialLabs, but one reviewer said that it works with multiple people viewing it- I'm slightly sceptical. :-) https://arstechnica.com/gadgets/2023/01/asus-new-16-inch-workstation-laptop… Tom, you suggested that VR headsets were the best option a few years ago. What are your thoughts about this release? (Replies from non-Tom individuals are also welcomed.) Thanks! -Jonathan On Thu, Jun 13, 2019 at 10:32 PM Jonathan Sheehan <jonathan.sheehan(a)gmail.com> wrote: > > Thanks for the information, Tom- I was hoping somebody had better > news. It sounds like we're in a dead zone between the epochs of stereo > glasses and VR headsets. > > I suppose I'll have to start looking into VR systems. > > Thanks, > -Jonathan > > > On Mon, Jun 10, 2019 at 1:19 PM Tom Goddard <goddard(a)sonic.net> wrote: > > > > Hi Jonathan, > > > > Nvidia was the main distributor of 3D active LCD stereo glasses, and they dropped support for their Nvidia 3D Vision glasses and emitters in April 2019 as described here > > > > https://www.ghacks.net/2019/03/11/nvidia-3d-vision-end-of-support/ > > > > https://nvidia.custhelp.com/app/answers/detail/a_id/4781 > > > > Nvidia graphics driver released after April 2019 won't work with these glasses. You can continue to use Nvidia 3D Vision with old graphics drivers, maybe that is the best option. > > > > The LCD active stereo glasses technology appears to be nearly dead. If you try to use it be prepared for problems getting it to work and to pay a lot of money. We have been using active stereo LCD with a stereo projector for many decades. We use it infrequently now, instead using virtual reality headsets with our ChimeraX software. Virtual reality headsets give better immersion but don't yet easily handle having a half-dozen people viewing the same scene in the same room (would require 6 separate computers driving the 6 headsets). > > > > Here are details on the old LCD glasses projector system we are have which is working: > > > > 10 pairs of RealD CE4 and CE4s stereo glasses ($280 each) > > > > http://www.colorlinkjapan.com/brand/product/ce4s.html > > > > A Stereo Graphics EXXR extra long range emitter ($780) > > > > http://www.colorlinkjapan.com/brand/product/exxr-emitter.html > > > > A Christie Mirage S+4K projector circa 2006, (1440 x 1050 resolution), no longer in production > > > > https://www.projectorcentral.com/Christie-Mirage_S+4K.htm > > > > An Nvidia Quadro P6000 graphics card ($3700) > > > > https://www.amazon.com/PNY-Quadro-P6000-Graphic-Card-x/dp/B01M0S2FKR > > > > with a stereo emitter bracket > > > > http://www8.hp.com/h20195/v2/GetPDF.aspx/c04658472.pdf > > > > This system breaks about once a year after graphics driver updates, the fix usually being to set the projector to be either the primary display or secondary display (depends on graphics driver version). Also the glasses fail at maybe the rate of 1 out of 10 pairs per year. > > > > Tom > > > > > > On Jun 7, 2019, at 8:30 AM, Jonathan Sheehan <jonathan.sheehan(a)gmail.com> wrote: > > > > Hi all, > > > > What stereo setup are people recommending these days? It seems that > > NVIDIA has discontinued the 3D Vision 2 goggles+emitter. Is there > > something else that is available and works well? > > > > Thanks for any advice! > > -Jonathan > > _______________________________________________ > > Chimera-users mailing list: Chimera-users(a)cgl.ucsf.edu > > Manage subscription: http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users > > > >

3 3

Coloring according to the secondary structure
by Enrico Martinez 06 Jan '23

06 Jan '23

Dear ChimeraX users! I need to color cartoon representation of a protein according to its secondary structure (e.g. to color helix in red, b-stands in blue and coils in violet etc). I have already seen a built-in plugin in UCSF Chimera (but not in Chimera-X) that could do this job! Could you suggest commands that would be useful in Chimera-X? Many thanks in advance Cheers Enrico

2 2

Unable to find the distance between centroid & any other atom in chimeraX
by Prathvi Singh 05 Jan '23

05 Jan '23

Hi Elaine, I declared the centroid of a tyrosine residue's ring by first selecting its ring members & then using the command: *define centroid sel radius 0.5* Next, when I try to find the distance between the centroid, I get an error whose screenshot I have attached with the email. The command that I am using to find the distance are: *dist sel* dist #1.1 #1.2 (here the distance is being calculated between two centroids) Please help - Prathvi Singh, Research Fellow, Department of Biological Sciences & Bioengineering, Indian Institute of Technology, Kanpur-208016

2 1

Unfocus in chimeraX using command line
by Prathvi Singh 05 Jan '23

05 Jan '23

Hi Elaine, Is there any equivalent of the "~focus" command in chimeraX? -- Prathvi Singh, Research Fellow, Department of Biological Sciences & Bioengineering, Indian Institute of Technology, Kanpur-208016

2 1

Re: [chimerax-users] Remote control / Threading in ChimeraX
by Albert Smith-Penzel 05 Jan '23

05 Jan '23

Hi Tom, Thanks for the explanation. I’ll have to think and investigate a bit to see if my thread use is causing some problems elsewhere. Most of our code has been pretty stable, but there are still a few occasional glitches that I haven’t completely understood. Anyway, session.ui.thread_safe is perfect for resolving the issues we have with adding the surface to the empty session and running commands remotely. The curl solution also works; I had something like this implemented previously for some other purposes, but my own preference is to send commands/data directly rather than writing out a file to be read on the other end. Thanks again! Cheers, Albert ––––––––––––––––––––––––––––––––––– Albert Smith-Penzel, Ph.D. DFG Project Leader Institute for Medical Physics and Biophysics Leipzig University 04107 Leipzig, Germany E-mail: albert.smith-penzel(a)uni-leipzig.de <mailto:albert.smith-penzel@medizin.uni-leipzig.de> > On 4. Jan 2023, at 21:41, Tom Goddard <goddard(a)sonic.net> wrote: > > WARNUNG: Diese E-Mail kam von außerhalb der Organisation. Klicken Sie nicht auf Links oder öffnen Sie keine Anhänge, es sei denn, Sie kennen den Absender und wissen, dass der Inhalt sicher ist. > > WARNUNG: Diese E-Mail kam von außerhalb der Organisation. Klicken Sie nicht auf Links oder öffnen Sie keine Anhänge, es sei denn, Sie kennen den Absender und wissen, dass der Inhalt sicher ist. > > Hi Albert, > > ChimeraX is not designed to run arbitrary Python code multithreaded. There are many reasons. First the Qt window toolkit is not thread safe -- all Qt GUI calls have to be done in the main thread. Violating that often results in a crash. But even without the GUI, imagine the main thread is looping over the list of open models while a second thread decides to add a model to that list -- this could break the looping thread many ways. Python has a global interpreter lock which means it only can run one thread at a time. Still if you pause one thread and then change the state of the ChimeraX session in another thread, the first thread could throw errors in countless ways when it resumes at an arbitrary point in the code with the data structures changed. In general very little software is thread safe because it is very difficult to implement. Usually that is only done by making threads not modify any data structures used by other threads. > > Here are two solutions. 1) Have your listener thread run the Python code it receives in the main thread rather than the listener thread. There is a ChimeraX function that takes a function and queues it to be run in the main thread. > > session.ui.thread_safe(func, args...) > > Here is the code if you want to see how it works > > https://github.com/RBVI/ChimeraX/blob/2e33222fabb48edec76df6e82a10efedb54a8… > > A second solution is to use the ChimeraX remotecontrol communication for everything. That mechanism just runs ChimeraX commands, but you can run Python code using it by just sending the command "open script27.py" where you put the Python you want to run in a file script27.py. > > Tom > > >> On Jan 4, 2023, at 4:40 AM, Albert Smith-Penzel via ChimeraX-users <chimerax-users(a)cgl.ucsf.edu> wrote: >> >> Hello, >> >> I am working on a Python project that uses ChimeraX (currently running 1.5 to interactively view dynamics data by mapping it onto a molecule. All the figures/movies in this paper, https://www.nature.com/articles/s41467-021-27417-y <https://www.nature.com/articles/s41467-021-27417-y>, were made possible by Chimera! Also thanks to Tom Goddard for explaining to me how some of those figures could be made awhile ago (https://rbvi.github.io/chimerax-recipes/spherical_harmonics/spherical_harmo… <https://rbvi.github.io/chimerax-recipes/spherical_harmonics/spherical_harmo…> ). >> >> So we’re controlling ChimeraX remotely; we launch a session and then can send commands from Python over to Chimera and edit images from within our python code (while still allowing us to interact via the ChimeraX interface). The basic way this works is as follows >> >> launch chimeraX from our code, including a short python script >> python script imports our libraries, launches “cmxr” (chimeraX remote) >> cmxr uses multiprocessing.connection.Listener to listen for commands/data coming from Client in our python code >> !!!The Listener runs in a thread (threading.Thread) so that we continue to be able to interact normally with ChimeraX (launching the Listener without using a thread freezes ChimeraX while waiting for commands, rendering it unusable) >> Then, we just tell it to run some function(s) that we’ve imported into ChimeraX, and provide some data, atom ids, etc. over the Client/Listener connection >> >> There are a few things that I can’t figure how to make work via this process. We’ve had a few long-standing workarounds, but I’ve always been a little disturbed that we can’t get this all to work from python alone. At the moment, when we launch ChimeraX, we also enable remote control via curl, and some commands are then sent that way (remotecontrol rest start port XXXX) >> >> In particular, if I want to send commands that would normally be executed at ChimeraX’s command line, in principle I should be able to use chimerax.core.commands.run. If I execute “run” from a simple python script, everything works, but if I run it from within a thread (as required by our Listener setup), then ChimeraX freezes. So, we send these commands via curl on the remotecontrol instead. >> >> Second, is there a right way to open models/surfaces from the ChimeraX python shell? At the moment, we also open pdbs via curl commands. For pdbs, maybe I just don’t know the right python command. However, I am re-implementing spherical harmonics referenced above. Once I have prepared the surface (“s”), if I add it to session.models, it crashes chimeraX if it is being added within the listener thread, and if there are no models already open. If another model is already open, then it works just fine, and if I load the same modules/execute the same commands from within the python shell provided in ChimeraX, it also works (regardless of other models being open). It seems like maybe ChimeraX isn’t able to fully initialize the session from within a thread. >> >> I’ve grouped these problems together because I guess mainly they have to do with being unable to do some actions from within threads, and if there’s an explanation/workaround for dealing with that. >> >> Thanks for your help! >> >> Cheers, >> Albert >> >> >> >> ––––––––––––––––––––––––––––––––––– >> Albert Smith-Penzel, Ph.D. >> DFG Project Leader >> Institute for Medical Physics and Biophysics >> Leipzig University >> 04107 Leipzig, Germany >> E-mail: albert.smith-penzel(a)uni-leipzig.de <mailto:albert.smith-penzel@medizin.uni-leipzig.de> >> >> >> >> >> >> _______________________________________________ >> ChimeraX-users mailing list >> ChimeraX-users(a)cgl.ucsf.edu >> Manage subscription: >> https://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users >

1 0

Remote control / Threading in ChimeraX
by Albert Smith-Penzel 04 Jan '23

04 Jan '23

Hello, I am working on a Python project that uses ChimeraX (currently running 1.5 to interactively view dynamics data by mapping it onto a molecule. All the figures/movies in this paper, https://www.nature.com/articles/s41467-021-27417-y <https://www.nature.com/articles/s41467-021-27417-y>, were made possible by Chimera! Also thanks to Tom Goddard for explaining to me how some of those figures could be made awhile ago (https://rbvi.github.io/chimerax-recipes/spherical_harmonics/spherical_harmo… <https://rbvi.github.io/chimerax-recipes/spherical_harmonics/spherical_harmo…> ). So we’re controlling ChimeraX remotely; we launch a session and then can send commands from Python over to Chimera and edit images from within our python code (while still allowing us to interact via the ChimeraX interface). The basic way this works is as follows launch chimeraX from our code, including a short python script python script imports our libraries, launches “cmxr” (chimeraX remote) cmxr uses multiprocessing.connection.Listener to listen for commands/data coming from Client in our python code !!!The Listener runs in a thread (threading.Thread) so that we continue to be able to interact normally with ChimeraX (launching the Listener without using a thread freezes ChimeraX while waiting for commands, rendering it unusable) Then, we just tell it to run some function(s) that we’ve imported into ChimeraX, and provide some data, atom ids, etc. over the Client/Listener connection There are a few things that I can’t figure how to make work via this process. We’ve had a few long-standing workarounds, but I’ve always been a little disturbed that we can’t get this all to work from python alone. At the moment, when we launch ChimeraX, we also enable remote control via curl, and some commands are then sent that way (remotecontrol rest start port XXXX) In particular, if I want to send commands that would normally be executed at ChimeraX’s command line, in principle I should be able to use chimerax.core.commands.run. If I execute “run” from a simple python script, everything works, but if I run it from within a thread (as required by our Listener setup), then ChimeraX freezes. So, we send these commands via curl on the remotecontrol instead. Second, is there a right way to open models/surfaces from the ChimeraX python shell? At the moment, we also open pdbs via curl commands. For pdbs, maybe I just don’t know the right python command. However, I am re-implementing spherical harmonics referenced above. Once I have prepared the surface (“s”), if I add it to session.models, it crashes chimeraX if it is being added within the listener thread, and if there are no models already open. If another model is already open, then it works just fine, and if I load the same modules/execute the same commands from within the python shell provided in ChimeraX, it also works (regardless of other models being open). It seems like maybe ChimeraX isn’t able to fully initialize the session from within a thread. I’ve grouped these problems together because I guess mainly they have to do with being unable to do some actions from within threads, and if there’s an explanation/workaround for dealing with that. Thanks for your help! Cheers, Albert ––––––––––––––––––––––––––––––––––– Albert Smith-Penzel, Ph.D. DFG Project Leader Institute for Medical Physics and Biophysics Leipzig University 04107 Leipzig, Germany E-mail: albert.smith-penzel(a)uni-leipzig.de <mailto:albert.smith-penzel@medizin.uni-leipzig.de>

2 1

How to get the average pTM scores from an Alphafold multimer prediction
by Chen, Henry 04 Jan '23

04 Jan '23

Dear Chimarax devs/users, I'm using ChimeraX Alphafold tool to do a multimer structure prediction. I was wondering how can I determine the average pTM score for each structure prediction? I get PAE.png and plddt.png file, but not a text file or anything that tells me the average pTM. (I was hoping the ranking file would give the average score but it does not) The models come with the pdb file as well as the scores.json files, but again I don't see anything that tells me the average pTM for the structure. Clearly the pTM per residue scores are saved in my pdb file, is there a command to average them all up? Thanks! Henry

3 3