How to align two molecules using a subset of atoms in ChimeraX?
Dear ChimeraX support team, I tried the following command first - it worked well. ================== match #1/A to #2/A ================== Then I tried to align the two molecules using parts of them - it didn't work and complained that "No chains in reference structure PROT.pdb #2 compatible with BLOSUM-62 similarity matrix." ================== matchmaker #1/A:10-20@CA to #2/B:10-20@CA ================== System Info: ================== Platform: Linux-3.10.0-862.el7.x86_64-x86_64-with-redhat-7.5-Maipo ChimeraX version: 0.9 (2019-02-05) ================== Thank a bunch! Steven -- Steve Chou
Hi Steve, I couldn’t tell exactly without having those structures, but did you really mean chain B of model #2? Your first command (the one that works) has chain A of both models. However, if you are just trying to match exactly those atoms, use the “align” command, not matchmaker. For example, something like align #1/A:10-20@ca to #2/A:10-20@ca <http://rbvi.ucsf.edu/chimerax/docs/user/commands/align.html> Align uses the atoms you tell it to use and pairs them in order. Matchmaker figures out the residue pairing automatically by making a sequence alignment… this is especially useful for aligning structures that happen to have different residue numbering, homologs with different sequence lengths, etc. When if figures out the pairing, it automatically only uses CA atoms of each pair, so you wouldn’t need to give them in the command. <http://rbvi.ucsf.edu/chimerax/docs/user/commands/matchmaker.html> So generally, one would just specify the whole chain and not give specific residues with “matchmaker” except to suppress using parts of the structure you want to exclude from the sequence alignment. Nevertheless, the error message is unexpected so I will report it as a bug. I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Feb 8, 2019, at 12:46 PM, Steve Chou <stevezchou@gmail.com> wrote:
Dear ChimeraX support team, I tried the following command first - it worked well. ================== match #1/A to #2/A ================== Then I tried to align the two molecules using parts of them - it didn't work and complained that "No chains in reference structure PROT.pdb #2 compatible with BLOSUM-62 similarity matrix." ================== matchmaker #1/A:10-20@CA to #2/B:10-20@CA ================== System Info: ================== Platform: Linux-3.10.0-862.el7.x86_64-x86_64-with-redhat-7.5-Maipo ChimeraX version: 0.9 (2019-02-05) ================== Thank a bunch! Steven
-- Steve Chou
_______________________________________________ ChimeraX-users mailing list ChimeraX-users@cgl.ucsf.edu Manage subscription: http://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users
Hi Elaine, Thanks for your prompt response! You're right. ==================================== match #1/A@CA:10-20 to #2/B@CA:10-20 ==================================== match #1/A@CA:10-30 to #2/B@CA:10-30 ==================================== The above two gave the same result, because it used the same atoms for the alignment. But the following order ("@CA goes after :10-337) does not work for 'match'. =================================== match #1/A:10-20@CA to #2/B:10-20@CA =================================== The order ("@CA goes after :10-337) works for 'align'. =================================== align #1/A:10-20@CA to #2/B:10-20@CA =================================== All the best, Steve On Fri, Feb 8, 2019 at 4:07 PM Elaine Meng <meng@cgl.ucsf.edu> wrote:
Hi Steve, I couldn’t tell exactly without having those structures, but did you really mean chain B of model #2? Your first command (the one that works) has chain A of both models.
However, if you are just trying to match exactly those atoms, use the “align” command, not matchmaker. For example, something like
align #1/A:10-20@ca to #2/A:10-20@ca
<http://rbvi.ucsf.edu/chimerax/docs/user/commands/align.html>
Align uses the atoms you tell it to use and pairs them in order. Matchmaker figures out the residue pairing automatically by making a sequence alignment… this is especially useful for aligning structures that happen to have different residue numbering, homologs with different sequence lengths, etc. When if figures out the pairing, it automatically only uses CA atoms of each pair, so you wouldn’t need to give them in the command.
<http://rbvi.ucsf.edu/chimerax/docs/user/commands/matchmaker.html>
So generally, one would just specify the whole chain and not give specific residues with “matchmaker” except to suppress using parts of the structure you want to exclude from the sequence alignment. Nevertheless, the error message is unexpected so I will report it as a bug.
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Feb 8, 2019, at 12:46 PM, Steve Chou <stevezchou@gmail.com> wrote:
Dear ChimeraX support team, I tried the following command first - it worked well. ================== match #1/A to #2/A ================== Then I tried to align the two molecules using parts of them - it didn't work and complained that "No chains in reference structure PROT.pdb #2 compatible with BLOSUM-62 similarity matrix." ================== matchmaker #1/A:10-20@CA to #2/B:10-20@CA ================== System Info: ================== Platform: Linux-3.10.0-862.el7.x86_64-x86_64-with-redhat-7.5-Maipo ChimeraX version: 0.9 (2019-02-05) ================== Thanks a bunch! Steve
-- Steve Chou
_______________________________________________ ChimeraX-users mailing list ChimeraX-users@cgl.ucsf.edu Manage subscription: http://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users
-- Steve Chou
Yes, as I mentioned, the error when specific residues are given to “matchmaker” seems to be a bug, which I’ve made a ticket for in our bug database. The reason your commands “worked” with @ca before the specific residue numbers is that when you give the @ca first, you are in effect specifying the whole chain (alpha-carbon of every residue in the whole chain, plus all atoms of :10-20). In other words, this specifies 11 atoms #1/A:10-20@ca … whereas this specifies ALL of the CA atoms of chain A, plus all other atoms of residues 10-20 in that chain #1/A@ca:10-20 <http://rbvi.ucsf.edu/chimerax/docs/user/commands/atomspec.html#hierarchy> Again, giving specific residues to “matchmaker” is not the way it would normally be used. The pruning (fit iteration, which is on by default) will already remove parts that don’t superimpose well, like floppy loops. <http://rbvi.ucsf.edu/chimerax/docs/user/commands/matchmaker.html> Elaine
On Feb 8, 2019, at 1:19 PM, Steve Chou <stevezchou@gmail.com> wrote:
Hi Elaine, Thanks for your prompt response! You're right. ==================================== match #1/A@CA:10-20 to #2/B@CA:10-20 ==================================== match #1/A@CA:10-30 to #2/B@CA:10-30 ==================================== The above two gave the same result, because it used the same atoms for the alignment. But the following order ("@CA goes after :10-337) does not work for 'match'. =================================== match #1/A:10-20@CA to #2/B:10-20@CA =================================== The order ("@CA goes after :10-337) works for 'align'. =================================== align #1/A:10-20@CA to #2/B:10-20@CA =================================== All the best, Steve
On Fri, Feb 8, 2019 at 4:07 PM Elaine Meng <meng@cgl.ucsf.edu> wrote: Hi Steve, I couldn’t tell exactly without having those structures, but did you really mean chain B of model #2? Your first command (the one that works) has chain A of both models.
However, if you are just trying to match exactly those atoms, use the “align” command, not matchmaker. For example, something like
align #1/A:10-20@ca to #2/A:10-20@ca
<http://rbvi.ucsf.edu/chimerax/docs/user/commands/align.html>
Align uses the atoms you tell it to use and pairs them in order. Matchmaker figures out the residue pairing automatically by making a sequence alignment… this is especially useful for aligning structures that happen to have different residue numbering, homologs with different sequence lengths, etc. When if figures out the pairing, it automatically only uses CA atoms of each pair, so you wouldn’t need to give them in the command.
<http://rbvi.ucsf.edu/chimerax/docs/user/commands/matchmaker.html>
So generally, one would just specify the whole chain and not give specific residues with “matchmaker” except to suppress using parts of the structure you want to exclude from the sequence alignment. Nevertheless, the error message is unexpected so I will report it as a bug.
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Feb 8, 2019, at 12:46 PM, Steve Chou <stevezchou@gmail.com> wrote:
Dear ChimeraX support team, I tried the following command first - it worked well. ================== match #1/A to #2/A ================== Then I tried to align the two molecules using parts of them - it didn't work and complained that "No chains in reference structure PROT.pdb #2 compatible with BLOSUM-62 similarity matrix." ================== matchmaker #1/A:10-20@CA to #2/B:10-20@CA ================== System Info: ================== Platform: Linux-3.10.0-862.el7.x86_64-x86_64-with-redhat-7.5-Maipo ChimeraX version: 0.9 (2019-02-05) ================== Thanks a bunch! Steve
-- Steve Chou
_______________________________________________ ChimeraX-users mailing list ChimeraX-users@cgl.ucsf.edu Manage subscription: http://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users
-- Steve Chou
_______________________________________________ ChimeraX-users mailing list ChimeraX-users@cgl.ucsf.edu Manage subscription: http://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users
participants (2)
-
Elaine Meng -
Steve Chou