(CC-ing chimerax-users@cgl.ucsf.edu since that is the recommended address for ChimeraX questions as long as they do not include private data. Email sent to me directly has more chance of falling through the cracks.) Dear Dr. Sharyn Endow, The Matchmaker algorithm in ChimeraX is the same as in Chimera. So everything I said before about Chimera's Matchmaker is also the case in ChimeraX. If you have the "iteration" option turned on it doesn't use all aligned columns but instead prunes residue pairs for a better fit of the core. If you are showing the sequence alignment, it shows what set of positions were used for the final fit as a light orange colored box. As before, it uses only the CA-CA pairs within those columns, not the side chains. These overall RMSD values are reported in the Log. "pruned" refers only to the fit iteration process, where iteration prunes away some pairs so that they are not used in the final fit. See "Fitting" at the bottom of the "Matchmaker" help: <https://rbvi.ucsf.edu/chimerax/docs/user/tools/matchmaker.html#fitting> Or, if you are using the "matchmaker" or "mmaker" command, see the "cutoffDistance" option for the description of iteration and pruning: <https://rbvi.ucsf.edu/chimerax/docs/user/commands/matchmaker.html#options> If you mean the RMSD header across the top of the sequence alignment, that is a different thing calculated per column (not for the whole structures overall), and you can control in the Sequence Viewer Settings whether this RMSD histogram is calculated CA-only or backbone-only. Neither of them uses the sidechains. If a pairwise alignment, "CA RMSD" boils down to just the CA-CA distance. Simply selecting some range of residues on the sequence alignment doesn't change anything. See help for Sequence Viewer, its header lines, and its Settings: <https://rbvi.ucsf.edu/chimerax/docs/user/tools/sequenceviewer.html> <https://rbvi.ucsf.edu/chimerax/docs/user/tools/sequenceviewer.html#headers> <https://rbvi.ucsf.edu/chimerax/docs/user/tools/sequenceviewer.html#settings> Finally as mentioned in my previous answer about Chimera, you can use the "rmsd" command to calculate RMSDs over any sets of atoms that you choose (with or without sidechains) without changing the current positions of the structures. The more tricky part is that you have to specify in the command which atoms to pair with which atoms, since it is not calculated automatically like it is in Matchmaker. The syntax is a little different than in Chimera: there is a "to" between the two sets of atomspecs, and in ChimeraX the slash symbol "/" is used to indicate chains. "rmsd" command <https://rbvi.ucsf.edu/chimerax/docs/user/commands/rmsd.html> ChimeraX atomspec including many examples: <https://rbvi.ucsf.edu/chimerax/docs/user/commands/atomspec.html> I hope this helps, 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 Mar 23, 2024, at 8:09 AM, Sharyn Endow, Ph.D. <sharyn.endow@duke.edu> wrote:
Dear Dr Elaine Meng,
We wrote to you a few years ago regarding Chimera RMSD and you were kind enough to send the information below. We now have a few Q's regarding Chimera X Matchmaker, which displays RMSD values, e.g., when all beta-strands are selected in the Sequence Viewer window.
• We would like to know what the RMSD values correspond to after aligning two protein sequences using Matchmaker Iwhich uses only only one point per residue for fitting) and selecting all b-strands in the Sequence Viewer window. Is a single RMSD value calculated for each residue to obtain the average RMSD for the "pruned" or all atom pairs and, if so, is the RMSD value between the single point per residue that was used for fitting?
• Is there some way to obtain the C-alpha RMSD values but not the side chain values, after aligning two protein chains in Matchmaker in Chimera X?
• What does "pruned" atom pairs mean in this context?
T hank you in advance for your help -
Sincerely,
Sharyn A. Endow, PhD Professor of Cell Biology, DUMC DUS, Cell Biology PO Box 3709 / 245/247 Sands Bldg / 303 Research Drive Durham, NC 27710 USA T 919 684-4311 F 919 681-9929
From: Elaine Meng <meng@cgl.ucsf.edu> Date: January 25, 2021 at 4:26:34 PM EST To: "Sharyn Endow, Ph.D." <sharyn.endow@duke.edu> Cc: Chimera <chimera-users@cgl.ucsf.edu> Subject: Chimera RMSD Reply-To: Chimera <chimera-users@cgl.ucsf.edu>
From: "Sharyn Endow, Ph.D." <sharyn.endow@duke.edu> Subject: Chimera RMSD Date: January 23, 2021 at 11:27:17 AM PST
Thank you for your work in developing Chimera. We are trying to use Chimera RMSD to quantitate twisting of a beta-sheet and would like to know what the syntax is to list rmsd values for specified residues corresponding to a beta-strand, not just for single residues. Is it possible to list these values using RMSD? If so, we would greatly appreciate knowing how to get started.
Sincerely, Sharyn
Hi Sharyn, You can specify any sets of atoms/residues that you like in the RMSD command, as long as there are equal numbers of atoms in the two sets that you specify (from two different models). They will be paired in the order specified. We have a command-line "atom specification" syntax for that kind of thing. There is a longish page with lots of examples here: <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/frameatom_spec.html >
In your case, you may need to specify residue numbers, chain ID, and possibly atom names if you just want to use alpha-carbons or backbone as opposed to all atoms of the residues. E.g. something like
rmsd #0:12.A,57-63.A #1:14.A,59-65.A
... for all atoms of model 0 residues 12 and 57-63 in chain A vs. all atoms of model 1 residues 14 and 59-65 in chain A. Or to use CA-only or backbone only:
rmsd #0:12.A,57-63.A@ca #1:14.A,59-65.A@ca rmsd #0:12.A,57-63.A@n,ca,c,o #1:14.A,59-65.A@n,ca,c,o
Of course the numbers could be the same in the two models, I just was trying to give a more general example. I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
P.S. the chimera-users@cgl.ucsf.edu address CC'd here is recommended for general Chimera how-to or is-it-possible types of questions
Dear Dr Meng, Thank you for your reply. I am even more confused than ever about the RMSD output by Matchmaker when specific secondary elements are being compared. My Q is whether the output by Matchmaker is RMSD of the C-alphas only of the selected structural elements, if the following is done: Display two protein sequences In the Sequence Viewer, select all beta-strands for one of the two proteins Run Matchmaker by selecting Chain Pairing (and selecting the two chains for comparison) For Alignment, do not select Show pairwise sequence alignments, use default other settings For Fitting, select Iterate by pruning long atom pairs AND select Also restrict to selection The output in the log is below for my two proteins: Matchmaker NCD_crystal_refine_31_ChainA.pdb, chain A (#2) with NCD_crystal_refine_31.pdb, chain B (#1), sequence alignment score = 305.9 RMSD between 70 pruned atom pairs is 0.717 angstroms; (across all 80 pairs: 1.161) Again, my Q is whether the output by Matchmaker is RMSD of the C-alphas only of the selected structural elements or, if not, what does the RMSD = 0.717 for the 70 pruned atom pairs and 1.161 for the 80 atom pairs correspond to? Is it any atom that was used by Matchmaker for pairing the two sequences, or the same atoms in corresponding residues in the two protein chains? Thank you in advance for a further clarification - Sharyn Endow Sharyn A. Endow, PhD Professor of Cell Biology, DUMC DUS, Cell Biology PO Box 3709 / 245/247 Sands Bldg / 303 Research Drive Durham, NC 27710 USA T 919 684-4311 F 919 681-9929 ________________________________ From: Elaine Meng <meng@cgl.ucsf.edu> Sent: Sunday, March 24, 2024 2:32 PM To: Sharyn Endow, Ph.D. <sharyn.endow@duke.edu> Cc: ChimeraX Users Help <chimerax-users@cgl.ucsf.edu> Subject: Re: Chimera X Matchmaker RMSD (CC-ing chimerax-users@cgl.ucsf.edu since that is the recommended address for ChimeraX questions as long as they do not include private data. Email sent to me directly has more chance of falling through the cracks.) Dear Dr. Sharyn Endow, The Matchmaker algorithm in ChimeraX is the same as in Chimera. So everything I said before about Chimera's Matchmaker is also the case in ChimeraX. If you have the "iteration" option turned on it doesn't use all aligned columns but instead prunes residue pairs for a better fit of the core. If you are showing the sequence alignment, it shows what set of positions were used for the final fit as a light orange colored box. As before, it uses only the CA-CA pairs within those columns, not the side chains. These overall RMSD values are reported in the Log. "pruned" refers only to the fit iteration process, where iteration prunes away some pairs so that they are not used in the final fit. See "Fitting" at the bottom of the "Matchmaker" help: <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/tools/m... > Or, if you are using the "matchmaker" or "mmaker" command, see the "cutoffDistance" option for the description of iteration and pruning: <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/command... > If you mean the RMSD header across the top of the sequence alignment, that is a different thing calculated per column (not for the whole structures overall), and you can control in the Sequence Viewer Settings whether this RMSD histogram is calculated CA-only or backbone-only. Neither of them uses the sidechains. If a pairwise alignment, "CA RMSD" boils down to just the CA-CA distance. Simply selecting some range of residues on the sequence alignment doesn't change anything. See help for Sequence Viewer, its header lines, and its Settings: <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/tools/s... > <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/tools/s... > <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/tools/s... > Finally as mentioned in my previous answer about Chimera, you can use the "rmsd" command to calculate RMSDs over any sets of atoms that you choose (with or without sidechains) without changing the current positions of the structures. The more tricky part is that you have to specify in the command which atoms to pair with which atoms, since it is not calculated automatically like it is in Matchmaker. The syntax is a little different than in Chimera: there is a "to" between the two sets of atomspecs, and in ChimeraX the slash symbol "/" is used to indicate chains. "rmsd" command <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/command... > ChimeraX atomspec including many examples: <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/command... > I hope this helps, 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 Mar 23, 2024, at 8:09 AM, Sharyn Endow, Ph.D. <sharyn.endow@duke.edu> wrote:
Dear Dr Elaine Meng,
We wrote to you a few years ago regarding Chimera RMSD and you were kind enough to send the information below. We now have a few Q's regarding Chimera X Matchmaker, which displays RMSD values, e.g., when all beta-strands are selected in the Sequence Viewer window.
• We would like to know what the RMSD values correspond to after aligning two protein sequences using Matchmaker Iwhich uses only only one point per residue for fitting) and selecting all b-strands in the Sequence Viewer window. Is a single RMSD value calculated for each residue to obtain the average RMSD for the "pruned" or all atom pairs and, if so, is the RMSD value between the single point per residue that was used for fitting?
• Is there some way to obtain the C-alpha RMSD values but not the side chain values, after aligning two protein chains in Matchmaker in Chimera X?
• What does "pruned" atom pairs mean in this context?
T hank you in advance for your help -
Sincerely,
Sharyn A. Endow, PhD Professor of Cell Biology, DUMC DUS, Cell Biology PO Box 3709 / 245/247 Sands Bldg / 303 Research Drive Durham, NC 27710 USA T 919 684-4311 F 919 681-9929
From: Elaine Meng <meng@cgl.ucsf.edu> Date: January 25, 2021 at 4:26:34 PM EST To: "Sharyn Endow, Ph.D." <sharyn.endow@duke.edu> Cc: Chimera <chimera-users@cgl.ucsf.edu> Subject: Chimera RMSD Reply-To: Chimera <chimera-users@cgl.ucsf.edu>
From: "Sharyn Endow, Ph.D." <sharyn.endow@duke.edu> Subject: Chimera RMSD Date: January 23, 2021 at 11:27:17 AM PST
Thank you for your work in developing Chimera. We are trying to use Chimera RMSD to quantitate twisting of a beta-sheet and would like to know what the syntax is to list rmsd values for specified residues corresponding to a beta-strand, not just for single residues. Is it possible to list these values using RMSD? If so, we would greatly appreciate knowing how to get started.
Sincerely, Sharyn
Hi Sharyn, You can specify any sets of atoms/residues that you like in the RMSD command, as long as there are equal numbers of atoms in the two sets that you specify (from two different models). They will be paired in the order specified. We have a command-line "atom specification" syntax for that kind of thing. There is a longish page with lots of examples here: <https://urldefense.com/v3/__https://www.rbvi.ucsf.edu/chimera/docs/UsersGuid... >
In your case, you may need to specify residue numbers, chain ID, and possibly atom names if you just want to use alpha-carbons or backbone as opposed to all atoms of the residues. E.g. something like
rmsd #0:12.A,57-63.A #1:14.A,59-65.A
... for all atoms of model 0 residues 12 and 57-63 in chain A vs. all atoms of model 1 residues 14 and 59-65 in chain A. Or to use CA-only or backbone only:
rmsd #0:12.A,57-63.A@ca #1:14.A,59-65.A@ca rmsd #0:12.A,57-63.A@n,ca,c,o #1:14.A,59-65.A@n,ca,c,o
Of course the numbers could be the same in the two models, I just was trying to give a more general example. I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
P.S. the chimera-users@cgl.ucsf.edu address CC'd here is recommended for general Chimera how-to or is-it-possible types of questions
Matchmaker fitting uses 1 point per residue, exactly the same as in Chimera. If you look in the Log after running Matchmaker it says how many CA-CA pairs were used in the fit and thus in the reported RMSD. So if you made the selection and also turned on "Also restrict to selection" before running Matchmaker, then this smaller number of CA-CA pairs (compared to the larger number of pairs if you had used the whole structures) used for fitting should be reported in the Log. Eg if the Log says something like this after running Matchmaker: RMSD between 147 pruned atom pairs is 0.425 angstroms; (across all 305 pairs: 7.702) It means that 147 columns in the sequence alignment were used in the final fit and the RMSD over these 147 pairs of atoms (147 CA-CA pairs) was 0.425 angstroms. If you consider all the columns in the alignment where both structures have a residue (not gap) and in the case of the "restrict to selection" option, these residues were selected, then the RMSD over those 305 CA-CA pairs was 7.701. You can look at the number of pairs reported and see if that agrees with your expectation compared to the parts you selected to use for fitting. It would be very helpful to turn on showing the sequence alignment because then you can confirm which parts were used for the final fit and thus the reported RMSD, since they will be enclosed in a light orange box. Then you can close the sequence alignment window if you don't need it any more. If you don't want fit iteration (don't want pruning) and you simply want to use ALL of the selected parts in the fitting, then turn off the fit iteration option in the "Fitting" section. Elaine
On Mar 24, 2024, at 1:09 PM, Sharyn Endow, Ph.D. <sharyn.endow@duke.edu> wrote:
Dear Dr Meng,
Thank you for your reply. I am even more confused than ever about the RMSD output by Matchmaker when specific secondary elements are being compared. My Q is whether the output by Matchmaker is RMSD of the C-alphas only of the selected structural elements, if the following is done:
Display two protein sequences In the Sequence Viewer, select all beta-strands for one of the two proteins Run Matchmaker by selecting Chain Pairing (and selecting the two chains for comparison) For Alignment, do not select Show pairwise sequence alignments, use default other settings For Fitting, select Iterate by pruning long atom pairs AND select Also restrict to selection
The output in the log is below for my two proteins:
Matchmaker NCD_crystal_refine_31_ChainA.pdb, chain A (#2) with NCD_crystal_refine_31.pdb, chain B (#1), sequence alignment score = 305.9 RMSD between 70 pruned atom pairs is 0.717 angstroms; (across all 80 pairs: 1.161)
Again, my Q is whether the output by Matchmaker is RMSD of the C-alphas only of the selected structural elementsor, if not, what does the RMSD = 0.717 for the 70 pruned atom pairs and 1.161 for the 80 atom pairs correspond to?Is it any atom that was used by Matchmaker for pairing the two sequences, or the same atoms in corresponding residues in the two protein chains?
Thank you in advance for a further clarification -
Sharyn Endow
Sharyn A. Endow, PhD Professor of Cell Biology, DUMC DUS, Cell Biology PO Box 3709 / 245/247 Sands Bldg / 303 Research Drive Durham, NC 27710 USA T 919 684-4311 F 919 681-9929
From: Elaine Meng <meng@cgl.ucsf.edu> Sent: Sunday, March 24, 2024 2:32 PM To: Sharyn Endow, Ph.D. <sharyn.endow@duke.edu> Cc: ChimeraX Users Help <chimerax-users@cgl.ucsf.edu> Subject: Re: Chimera X Matchmaker RMSD
(CC-ing chimerax-users@cgl.ucsf.edu since that is the recommended address for ChimeraX questions as long as they do not include private data. Email sent to me directly has more chance of falling through the cracks.)
Dear Dr. Sharyn Endow, The Matchmaker algorithm in ChimeraX is the same as in Chimera. So everything I said before about Chimera's Matchmaker is also the case in ChimeraX.
If you have the "iteration" option turned on it doesn't use all aligned columns but instead prunes residue pairs for a better fit of the core. If you are showing the sequence alignment, it shows what set of positions were used for the final fit as a light orange colored box. As before, it uses only the CA-CA pairs within those columns, not the side chains. These overall RMSD values are reported in the Log. "pruned" refers only to the fit iteration process, where iteration prunes away some pairs so that they are not used in the final fit.
See "Fitting" at the bottom of the "Matchmaker" help: <https://rbvi.ucsf.edu/chimerax/docs/user/tools/matchmaker.html#fitting >
Or, if you are using the "matchmaker" or "mmaker" command, see the "cutoffDistance" option for the description of iteration and pruning: <https://rbvi.ucsf.edu/chimerax/docs/user/commands/matchmaker.html#options >
If you mean the RMSD header across the top of the sequence alignment, that is a different thing calculated per column (not for the whole structures overall), and you can control in the Sequence Viewer Settings whether this RMSD histogram is calculated CA-only or backbone-only. Neither of them uses the sidechains. If a pairwise alignment, "CA RMSD" boils down to just the CA-CA distance. Simply selecting some range of residues on the sequence alignment doesn't change anything.
See help for Sequence Viewer, its header lines, and its Settings: <https://rbvi.ucsf.edu/chimerax/docs/user/tools/sequenceviewer.html > <https://rbvi.ucsf.edu/chimerax/docs/user/tools/sequenceviewer.html#headers > <https://rbvi.ucsf.edu/chimerax/docs/user/tools/sequenceviewer.html#settings >
Finally as mentioned in my previous answer about Chimera, you can use the "rmsd" command to calculate RMSDs over any sets of atoms that you choose (with or without sidechains) without changing the current positions of the structures. The more tricky part is that you have to specify in the command which atoms to pair with which atoms, since it is not calculated automatically like it is in Matchmaker. The syntax is a little different than in Chimera: there is a "to" between the two sets of atomspecs, and in ChimeraX the slash symbol "/" is used to indicate chains.
"rmsd" command <https://rbvi.ucsf.edu/chimerax/docs/user/commands/rmsd.html >
ChimeraX atomspec including many examples: <https://rbvi.ucsf.edu/chimerax/docs/user/commands/atomspec.html >
I hope this helps, 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 Mar 23, 2024, at 8:09 AM, Sharyn Endow, Ph.D. <sharyn.endow@duke.edu> wrote:
Dear Dr Elaine Meng,
We wrote to you a few years ago regarding Chimera RMSD and you were kind enough to send the information below. We now have a few Q's regarding Chimera X Matchmaker, which displays RMSD values, e.g., when all beta-strands are selected in the Sequence Viewer window.
• We would like to know what the RMSD values correspond to after aligning two protein sequences using Matchmaker Iwhich uses only only one point per residue for fitting) and selecting all b-strands in the Sequence Viewer window. Is a single RMSD value calculated for each residue to obtain the average RMSD for the "pruned" or all atom pairs and, if so, is the RMSD value between the single point per residue that was used for fitting?
• Is there some way to obtain the C-alpha RMSD values but not the side chain values, after aligning two protein chains in Matchmaker in Chimera X?
• What does "pruned" atom pairs mean in this context?
T hank you in advance for your help -
Sincerely,
Sharyn A. Endow, PhD Professor of Cell Biology, DUMC DUS, Cell Biology PO Box 3709 / 245/247 Sands Bldg / 303 Research Drive Durham, NC 27710 USA T 919 684-4311 F 919 681-9929
From: Elaine Meng <meng@cgl.ucsf.edu> Date: January 25, 2021 at 4:26:34 PM EST To: "Sharyn Endow, Ph.D." <sharyn.endow@duke.edu> Cc: Chimera <chimera-users@cgl.ucsf.edu> Subject: Chimera RMSD Reply-To: Chimera <chimera-users@cgl.ucsf.edu>
From: "Sharyn Endow, Ph.D." <sharyn.endow@duke.edu> Subject: Chimera RMSD Date: January 23, 2021 at 11:27:17 AM PST
Thank you for your work in developing Chimera. We are trying to use Chimera RMSD to quantitate twisting of a beta-sheet and would like to know what the syntax is to list rmsd values for specified residues corresponding to a beta-strand, not just for single residues. Is it possible to list these values using RMSD? If so, we would greatly appreciate knowing how to get started.
Sincerely, Sharyn
Hi Sharyn, You can specify any sets of atoms/residues that you like in the RMSD command, as long as there are equal numbers of atoms in the two sets that you specify (from two different models). They will be paired in the order specified. We have a command-line "atom specification" syntax for that kind of thing. There is a longish page with lots of examples here: <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/frameatom_spec.html >
In your case, you may need to specify residue numbers, chain ID, and possibly atom names if you just want to use alpha-carbons or backbone as opposed to all atoms of the residues. E.g. something like
rmsd #0:12.A,57-63.A #1:14.A,59-65.A
... for all atoms of model 0 residues 12 and 57-63 in chain A vs. all atoms of model 1 residues 14 and 59-65 in chain A. Or to use CA-only or backbone only:
rmsd #0:12.A,57-63.A@ca #1:14.A,59-65.A@ca rmsd #0:12.A,57-63.A@n,ca,c,o #1:14.A,59-65.A@n,ca,c,o
Of course the numbers could be the same in the two models, I just was trying to give a more general example. I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
P.S. the chimera-users@cgl.ucsf.edu address CC'd here is recommended for general Chimera how-to or is-it-possible types of questions
Thank you - As I understand it, and from turning on the Show sequence alignment, the RMSD values correspond to the Calphas that were used in the pairwise fit. But let me know if this NOT the case. This clarifies the major Q that I was posing and was confused by in the previous? Matchmaker documentation stating that ANY atom is used for the alignment, not just the protein Calphas. Thank you again - Sharyn Sharyn A. Endow, PhD Professor of Cell Biology, DUMC DUS, Cell Biology PO Box 3709 / 245/247 Sands Bldg / 303 Research Drive Durham, NC 27710 USA T 919 684-4311 F 919 681-9929 ________________________________ From: Elaine Meng <meng@cgl.ucsf.edu> Sent: Sunday, March 24, 2024 4:35 PM To: Sharyn Endow, Ph.D. <sharyn.endow@duke.edu> Cc: ChimeraX Users Help <chimerax-users@cgl.ucsf.edu> Subject: Re: Chimera X Matchmaker RMSD Matchmaker fitting uses 1 point per residue, exactly the same as in Chimera. If you look in the Log after running Matchmaker it says how many CA-CA pairs were used in the fit and thus in the reported RMSD. So if you made the selection and also turned on "Also restrict to selection" before running Matchmaker, then this smaller number of CA-CA pairs (compared to the larger number of pairs if you had used the whole structures) used for fitting should be reported in the Log. Eg if the Log says something like this after running Matchmaker: RMSD between 147 pruned atom pairs is 0.425 angstroms; (across all 305 pairs: 7.702) It means that 147 columns in the sequence alignment were used in the final fit and the RMSD over these 147 pairs of atoms (147 CA-CA pairs) was 0.425 angstroms. If you consider all the columns in the alignment where both structures have a residue (not gap) and in the case of the "restrict to selection" option, these residues were selected, then the RMSD over those 305 CA-CA pairs was 7.701. You can look at the number of pairs reported and see if that agrees with your expectation compared to the parts you selected to use for fitting. It would be very helpful to turn on showing the sequence alignment because then you can confirm which parts were used for the final fit and thus the reported RMSD, since they will be enclosed in a light orange box. Then you can close the sequence alignment window if you don't need it any more. If you don't want fit iteration (don't want pruning) and you simply want to use ALL of the selected parts in the fitting, then turn off the fit iteration option in the "Fitting" section. Elaine
On Mar 24, 2024, at 1:09 PM, Sharyn Endow, Ph.D. <sharyn.endow@duke.edu> wrote:
Dear Dr Meng,
Thank you for your reply. I am even more confused than ever about the RMSD output by Matchmaker when specific secondary elements are being compared. My Q is whether the output by Matchmaker is RMSD of the C-alphas only of the selected structural elements, if the following is done:
Display two protein sequences In the Sequence Viewer, select all beta-strands for one of the two proteins Run Matchmaker by selecting Chain Pairing (and selecting the two chains for comparison) For Alignment, do not select Show pairwise sequence alignments, use default other settings For Fitting, select Iterate by pruning long atom pairs AND select Also restrict to selection
The output in the log is below for my two proteins:
Matchmaker NCD_crystal_refine_31_ChainA.pdb, chain A (#2) with NCD_crystal_refine_31.pdb, chain B (#1), sequence alignment score = 305.9 RMSD between 70 pruned atom pairs is 0.717 angstroms; (across all 80 pairs: 1.161)
Again, my Q is whether the output by Matchmaker is RMSD of the C-alphas only of the selected structural elementsor, if not, what does the RMSD = 0.717 for the 70 pruned atom pairs and 1.161 for the 80 atom pairs correspond to?Is it any atom that was used by Matchmaker for pairing the two sequences, or the same atoms in corresponding residues in the two protein chains?
Thank you in advance for a further clarification -
Sharyn Endow
Sharyn A. Endow, PhD Professor of Cell Biology, DUMC DUS, Cell Biology PO Box 3709 / 245/247 Sands Bldg / 303 Research Drive Durham, NC 27710 USA T 919 684-4311 F 919 681-9929
From: Elaine Meng <meng@cgl.ucsf.edu> Sent: Sunday, March 24, 2024 2:32 PM To: Sharyn Endow, Ph.D. <sharyn.endow@duke.edu> Cc: ChimeraX Users Help <chimerax-users@cgl.ucsf.edu> Subject: Re: Chimera X Matchmaker RMSD
(CC-ing chimerax-users@cgl.ucsf.edu since that is the recommended address for ChimeraX questions as long as they do not include private data. Email sent to me directly has more chance of falling through the cracks.)
Dear Dr. Sharyn Endow, The Matchmaker algorithm in ChimeraX is the same as in Chimera. So everything I said before about Chimera's Matchmaker is also the case in ChimeraX.
If you have the "iteration" option turned on it doesn't use all aligned columns but instead prunes residue pairs for a better fit of the core. If you are showing the sequence alignment, it shows what set of positions were used for the final fit as a light orange colored box. As before, it uses only the CA-CA pairs within those columns, not the side chains. These overall RMSD values are reported in the Log. "pruned" refers only to the fit iteration process, where iteration prunes away some pairs so that they are not used in the final fit.
See "Fitting" at the bottom of the "Matchmaker" help: <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/tools/m... >
Or, if you are using the "matchmaker" or "mmaker" command, see the "cutoffDistance" option for the description of iteration and pruning: <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/command... >
If you mean the RMSD header across the top of the sequence alignment, that is a different thing calculated per column (not for the whole structures overall), and you can control in the Sequence Viewer Settings whether this RMSD histogram is calculated CA-only or backbone-only. Neither of them uses the sidechains. If a pairwise alignment, "CA RMSD" boils down to just the CA-CA distance. Simply selecting some range of residues on the sequence alignment doesn't change anything.
See help for Sequence Viewer, its header lines, and its Settings: <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/tools/s... > <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/tools/s... > <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/tools/s... >
Finally as mentioned in my previous answer about Chimera, you can use the "rmsd" command to calculate RMSDs over any sets of atoms that you choose (with or without sidechains) without changing the current positions of the structures. The more tricky part is that you have to specify in the command which atoms to pair with which atoms, since it is not calculated automatically like it is in Matchmaker. The syntax is a little different than in Chimera: there is a "to" between the two sets of atomspecs, and in ChimeraX the slash symbol "/" is used to indicate chains.
"rmsd" command <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/command... >
ChimeraX atomspec including many examples: <https://urldefense.com/v3/__https://rbvi.ucsf.edu/chimerax/docs/user/command... >
I hope this helps, 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 Mar 23, 2024, at 8:09 AM, Sharyn Endow, Ph.D. <sharyn.endow@duke.edu> wrote:
Dear Dr Elaine Meng,
We wrote to you a few years ago regarding Chimera RMSD and you were kind enough to send the information below. We now have a few Q's regarding Chimera X Matchmaker, which displays RMSD values, e.g., when all beta-strands are selected in the Sequence Viewer window.
• We would like to know what the RMSD values correspond to after aligning two protein sequences using Matchmaker Iwhich uses only only one point per residue for fitting) and selecting all b-strands in the Sequence Viewer window. Is a single RMSD value calculated for each residue to obtain the average RMSD for the "pruned" or all atom pairs and, if so, is the RMSD value between the single point per residue that was used for fitting?
• Is there some way to obtain the C-alpha RMSD values but not the side chain values, after aligning two protein chains in Matchmaker in Chimera X?
• What does "pruned" atom pairs mean in this context?
T hank you in advance for your help -
Sincerely,
Sharyn A. Endow, PhD Professor of Cell Biology, DUMC DUS, Cell Biology PO Box 3709 / 245/247 Sands Bldg / 303 Research Drive Durham, NC 27710 USA T 919 684-4311 F 919 681-9929
From: Elaine Meng <meng@cgl.ucsf.edu> Date: January 25, 2021 at 4:26:34 PM EST To: "Sharyn Endow, Ph.D." <sharyn.endow@duke.edu> Cc: Chimera <chimera-users@cgl.ucsf.edu> Subject: Chimera RMSD Reply-To: Chimera <chimera-users@cgl.ucsf.edu>
From: "Sharyn Endow, Ph.D." <sharyn.endow@duke.edu> Subject: Chimera RMSD Date: January 23, 2021 at 11:27:17 AM PST
Thank you for your work in developing Chimera. We are trying to use Chimera RMSD to quantitate twisting of a beta-sheet and would like to know what the syntax is to list rmsd values for specified residues corresponding to a beta-strand, not just for single residues. Is it possible to list these values using RMSD? If so, we would greatly appreciate knowing how to get started.
Sincerely, Sharyn
Hi Sharyn, You can specify any sets of atoms/residues that you like in the RMSD command, as long as there are equal numbers of atoms in the two sets that you specify (from two different models). They will be paired in the order specified. We have a command-line "atom specification" syntax for that kind of thing. There is a longish page with lots of examples here: <https://urldefense.com/v3/__https://www.rbvi.ucsf.edu/chimera/docs/UsersGuid... >
In your case, you may need to specify residue numbers, chain ID, and possibly atom names if you just want to use alpha-carbons or backbone as opposed to all atoms of the residues. E.g. something like
rmsd #0:12.A,57-63.A #1:14.A,59-65.A
... for all atoms of model 0 residues 12 and 57-63 in chain A vs. all atoms of model 1 residues 14 and 59-65 in chain A. Or to use CA-only or backbone only:
rmsd #0:12.A,57-63.A@ca #1:14.A,59-65.A@ca rmsd #0:12.A,57-63.A@n,ca,c,o #1:14.A,59-65.A@n,ca,c,o
Of course the numbers could be the same in the two models, I just was trying to give a more general example. I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
P.S. the chimera-users@cgl.ucsf.edu address CC'd here is recommended for general Chimera how-to or is-it-possible types of questions
The two RMSD values that Matchmaker gives in the Log are based on (1) CA-CA pairs used in the fit, and (2) overall (all CA-CA pairs in columns of the alignment that have a residue from each structure, i.e. neither is a gap). In my example Log message 147 pairs were used in the fit, and the overall alignment had 305 pairs, and that's why you get two RMSD values:
RMSD between 147 pruned atom pairs is 0.425 angstroms; (across all 305 pairs: 7.702)
This is true regardless of whether or not you actually display the sequence alignment. I merely suggested you should show the sequence alignment because it will display a light orange box around the positions that were used in the final fit, to make it easier for you see whether they are the positions that you expected or not. If you want to use any other atoms for fitting you would use the "match" command in Chimera or the "align" command in ChimeraX. The "match" command in Chimera is NOT the same as the "matchmaker" command. Maybe that is where the confusion arose. To report RMSD using any atoms (without fitting or moving anything) you would use the "rmsd" command which is in both programs. 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 Mar 24, 2024, at 1:42 PM, Sharyn Endow, Ph.D. via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:
Thank you -
As I understand it, and from turning on the Show sequence alignment, the RMSD values correspond to the Calphas that were used in the pairwise fit. But let me know if this NOT the case.
This clarifies the major Q that I was posing and was confused by in the previous? Matchmaker documentation stating that ANY atom is used for the alignment, not just the protein Calphas.
Thank you again -
Sharyn
Sharyn A. Endow, PhD Professor of Cell Biology, DUMC DUS, Cell Biology PO Box 3709 / 245/247 Sands Bldg / 303 Research Drive Durham, NC 27710 USA T 919 684-4311 F 919 681-9929
From: Elaine Meng <meng@cgl.ucsf.edu> Sent: Sunday, March 24, 2024 4:35 PM To: Sharyn Endow, Ph.D. <sharyn.endow@duke.edu> Cc: ChimeraX Users Help <chimerax-users@cgl.ucsf.edu> Subject: Re: Chimera X Matchmaker RMSD
Matchmaker fitting uses 1 point per residue, exactly the same as in Chimera. If you look in the Log after running Matchmaker it says how many CA-CA pairs were used in the fit and thus in the reported RMSD. So if you made the selection and also turned on "Also restrict to selection" before running Matchmaker, then this smaller number of CA-CA pairs (compared to the larger number of pairs if you had used the whole structures) used for fitting should be reported in the Log.
Eg if the Log says something like this after running Matchmaker:
RMSD between 147 pruned atom pairs is 0.425 angstroms; (across all 305 pairs: 7.702)
It means that 147 columns in the sequence alignment were used in the final fit and the RMSD over these 147 pairs of atoms (147 CA-CA pairs) was 0.425 angstroms. If you consider all the columns in the alignment where both structures have a residue (not gap) and in the case of the "restrict to selection" option, these residues were selected, then the RMSD over those 305 CA-CA pairs was 7.701.
You can look at the number of pairs reported and see if that agrees with your expectation compared to the parts you selected to use for fitting.
It would be very helpful to turn on showing the sequence alignment because then you can confirm which parts were used for the final fit and thus the reported RMSD, since they will be enclosed in a light orange box. Then you can close the sequence alignment window if you don't need it any more.
If you don't want fit iteration (don't want pruning) and you simply want to use ALL of the selected parts in the fitting, then turn off the fit iteration option in the "Fitting" section.
Elaine
On Mar 24, 2024, at 1:09 PM, Sharyn Endow, Ph.D. <sharyn.endow@duke.edu> wrote:
Dear Dr Meng,
Thank you for your reply. I am even more confused than ever about the RMSD output by Matchmaker when specific secondary elements are being compared. My Q is whether the output by Matchmaker is RMSD of the C-alphas only of the selected structural elements, if the following is done:
Display two protein sequences In the Sequence Viewer, select all beta-strands for one of the two proteins Run Matchmaker by selecting Chain Pairing (and selecting the two chains for comparison) For Alignment, do not select Show pairwise sequence alignments, use default other settings For Fitting, select Iterate by pruning long atom pairs AND select Also restrict to selection
The output in the log is below for my two proteins:
Matchmaker NCD_crystal_refine_31_ChainA.pdb, chain A (#2) with NCD_crystal_refine_31.pdb, chain B (#1), sequence alignment score = 305.9 RMSD between 70 pruned atom pairs is 0.717 angstroms; (across all 80 pairs: 1.161)
Again, my Q is whether the output by Matchmaker is RMSD of the C-alphas only of the selected structural elementsor, if not, what does the RMSD = 0.717 for the 70 pruned atom pairs and 1.161 for the 80 atom pairs correspond to?Is it any atom that was used by Matchmaker for pairing the two sequences, or the same atoms in corresponding residues in the two protein chains?
Thank you in advance for a further clarification -
Sharyn Endow
Sharyn A. Endow, PhD Professor of Cell Biology, DUMC DUS, Cell Biology PO Box 3709 / 245/247 Sands Bldg / 303 Research Drive Durham, NC 27710 USA T 919 684-4311 F 919 681-9929
From: Elaine Meng <meng@cgl.ucsf.edu> Sent: Sunday, March 24, 2024 2:32 PM To: Sharyn Endow, Ph.D. <sharyn.endow@duke.edu> Cc: ChimeraX Users Help <chimerax-users@cgl.ucsf.edu> Subject: Re: Chimera X Matchmaker RMSD
(CC-ing chimerax-users@cgl.ucsf.edu since that is the recommended address for ChimeraX questions as long as they do not include private data. Email sent to me directly has more chance of falling through the cracks.)
Dear Dr. Sharyn Endow, The Matchmaker algorithm in ChimeraX is the same as in Chimera. So everything I said before about Chimera's Matchmaker is also the case in ChimeraX.
If you have the "iteration" option turned on it doesn't use all aligned columns but instead prunes residue pairs for a better fit of the core. If you are showing the sequence alignment, it shows what set of positions were used for the final fit as a light orange colored box. As before, it uses only the CA-CA pairs within those columns, not the side chains. These overall RMSD values are reported in the Log. "pruned" refers only to the fit iteration process, where iteration prunes away some pairs so that they are not used in the final fit.
See "Fitting" at the bottom of the "Matchmaker" help: <https://rbvi.ucsf.edu/chimerax/docs/user/tools/matchmaker.html#fitting >
Or, if you are using the "matchmaker" or "mmaker" command, see the "cutoffDistance" option for the description of iteration and pruning: <https://rbvi.ucsf.edu/chimerax/docs/user/commands/matchmaker.html#options >
If you mean the RMSD header across the top of the sequence alignment, that is a different thing calculated per column (not for the whole structures overall), and you can control in the Sequence Viewer Settings whether this RMSD histogram is calculated CA-only or backbone-only. Neither of them uses the sidechains. If a pairwise alignment, "CA RMSD" boils down to just the CA-CA distance. Simply selecting some range of residues on the sequence alignment doesn't change anything.
See help for Sequence Viewer, its header lines, and its Settings: <https://rbvi.ucsf.edu/chimerax/docs/user/tools/sequenceviewer.html > <https://rbvi.ucsf.edu/chimerax/docs/user/tools/sequenceviewer.html#headers > <https://rbvi.ucsf.edu/chimerax/docs/user/tools/sequenceviewer.html#settings >
Finally as mentioned in my previous answer about Chimera, you can use the "rmsd" command to calculate RMSDs over any sets of atoms that you choose (with or without sidechains) without changing the current positions of the structures. The more tricky part is that you have to specify in the command which atoms to pair with which atoms, since it is not calculated automatically like it is in Matchmaker. The syntax is a little different than in Chimera: there is a "to" between the two sets of atomspecs, and in ChimeraX the slash symbol "/" is used to indicate chains.
"rmsd" command <https://rbvi.ucsf.edu/chimerax/docs/user/commands/rmsd.html >
ChimeraX atomspec including many examples: <https://rbvi.ucsf.edu/chimerax/docs/user/commands/atomspec.html >
I hope this helps, 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 Mar 23, 2024, at 8:09 AM, Sharyn Endow, Ph.D. <sharyn.endow@duke.edu> wrote:
Dear Dr Elaine Meng,
We wrote to you a few years ago regarding Chimera RMSD and you were kind enough to send the information below. We now have a few Q's regarding Chimera X Matchmaker, which displays RMSD values, e.g., when all beta-strands are selected in the Sequence Viewer window.
• We would like to know what the RMSD values correspond to after aligning two protein sequences using Matchmaker Iwhich uses only only one point per residue for fitting) and selecting all b-strands in the Sequence Viewer window. Is a single RMSD value calculated for each residue to obtain the average RMSD for the "pruned" or all atom pairs and, if so, is the RMSD value between the single point per residue that was used for fitting?
• Is there some way to obtain the C-alpha RMSD values but not the side chain values, after aligning two protein chains in Matchmaker in Chimera X?
• What does "pruned" atom pairs mean in this context?
T hank you in advance for your help -
Sincerely,
Sharyn A. Endow, PhD Professor of Cell Biology, DUMC DUS, Cell Biology PO Box 3709 / 245/247 Sands Bldg / 303 Research Drive Durham, NC 27710 USA T 919 684-4311 F 919 681-9929
From: Elaine Meng <meng@cgl.ucsf.edu> Date: January 25, 2021 at 4:26:34 PM EST To: "Sharyn Endow, Ph.D." <sharyn.endow@duke.edu> Cc: Chimera <chimera-users@cgl.ucsf.edu> Subject: Chimera RMSD Reply-To: Chimera <chimera-users@cgl.ucsf.edu>
From: "Sharyn Endow, Ph.D." <sharyn.endow@duke.edu> Subject: Chimera RMSD Date: January 23, 2021 at 11:27:17 AM PST
Thank you for your work in developing Chimera. We are trying to use Chimera RMSD to quantitate twisting of a beta-sheet and would like to know what the syntax is to list rmsd values for specified residues corresponding to a beta-strand, not just for single residues. Is it possible to list these values using RMSD? If so, we would greatly appreciate knowing how to get started.
Sincerely, Sharyn
Hi Sharyn, You can specify any sets of atoms/residues that you like in the RMSD command, as long as there are equal numbers of atoms in the two sets that you specify (from two different models). They will be paired in the order specified. We have a command-line "atom specification" syntax for that kind of thing. There is a longish page with lots of examples here: <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/frameatom_spec.html >
In your case, you may need to specify residue numbers, chain ID, and possibly atom names if you just want to use alpha-carbons or backbone as opposed to all atoms of the residues. E.g. something like
rmsd #0:12.A,57-63.A #1:14.A,59-65.A
... for all atoms of model 0 residues 12 and 57-63 in chain A vs. all atoms of model 1 residues 14 and 59-65 in chain A. Or to use CA-only or backbone only:
rmsd #0:12.A,57-63.A@ca #1:14.A,59-65.A@ca rmsd #0:12.A,57-63.A@n,ca,c,o #1:14.A,59-65.A@n,ca,c,o
Of course the numbers could be the same in the two models, I just was trying to give a more general example. I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
P.S. the chimera-users@cgl.ucsf.edu address CC'd here is recommended for general Chimera how-to or is-it-possible types of questions
_______________________________________________ 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 (2)
-
Elaine Meng -
Sharyn Endow, Ph.D.