Surface color by homology
Hello, I am trying to color the surface of my model according to sequence conservation. I opened the MSA, associated the proper sequence with the chain I wish to color, and then used the command "color byattr seq_conservation". But this results in the colors not blending at the surface, which is not very pleasant looking. Is there a way to make the transitions between adjacent colors smoother, similar to when we color the surface, for e.g., by electrostatic potential? In legacy chimera I used to follow the "Structure... Render by Conservation" in the Multalign Viewer and then choose the attribute mavConservation, which used to do the trick Thanks in advance! Pedro Bule -- *Pedro Bule* *Assistant Professor* GLYCOBIOLOGY AND STRUCTURAL ENZYMOLOGY LAB *P:* (+351) 21 365 28 84 // 431379 | pedrobule@fmv.ulisboa.pt <email@fmv.ulisboa.pt> *Centro de Investigação Interdisciplinar em Sanidade Animal* Faculdade de Medicina Veterinária Av. Universidade Técnica | 1300-477 Lisboa | PORTUGAL
Hi Pedro, Coloring by conservation is a different situation than coloring by electrostatic potential. The electrostatic potential is varying continuously in space (hence the gradual change in color), whereas the conservation is a single value per residue. Thus, there is a single color for the surface patch that comes from the same residue. So what you see is the true representation of the results. In Chimera, it looked blended only as an artifact, because the surface triangulation did not try to make sharp boundaries per atomic patch; the jagged triangles from adjacent atoms were next to each other and made the color look blended. In ChimeraX, although you can turn off the sharp boundaries, doing that loses the linkage between the surface patch and the atom so you cannot directly color the surface by the property of the atoms (e.g. bfactor) or residues (e.g. conservation). If you *really* want to try using the ChimeraX surfaces with sharp boundaries turned off, then you have to do a multistep process of coloring the atoms and then using color zone to color the surface to match the nearby atoms. Even if you do that, you can see the jagged edges instead of smooth color blending because the conservation is a single value per residue (not continuously varying in space like electrostatic potential). Here is an example with coloring by bfactor, but it would be analogous to do it for conservation. open 2gbp surface sharpBoundaries false color byattr bfactor target a color zone #1.2 near #1 dist 4 It is very ugly in my opinion, does not look like smooth blending. Maybe you could try to get a finer triangulation with other "surface" command options, but in my opinion it would probably be a waste of time. You can see the difference with sharp boundaries true vs. false: surface sharp true Again, the sharper boundaries are a more true representation of the values that you are trying to show. I hope this makes sense, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Resource for Biocomputing, Visualization, and Informatics Department of Pharmaceutical Chemistry University of California, San Francisco
On Feb 14, 2024, at 4:17 AM, Pedro Bule via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:
Hello,
I am trying to color the surface of my model according to sequence conservation. I opened the MSA, associated the proper sequence with the chain I wish to color, and then used the command "color byattr seq_conservation". But this results in the colors not blending at the surface, which is not very pleasant looking. Is there a way to make the transitions between adjacent colors smoother, similar to when we color the surface, for e.g., by electrostatic potential? In legacy chimera I used to follow the "Structure... Render by Conservation" in the Multalign Viewer and then choose the attribute mavConservation, which used to do the trick
Thanks in advance!
Pedro Bule
Hi Pedro, Chimera had a nicer triangulation of the molecular surface than ChimeraX (but the Chimera molecular surface often crashed on large molecules), and that is why it gives nicer blurred conservation coloring than what Elaine showed. The ChimeraX surface option gridSpacing (default 0.5 Angstroms) can make the surface triangles smaller which makes the the jagged edges less, but also makes the blur width less, so it almost looks like the sharp edges. So I don't think that helps. surface #1 gridSpacing 0.2 color zone #1 near #1 distance 5 Tom Here is the appearance of the color zone with grid spacing 0.5 versus 0.2. 
On Feb 14, 2024, at 9:11 AM, Elaine Meng via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:
Hi Pedro, Coloring by conservation is a different situation than coloring by electrostatic potential. The electrostatic potential is varying continuously in space (hence the gradual change in color), whereas the conservation is a single value per residue. Thus, there is a single color for the surface patch that comes from the same residue. So what you see is the true representation of the results.
In Chimera, it looked blended only as an artifact, because the surface triangulation did not try to make sharp boundaries per atomic patch; the jagged triangles from adjacent atoms were next to each other and made the color look blended.
In ChimeraX, although you can turn off the sharp boundaries, doing that loses the linkage between the surface patch and the atom so you cannot directly color the surface by the property of the atoms (e.g. bfactor) or residues (e.g. conservation).
If you *really* want to try using the ChimeraX surfaces with sharp boundaries turned off, then you have to do a multistep process of coloring the atoms and then using color zone to color the surface to match the nearby atoms. Even if you do that, you can see the jagged edges instead of smooth color blending because the conservation is a single value per residue (not continuously varying in space like electrostatic potential).
Here is an example with coloring by bfactor, but it would be analogous to do it for conservation.
open 2gbp surface sharpBoundaries false color byattr bfactor target a color zone #1.2 near #1 dist 4
It is very ugly in my opinion, does not look like smooth blending. <Screen Shot 2024-02-14 at 9.08.30 AM.png>
Maybe you could try to get a finer triangulation with other "surface" command options, but in my opinion it would probably be a waste of time. You can see the difference with sharp boundaries true vs. false:
surface sharp true
<Screen Shot 2024-02-14 at 9.08.44 AM.png>
Again, the sharper boundaries are a more true representation of the values that you are trying to show.
I hope this makes sense, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Resource for Biocomputing, Visualization, and Informatics Department of Pharmaceutical Chemistry University of California, San Francisco
On Feb 14, 2024, at 4:17 AM, Pedro Bule via ChimeraX-users <chimerax-users@cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu>> wrote:
Hello,
I am trying to color the surface of my model according to sequence conservation. I opened the MSA, associated the proper sequence with the chain I wish to color, and then used the command "color byattr seq_conservation". But this results in the colors not blending at the surface, which is not very pleasant looking. Is there a way to make the transitions between adjacent colors smoother, similar to when we color the surface, for e.g., by electrostatic potential? In legacy chimera I used to follow the "Structure... Render by Conservation" in the Multalign Viewer and then choose the attribute mavConservation, which used to do the trick
Thanks in advance!
Pedro Bule
_______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/
Dear Elaine, Thank you very much for your quick and thorough response, which made perfect sense. You are right, it is indeed more accurate to use sharp edges. I guess that, for the intended purpose, which is to localize the active site of an enzyme, a bit of blurring wouldn't hurt, as I am not so worried about showing individual amino acid conservation rather than obtaining a nice, strongly colored patch. But in a way, I guess that is kind of cheating. I'll stick to sharp edges! Thank you very much Pedro Elaine Meng <meng@cgl.ucsf.edu> escreveu (quarta, 14/02/2024 à(s) 17:11):
Hi Pedro, Coloring by conservation is a different situation than coloring by electrostatic potential. The electrostatic potential is varying continuously in space (hence the gradual change in color), whereas the conservation is a single value per residue. Thus, there is a single color for the surface patch that comes from the same residue. So what you see is the true representation of the results.
In Chimera, it looked blended only as an artifact, because the surface triangulation did not try to make sharp boundaries per atomic patch; the jagged triangles from adjacent atoms were next to each other and made the color look blended.
In ChimeraX, although you can turn off the sharp boundaries, doing that loses the linkage between the surface patch and the atom so you cannot directly color the surface by the property of the atoms (e.g. bfactor) or residues (e.g. conservation).
If you *really* want to try using the ChimeraX surfaces with sharp boundaries turned off, then you have to do a multistep process of coloring the atoms and then using color zone to color the surface to match the nearby atoms. Even if you do that, you can see the jagged edges instead of smooth color blending because the conservation is a single value per residue (not continuously varying in space like electrostatic potential).
Here is an example with coloring by bfactor, but it would be analogous to do it for conservation.
open 2gbp surface sharpBoundaries false color byattr bfactor target a color zone #1.2 near #1 dist 4
It is very ugly in my opinion, does not look like smooth blending.
Maybe you could try to get a finer triangulation with other "surface" command options, but in my opinion it would probably be a waste of time. You can see the difference with sharp boundaries true vs. false:
surface sharp true
Again, the sharper boundaries are a more true representation of the values that you are trying to show.
I hope this makes sense, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Resource for Biocomputing, Visualization, and Informatics Department of Pharmaceutical Chemistry University of California, San Francisco
On Feb 14, 2024, at 4:17 AM, Pedro Bule via ChimeraX-users < chimerax-users@cgl.ucsf.edu> wrote:
Hello,
I am trying to color the surface of my model according to sequence conservation. I opened the MSA, associated the proper sequence with the chain I wish to color, and then used the command "color byattr seq_conservation". But this results in the colors not blending at the surface, which is not very pleasant looking. Is there a way to make the transitions between adjacent colors smoother, similar to when we color the surface, for e.g., by electrostatic potential? In legacy chimera I used to follow the "Structure... Render by Conservation" in the Multalign Viewer and then choose the attribute mavConservation, which used to do the trick
Thanks in advance!
Pedro Bule
Hi Pedro, I was curious if I could blur the coloring of the surface across multiple triangles to get a smoother blurred appearance. So I wrote some Python code that defines a "color smooth" command and it worked decently https://rbvi.github.io/chimerax-recipes/color_smooth/color_smooth.html Tom 
On Feb 15, 2024, at 2:08 AM, Pedro Bule via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:
Dear Elaine,
Thank you very much for your quick and thorough response, which made perfect sense. You are right, it is indeed more accurate to use sharp edges. I guess that, for the intended purpose, which is to localize the active site of an enzyme, a bit of blurring wouldn't hurt, as I am not so worried about showing individual amino acid conservation rather than obtaining a nice, strongly colored patch. But in a way, I guess that is kind of cheating. I'll stick to sharp edges!
Thank you very much
Pedro
Elaine Meng <meng@cgl.ucsf.edu <mailto:meng@cgl.ucsf.edu>> escreveu (quarta, 14/02/2024 à(s) 17:11):
Hi Pedro, Coloring by conservation is a different situation than coloring by electrostatic potential. The electrostatic potential is varying continuously in space (hence the gradual change in color), whereas the conservation is a single value per residue. Thus, there is a single color for the surface patch that comes from the same residue. So what you see is the true representation of the results.
In Chimera, it looked blended only as an artifact, because the surface triangulation did not try to make sharp boundaries per atomic patch; the jagged triangles from adjacent atoms were next to each other and made the color look blended.
In ChimeraX, although you can turn off the sharp boundaries, doing that loses the linkage between the surface patch and the atom so you cannot directly color the surface by the property of the atoms (e.g. bfactor) or residues (e.g. conservation).
If you *really* want to try using the ChimeraX surfaces with sharp boundaries turned off, then you have to do a multistep process of coloring the atoms and then using color zone to color the surface to match the nearby atoms. Even if you do that, you can see the jagged edges instead of smooth color blending because the conservation is a single value per residue (not continuously varying in space like electrostatic potential).
Here is an example with coloring by bfactor, but it would be analogous to do it for conservation.
open 2gbp surface sharpBoundaries false color byattr bfactor target a color zone #1.2 near #1 dist 4
It is very ugly in my opinion, does not look like smooth blending. <Screen Shot 2024-02-14 at 9.08.30 AM.png>
Maybe you could try to get a finer triangulation with other "surface" command options, but in my opinion it would probably be a waste of time. You can see the difference with sharp boundaries true vs. false:
surface sharp true
<Screen Shot 2024-02-14 at 9.08.44 AM.png>
Again, the sharper boundaries are a more true representation of the values that you are trying to show.
I hope this makes sense, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Resource for Biocomputing, Visualization, and Informatics Department of Pharmaceutical Chemistry University of California, San Francisco
On Feb 14, 2024, at 4:17 AM, Pedro Bule via ChimeraX-users <chimerax-users@cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu>> wrote:
Hello,
I am trying to color the surface of my model according to sequence conservation. I opened the MSA, associated the proper sequence with the chain I wish to color, and then used the command "color byattr seq_conservation". But this results in the colors not blending at the surface, which is not very pleasant looking. Is there a way to make the transitions between adjacent colors smoother, similar to when we color the surface, for e.g., by electrostatic potential? In legacy chimera I used to follow the "Structure... Render by Conservation" in the Multalign Viewer and then choose the attribute mavConservation, which used to do the trick
Thanks in advance!
Pedro Bule
_______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/
Dear Tom, That's awesome! Thank you so much. I have just tried your script on my model and it worked great! And it doesn't seem to introduce any mismatch between surface coloring and the underlying residues, so I'd say it does an excellent job. Thank you very much for the time you have dedicated to this Kind regards Pedro Tom Goddard <goddard@sonic.net> escreveu (sexta, 16/02/2024 à(s) 22:59):
Hi Pedro,
I was curious if I could blur the coloring of the surface across multiple triangles to get a smoother blurred appearance. So I wrote some Python code that defines a "color smooth" command and it worked decently
https://rbvi.github.io/chimerax-recipes/color_smooth/color_smooth.html
Tom
[image: bfactor_8vgz.png][image: bfactor_smooth.png]
On Feb 15, 2024, at 2:08 AM, Pedro Bule via ChimeraX-users < chimerax-users@cgl.ucsf.edu> wrote:
Dear Elaine,
Thank you very much for your quick and thorough response, which made perfect sense. You are right, it is indeed more accurate to use sharp edges. I guess that, for the intended purpose, which is to localize the active site of an enzyme, a bit of blurring wouldn't hurt, as I am not so worried about showing individual amino acid conservation rather than obtaining a nice, strongly colored patch. But in a way, I guess that is kind of cheating. I'll stick to sharp edges!
Thank you very much
Pedro
Elaine Meng <meng@cgl.ucsf.edu> escreveu (quarta, 14/02/2024 à(s) 17:11):
Hi Pedro, Coloring by conservation is a different situation than coloring by electrostatic potential. The electrostatic potential is varying continuously in space (hence the gradual change in color), whereas the conservation is a single value per residue. Thus, there is a single color for the surface patch that comes from the same residue. So what you see is the true representation of the results.
In Chimera, it looked blended only as an artifact, because the surface triangulation did not try to make sharp boundaries per atomic patch; the jagged triangles from adjacent atoms were next to each other and made the color look blended.
In ChimeraX, although you can turn off the sharp boundaries, doing that loses the linkage between the surface patch and the atom so you cannot directly color the surface by the property of the atoms (e.g. bfactor) or residues (e.g. conservation).
If you *really* want to try using the ChimeraX surfaces with sharp boundaries turned off, then you have to do a multistep process of coloring the atoms and then using color zone to color the surface to match the nearby atoms. Even if you do that, you can see the jagged edges instead of smooth color blending because the conservation is a single value per residue (not continuously varying in space like electrostatic potential).
Here is an example with coloring by bfactor, but it would be analogous to do it for conservation.
open 2gbp surface sharpBoundaries false color byattr bfactor target a color zone #1.2 near #1 dist 4
It is very ugly in my opinion, does not look like smooth blending. <Screen Shot 2024-02-14 at 9.08.30 AM.png>
Maybe you could try to get a finer triangulation with other "surface" command options, but in my opinion it would probably be a waste of time. You can see the difference with sharp boundaries true vs. false:
surface sharp true
<Screen Shot 2024-02-14 at 9.08.44 AM.png>
Again, the sharper boundaries are a more true representation of the values that you are trying to show.
I hope this makes sense, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Resource for Biocomputing, Visualization, and Informatics Department of Pharmaceutical Chemistry University of California, San Francisco
On Feb 14, 2024, at 4:17 AM, Pedro Bule via ChimeraX-users < chimerax-users@cgl.ucsf.edu> wrote:
Hello,
I am trying to color the surface of my model according to sequence conservation. I opened the MSA, associated the proper sequence with the chain I wish to color, and then used the command "color byattr seq_conservation". But this results in the colors not blending at the surface, which is not very pleasant looking. Is there a way to make the transitions between adjacent colors smoother, similar to when we color the surface, for e.g., by electrostatic potential? In legacy chimera I used to follow the "Structure... Render by Conservation" in the Multalign Viewer and then choose the attribute mavConservation, which used to do the trick
Thanks in advance!
Pedro Bule
_______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/
participants (3)
-
Elaine Meng -
Pedro Bule -
Tom Goddard