July 2020 - ChimeraX-users - RBVI Mailing Lists

Re: [chimerax-users] [Chimera-users] Difference in alignment of cofactor between Chimera and ChimeraX
by Elaine Meng 08 Jul '20

08 Jul '20

Hello Derek, You're welcome! Thought I should also mention that in matchmaker you can specify more than one pair of chains, i.e. use multiple pairs of chains for the fit. Which way is better depends on what you are trying to achieve: overall superposition of whole complex, or best fit of a single pair of chains. This multichain fit can be done in either Chimera or ChimeraX with a matchmaker command option. I'll use ChimeraX commands to illustrate: open 3u7q open 5n6y hide ribbon The ChimeraX log shows that 3u7q has four chains, two alpha (A,C) and two beta (B,D) whereas 5n6y has 6 chains, two alpha (A,D), two beta (B,E) and two delta (C,F). So perhaps you want to use all of the alpha and beta chains from the structures for the fit instead of only A with A. Now the trick is to figure out which alpha goes with which alpha and which beta with which beta. I might be lazy and first just try ABCD in 3u7q as pairing with ABDE in 5n6y and see if that looks right. I might also use coloring by polymer (figures out a color based on chain sequence) to make it clearer what's what. color bypolymer mm #2/A,B,D,E to #1/A,B,C,D pair ss <http://rbvi.ucsf.edu/chimerax/docs/user/commands/matchmaker.html> .... which actually looks right. The alphas and betas are colored the same within a given PDB structure but different between the structures, probably because of small sequence differences (something like one structure has a couple more residues on one end or the other; I didn't investigate, but it's expected). The Log reports: RMSD between 1026 pruned atom pairs is 1.314 angstroms; (across all 1783 pairs: 3.281) ...now to check the placement of cofactors: show ligand ~ribbon You can see a more symmetrical fit, with similar displacements on both sides. I hope this helps, Elaine > On Jul 7, 2020, at 9:39 AM, Derek Harris <derek.harris(a)aggiemail.usu.edu> wrote: > > Hello Elaine, > > Thank you very much for the rapid response. Following your commands, and seeing all the ligands, I now see what's happening. Just as you said; the proteins are each composed of two identical catalytic halves. In one half there is a nice alignment of the cofactors, but in the other half there is not. I was isolating, and only viewing, the chain where there isn't good alignment. > > Thanks again! > > Derek Harris > Postdoctoral Fellow > Seefeldt Lab > Department of Chemistry and Biochemistry > Utah State University > > > On Tue, Jul 7, 2020 at 9:35 AM Elaine Meng <meng(a)cgl.ucsf.edu> wrote: > Hello Derek, > It does not make sense for the cofactor alignment to be different if the protein alignment is the same, because the cofactors are just moved rigidly along with their proteins. Also the results should be identical between Chimera and ChimeraX, assuming you are using the same options. > > I get exactly the same alignment between these two proteins in the Chimera vs. ChimeraX. > > Chimera commands (including display ligand only): > open 3u7q > open 5n6y > mm #0 #1 > ~ribbon > show ligand > > ChimeraX commands: > open 3u7q > open 5n6y > mm #2 to #1 > ~ribbon > hide > show ligand > > In both programs, I get the same RMSD message to the Log: > > RMSD between 375 pruned atom pairs is 0.968 angstroms; (across all 450 pairs: 2.998) > > In both programs, matchmaker ends up choosing to align chain A to chain A. Thus the cofactors in chains A are better superimposed than the ones in the other chains. > > I hope this helps, > Elaine > ----- > Elaine C. Meng, Ph.D. > UCSF Chimera(X) team > Department of Pharmaceutical Chemistry > University of California, San Francisco > > > On Jul 6, 2020, at 6:47 PM, Derek Harris <derek.harris(a)aggiemail.usu.edu> wrote: > > > > Hello, > > > > I am trying to reproduce an alignment I had previously done in Chimera in ChimeraX with matchmaker command. > > > > The protein alignment between the two programs is the same, but in ChimeraX the active-site cofactor of the matchstruct isn't aligning appropriately to the refstruct cofactor and is unreasonably placed in relation to even its own protein. In Chimera the two cofactors align nicely. > > > > The two structures I am trying to align are 3U7Q and 5N6Y. > > > > Thanks in advance for any assistance you can offer! > > > > Derek Harris > > Postdoctoral Fellow > > Seefeldt Lab > > Department of Chemistry and Biochemistry > > Utah State University > > _______________________________________________ > Chimera-users mailing list: Chimera-users(a)cgl.ucsf.edu > Manage subscription: https://www.rbvi.ucsf.edu/mailman/listinfo/chimera-users

2 1

how to turn molecule centering a specific atom in ChimeraX
by Shin Inchul 07 Jul '20

07 Jul '20

Hi, I was trying to turn molecule around specific atom. Since I am a beginner of ChimeraX, I would like to have some working examples in the command line. Without it, I have difficulties in applying the usage as appeared in the user guide. In PyMOL, mouse middle button click on a atom sets the centering on the atom then use rotate command. I appreciate all the supports provided. Thank you. Inchul

3 3

Re: [chimerax-users] session file sizes
by Tom Goddard 07 Jul '20

07 Jul '20

To close molecular surfaces surface close #1 or just surf close If instead you do ~surface that hides the surface. I discourage use of these obscure "~" commands that Chimera used. Use "surface hide #1" or "surface close #1". If you just hide the surfaces they are still in the session file. You can also see the surface models in Model Panel by clicking the little triangle and the left of the model name to show submodel of your molecule. And you can choose those surface models and close them. Tom > On Jul 7, 2020, at 2:48 PM, Melissa Jurica <mjurica(a)ucsc.edu> wrote: > > That was my first guess, so I turned off the surface view. However, no change in file size. I don’t see how to delete the surface that I calculated- still trying to get the hang of ChimeraX after switching from Chimera. > > > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > Melissa S. Jurica, Ph.D. > Professor, Molecular, Cell & Developmental Biology > Center for Molecular Biology of RNA > University of California, Santa Cruz > 1156 High Street > Santa Cruz, CA 95064 > > Office: 450 Sinsheimer Labs Lab: 434 Sinsheimer Labs > Office phone (831) 459-4427 Lab phone (831) 459-2463 Fax (831) 459-3139 > http://www.mcd.ucsc.edu/faculty/jurica.html <http://www.mcd.ucsc.edu/faculty/jurica.html> > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > >> On Jul 7, 2020, at 2:41 PM, Tom Goddard <goddard(a)sonic.net <mailto:goddard@sonic.net>> wrote: >> >> Hi Melissa, >> >> Molecular surfaces make session files very large because they save all the surface triangles. Did you have a molecular surface in that large session? I have considered not saving the molecular surfaces and just recomputing them although that will make restoring the session slower as it tries to recompute the surface. >> >> Tom >> >> >>> On Jul 7, 2020, at 2:28 PM, Melissa Jurica <mjurica(a)ucsc.edu <mailto:mjurica@ucsc.edu>> wrote: >>> >>> What deterimines the file size of saved ChimeraX sessions (.csx files)? I saved a session of one structure (6Y5q) at 4 MB, and another structure (6y50) at 2.6 MB. However, when I saved a session with both structures open, it was a whopping 120.6 MB. Is there a way to reduce that file size? >>> >>> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ >>> Melissa S. Jurica, Ph.D. >>> Professor, Molecular, Cell & Developmental Biology >>> Center for Molecular Biology of RNA >>> University of California, Santa Cruz >>> 1156 High Street >>> Santa Cruz, CA 95064 >>> >>> Office: 450 Sinsheimer Labs Lab: 434 Sinsheimer Labs >>> Office phone (831) 459-4427 Lab phone (831) 459-2463 Fax (831) 459-3139 >>> http://www.mcd.ucsc.edu/faculty/jurica.html <http://www.mcd.ucsc.edu/faculty/jurica.html> >>> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ >>> >>> _______________________________________________ >>> 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> >> >

1 0

session file sizes
by Melissa Jurica 07 Jul '20

07 Jul '20

What deterimines the file size of saved ChimeraX sessions (.csx files)? I saved a session of one structure (6Y5q) at 4 MB, and another structure (6y50) at 2.6 MB. However, when I saved a session with both structures open, it was a whopping 120.6 MB. Is there a way to reduce that file size? ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Melissa S. Jurica, Ph.D. Professor, Molecular, Cell & Developmental Biology Center for Molecular Biology of RNA University of California, Santa Cruz 1156 High Street Santa Cruz, CA 95064 Office: 450 Sinsheimer Labs Lab: 434 Sinsheimer Labs Office phone (831) 459-4427 Lab phone (831) 459-2463 Fax (831) 459-3139 http://www.mcd.ucsc.edu/faculty/jurica.html <http://www.mcd.ucsc.edu/faculty/jurica.html> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

2 1

label_seq_id in mmcif files for Chimera
by Marcin Wojdyr 07 Jul '20

07 Jul '20

Hi, I'd like to generate mmCIF files (using my own software) that can be read by Chimera. The problem is that sometimes I don't know the full sequence (SEQRES), so I don't know values for _atom_site.label_seq_id. If I skip label_seq_id entirely, the file doesn't load. ChimeraX gives warning: "Skipping atom_site category: Missing column 'label_seq_id'". If I set all label_seq_id values to null ('.') I get warning: "Unable to infer polymer connectivity due to unspecified label_seq_id for residue "ARG" near line 125" and all residues are disconnected. How should the mmCIF look like so that Chimera infers the connectivity in the same way as it does for PDB files? Cheers, Marcin

2 1

Re: [chimerax-users] [ChimeraX] #3464: partial surface mesh
by Elaine Meng 07 Jul '20

07 Jul '20

Hi Rory, There is no "hydrophobic surface" ... maybe you mean the molecular surface shown as solid (not dot or mesh)? Chimera's presets include "hydrophobicity surface" which is perhaps misleading, simply means the molecular surface colored by hydrophobicity. What exactly are you trying to do? Show protein as mesh surface and ligand as solid surface? You can definitely do it (or any combination of mesh surface on some atoms, solid on some other atoms) by opening multiple copies of the structure, as we said in previous messages. Then in the solid-surface copy, only show the parts of the surface you want as solid. In the mesh-surface copy, show the parts you want as mesh. For example, here is an example that shows H-ras as mesh surface with its ligand GCP as solid surface (image attached below), commands: open 121p open 121p surf #1 & protein surf #2:gcp surf style #1 mesh transp #1 50 surfaces color #2 gold surfaces hide solvent This can all be done with the menu as well, but it would take a lot more typing to enter that whole procedure. I Cc'd chimerax-users(a)cgl.ucsf.edu which is the recommended address for asking ChimeraX questions (for future reference). I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco > On Jul 6, 2020, at 10:30 AM, Gordon, Rory <rory.gordon.19(a)ucl.ac.uk> wrote: > > Hi Guys, > > Further to previous email, just tried fiddling with the transparency settings. > > I can get the protein structure to change through the transparency gradient for hydrophobic surface. > > However, for the substrate/ligands there seems to be binary system where you can either have full transparency or none at all. > > The hydrophobic surface can be coloured both for the protein and substrate/ligand. > > Cheers, > > Rory > From: Gordon, Rory <rory.gordon.19(a)ucl.ac.uk> > Sent: 06 July 2020 18:26 > To: Tom Goddard <goddard(a)sonic.net>; Elaine Meng <meng(a)cgl.ucsf.edu> > Cc: Eric Pettersen <pett(a)cgl.ucsf.edu> > Subject: Re: [ChimeraX] #3464: partial surface mesh > > Hi Tom, Elaine and Eric > > I tried to establish the mesh surface with the whole protein structure selected through the GUI. I can only get the hydrophobic surface which I can colour with B-factor. > > The options of Mesh, Dot or Solid don't seem to change anything. > > Not sure how to get round this, so any help would be appreciated. > > Cheers, > > Rory > > > From: Tom Goddard <goddard(a)sonic.net> > Sent: 02 July 2020 18:24 > To: Elaine Meng <meng(a)cgl.ucsf.edu> > Cc: Gordon, Rory <rory.gordon.19(a)ucl.ac.uk>; Eric Pettersen <pett(a)cgl.ucsf.edu> > Subject: Re: [ChimeraX] #3464: partial surface mesh > > Another trick that makes these meshes look nicer is to make them about 50-80% transparent (command "transp #2 70"). It gives the mesh an airy look. > > Tom > > >> On Jul 2, 2020, at 10:17 AM, Elaine Meng <meng(a)cgl.ucsf.edu> wrote: >> >> Hi Rory, >> Here is an example image where I opened the same structure twice. I showed one copy with mesh surface and the second one with solid surface, and then simply hid parts of the solid surface. This is the way that Tom mentioned. >> Elaine >> >> <example.png> >> >> >>> On Jul 2, 2020, at 3:29 AM, ChimeraX <ChimeraX-bugs-admin(a)cgl.ucsf.edu> wrote: >>> >>> #3464: partial surface mesh >>> ---------------------------------------+---------------------------- >>> Reporter: rory.gordon.19@… | Owner: Eric Pettersen >>> Type: defect | Status: closed >>> Priority: normal | Milestone: >>> Component: Surface | Version: >>> Resolution: fixed | Keywords: >>> Blocked By: | Blocking: >>> Notify when closed: | Platform: all >>> Project: ChimeraX | >>> ---------------------------------------+---------------------------- >>> >>> Comment (by rory.gordon.19@…): >>> >>> {{{ >>> Hi Eric, >>> >>> Thanks for your email. >>> >>> I had actually seen a graphic in a publication which gave me that idea. >>> >>> Thanks anyway for the pointers. >>> >>> Cheers, >>> >>> Rory >>> ________________________________ >>> From: ChimeraX <ChimeraX-bugs-admin(a)cgl.ucsf.edu> >>> Sent: 01 July 2020 18:23 >>> Cc: goddard(a)cgl.ucsf.edu <goddard(a)cgl.ucsf.edu>; meng(a)cgl.ucsf.edu >>> <meng(a)cgl.ucsf.edu>; pett(a)cgl.ucsf.edu <pett(a)cgl.ucsf.edu>; Gordon, Rory >>> <rory.gordon.19(a)ucl.ac.uk> >>> Subject: Re: [ChimeraX] #3464: partial surface mesh (was: ChimeraX bug >>> report submission) >>> >>> #3464: partial surface mesh >>> ---------------------------------------+---------------------------- >>> Reporter: rory.gordon.19@… | Owner: Eric Pettersen >>> Type: defect | Status: accepted >>> Priority: normal | Milestone: >>> Component: Unassigned | Version: >>> Resolution: | Keywords: >>> Blocked By: | Blocking: >>> Notify when closed: | Platform: all >>> Project: ChimeraX | >>> ---------------------------------------+---------------------------- >>> Changes (by Eric Pettersen): >>> >>> * status: new => accepted >>> * cc: Tom Goddard, Elaine Meng (added) >>> * project: => ChimeraX >>> * platform: => all >>> * owner: (none) => Eric Pettersen >>> >>> >>> Comment: >>> >>> Hi Rory, >>> I don't believe you can change just part of a surface into mesh, >>> it's all or nothing. You can make that part of the surface partially >>> transparent instead with Actions→Surface→Transparency. >>> >>> --Eric >>> >>> Eric Pettersen >>> UCSF Computer Graphics Lab >>> >>> ----- >>> Tom, should the command "surface style (#!4 & sel) mesh" actually do >>> nothing (with part of the surface selected)? That seems surprising. >>> Seems like it should change the whole surface into mesh. Is that a bug >>> or >>> am I misunderstanding something? >>> >>> --Eric >>> >>> -- >>> Ticket URL: >>> <https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fplato.cgl…> >>> ChimeraX >>> <https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.rbvi.u…> >>> ChimeraX Issue Tracker >>> }}} >>> >>> -- >>> Ticket URL: <https://plato.cgl.ucsf.edu/trac/ChimeraX/ticket/3464#comment:7> >>> ChimeraX <http://www.rbvi.ucsf.edu/chimerax/> >>> ChimeraX Issue Tracker

2 1

artificial lines or seems on surface using soft lighting
by Steve Chou 06 Jul '20

06 Jul '20

Hi ChimeraX folks, When I used soft lighting, I saw lines/seams on the molecular surface (see the attached image). To reproduce this: ---------------------- load xxxx.pdb surface color violet lighting soft ---------------------- How to remove these artificial lines/seams? These lines also show up in the cutting plane in a section view. Many thanks in advance! Steve -- Steve Chou

2 2

Can I only clip some models, but not all of them?
by Steve Chou 05 Jul '20

05 Jul '20

Hi ChimeraX developers and users, When a ligand is embedded in the protein, can I only clip the surface of the protein, but not the ligand (sticks)? Many thanks, Steve -- Steve Chou

2 1

How to obtain size measures
by Andrea Dallapè 04 Jul '20

04 Jul '20

Dear UCSF Chimera and ChimeraX developers, I would like to ask you if there's a method to obtain the size measures of a protein in order to compare with the heights and lengths obtained by AFM. Could you please suggest me how could I obtain such measurements? Thank you in advance for your help and support, Andrea

2 2

color offset and surface type for mapping Electrostatic Potential
by Steve Chou 02 Jul '20

02 Jul '20

Hello everyone, I tried to map Electrostatic Potential (ESP) onto the molecular surface in ChimeraX v1.0 and PyMOL v2.1, but I found that the results were very different in these two programs. Therefore, I wrote to Elaine Meng and Tom Goddard for help. For your reference, I repost the ideals of our emails below. I first calculated the ESP with DelPhi v5.1 (a relatively old version) to get the .phi file (ESP). PyMOL commands for displaying ESP on the molecular surface are as follows: ------------------------------------------------ load 1ubq.pdb bg_color white remove resn hoh remove hydrogens hide line select heavy, name c*+o*+s*+n* show surface load 1ubq.phi, e_map ramp_new e_lvl, e_map, [-7, 0, 7] set surface_color, e_lvl, 1ubq set light_count, 1 set ambient, 0.5 set reflect, 0.2 ------------------------------------------------ ChimeraX commands for displaying ESP on the molecular surface are as follows: ------------------------------------------------ open 1ubq.pdb open 1ubq.phi hide #2 set bg white surface #1 color electrostatic #1 map #2 palette -7,red:0,white:7,blue material dull lighting simple ------------------------------------------------ In PyMOL, the colors (red/blue) are mainly on the residue (E/D, K/R) tips, while in ChimeraX, the colors are spread. Besides, the surface shapes are also slightly different in these two programs. See the image (left and middle panels) for the different colors and shapes. The color difference is caused by the color offset. The shape difference is also explained below. I changed the color offset to 0. Now the color looks similar in these two programs (left and right panels). The color offset (1.4 A) in ChimeraX gives a better idea of the possible charge interactions with other molecules. ------------------------------------------------ color electrostatic #1 map #2 palette -7,red:0,white:7,blue offset 0.0 ------------------------------------------------ The summary of the explanations from Tom: ------------------------------------------------ One idea I have for why PyMol shows redder tips is because ChimeraX is showing the potential 1.4 Angstroms away from the surface since that is about how far away a binding molecule would be. PyMol may be showing the ESP values exactly on the surface. You can try using 0 offset instead of 1.4 with the "color electrostatic" command "offset" parameter. https://www.cgl.ucsf.edu/chimerax/docs/user/commands/color.html#map ------------------------------------------------ The summary of the explanations from Elaine: ------------------------------------------------ As Tom mentioned, the rationale for the 1.4 A projection is that Chimera and ChimeraX show the solvent-excluded surface (SES), as opposed to the solvent-accessible surface (SAS). The SAS is approximately half a probe radius outward, and the SAS represents the closest possible approach of atomic centers of waters or ligands interacting with the protein, so the result is coloring by what values of the electrostatic potential (ESP) that such ligands would "see". The diagram in this page shows the relationships among SES, SAS, and probe. <http://rbvi.ucsf.edu/chimerax/docs/user/commands/surface.html> Possible reasons for difference in Pymol, if the ESP map itself is identical: (1) coloring by offset values or not. It has been my general understanding that when programs color the SES by ESP, the default approach is to color by the values one surface-probe radius outward. (2) shape/location of the surface. Is it an SES or SAS, or neither? Pymol surfaces often look more "melty" or smoothed to me than the SES in Chimera/ChimeraX. (3) the exact color-mapping, i.e. which colors at which values. Pymol ESP coloring often uses a duller-looking blue. ------------------------------------------------ All the best, Steve -- Steve Chou

1 0

ChimeraX-users July 2020