structural alignment
Friends, I have 10 conformations of the same RNA molecule obtained from 5 separate MD simulations. Now to compare the conformations, I want to make a structural alignment of these conformations and make an alignment map ( something like multiple seq. alignment for protein ) so as to show residues (or residue positions) that higher flexibility among these ten conformations. What is the best way to do it in Chimera. Thanks, Bala
Hi Bala, Since they are the same sequence, you don't need a sequence alignment, but we had to make a small fix in Chimera to allow it, so you would need the *next* daily build to follow this recipe. (or skip down to the lower part of this message for a different approach you can do now!) You would just show the sequence of one copy, then associate all the structures with that sequence, then show the RMSD header. The RMSD header bar heights represent the RMSD among residues at each position, and it is also available as the residue attribute mavRMSD for display with colors and "worms" using Render by Attribute. <http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/multalignviewer/mul...> (1) showing sequence. Choose Favorites... Sequence, pick any one of the RNA chains. Remember which one you chose. (2) associating. In the sequence window menu, choose Structures... Associations and set all of the other structure models to also associate with that same sequence, OK. This could also be done if the sequences were not identical but highly similar. (3) superimposing. In the sequence window menu, choose Structure... Match and superimpose everything. Don't turn on any of the options, just click OK. (4) showing RMSD header. In the sequence window menu, FIRST make the sequence area larger vertically by resizing the window (necessary to avoid a technical problem) and then choose Headers... RMSD. (5) showing RMSD values on a structure with color and/or "worms." You would probably hide all the structures except one (e.g. command: ~modeldisp #.2-end), then choose Tools... Depiction... Render by Attribute, and in that dialog, use the attribute of "residues" named "mavRMSD" of the same model as you are displaying, etc. You can adjust the mapping of values to color or worm radius however you like, as explained here: <http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/render/render.html#...> Actually, Eric just told me that you can do a similar thing now (without the fix I mentioned) by making a sequence alignment instead of taking the more elegant single-sequence approach. Here is that alternative recipe: After #1 above, (2) add one more copy of the sequence to make an alignment of two sequences. FIRST make the sequence area larger vertically by resizing the window (necessary to avoid a technical problem). From the sequence window menu, choose Edit... Add Sequence, choose From Structure, and choose any copy of the chain that is *different* than the one you chose in #1. In your case (RNA) you also need to change the Matrix to "Nucleic" before clicking OK. Now you should have an alignment of two sequences with all the other structures automatically already associated with the second sequence. Then carry on with #3-5 as above. Below I attach an image showing RMSD with colors blue->red->yellow and worms for an NMR ensemble of RNA, NDB entry 17ra. Only the first member of the ensemble is shown. I hope this helps, Elaine ---------- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
Dear Elaine, Thank you. But i have a problem in doing this because my RNA has two principal chains [two strands seperated by TER card, attached pdb file]. It would be nice if there is a way to output the rmsd so that i can make a box plot to show which residues are flexible among those conformations. Thank you, Bala On Wed, Jul 7, 2010 at 12:26 AM, Elaine Meng <meng@cgl.ucsf.edu> wrote:
Hi Bala, Since they are the same sequence, you don't need a sequence alignment, but we had to make a small fix in Chimera to allow it, so you would need the *next* daily build to follow this recipe. (or skip down to the lower part of this message for a different approach you can do now!) You would just show the sequence of one copy, then associate all the structures with that sequence, then show the RMSD header. The RMSD header bar heights represent the RMSD among residues at each position, and it is also available as the residue attribute mavRMSD for display with colors and "worms" using Render by Attribute. < http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/multalignviewer/mul...
(1) showing sequence. Choose Favorites... Sequence, pick any one of the RNA chains. Remember which one you chose. (2) associating. In the sequence window menu, choose Structures... Associations and set all of the other structure models to also associate with that same sequence, OK. This could also be done if the sequences were not identical but highly similar. (3) superimposing. In the sequence window menu, choose Structure... Match and superimpose everything. Don't turn on any of the options, just click OK. (4) showing RMSD header. In the sequence window menu, FIRST make the sequence area larger vertically by resizing the window (necessary to avoid a technical problem) and then choose Headers... RMSD. (5) showing RMSD values on a structure with color and/or "worms." You would probably hide all the structures except one (e.g. command: ~modeldisp #.2-end), then choose Tools... Depiction... Render by Attribute, and in that dialog, use the attribute of "residues" named "mavRMSD" of the same model as you are displaying, etc. You can adjust the mapping of values to color or worm radius however you like, as explained here: < http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/render/render.html#...
Actually, Eric just told me that you can do a similar thing now (without the fix I mentioned) by making a sequence alignment instead of taking the more elegant single-sequence approach. Here is that alternative recipe: After #1 above, (2) add one more copy of the sequence to make an alignment of two sequences. FIRST make the sequence area larger vertically by resizing the window (necessary to avoid a technical problem). From the sequence window menu, choose Edit... Add Sequence, choose From Structure, and choose any copy of the chain that is *different* than the one you chose in #1. In your case (RNA) you also need to change the Matrix to "Nucleic" before clicking OK. Now you should have an alignment of two sequences with all the other structures automatically already associated with the second sequence. Then carry on with #3-5 as above.
Below I attach an image showing RMSD with colors blue->red->yellow and worms for an NMR ensemble of RNA, NDB entry 17ra. Only the first member of the ensemble is shown.
I hope this helps, Elaine ---------- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
Hi Bala, You can repeat the whole process as described for each principal chain, but the RMSDs would be for the chains superimposed separately rather than both at the same time. Or, you could just strip out the TERs and pretend each model with two chains is a single chain, to perform the process on both together. I am not sure what a box plot is, but besides showing RMSDs with color or worms on the structure, you could (a) save the sequence window contents, which includes the RMSD header histogram, as an EPS file (see File... Save EPS in the sequence window menu). (b) save the mavRMSD values to a text file using Render by Attribute (see File... Save Attributes in that dialog). Then you could convert that file as needed to make input for some other plotting program that is able to make the kind of plot you want. Best, Elaine ---------- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco On Jul 6, 2010, at 3:59 PM, Bala subramanian wrote:
Dear Elaine, Thank you. But i have a problem in doing this because my RNA has two principal chains [two strands seperated by TER card, attached pdb file]. It would be nice if there is a way to output the rmsd so that i can make a box plot to show which residues are flexible among those conformations. Thank you, Bala
Hi Bala, The recommended single-sequence procedure now works (daily builds dated Jul 7 2010 or later), and it's no longer necessary to resize the window manually before showing the RMSD header. Elaine On Jul 6, 2010, at 3:26 PM, Elaine Meng wrote:
You would just show the sequence of one copy, then associate all the structures with that sequence, then show the RMSD header. The RMSD header bar heights represent the RMSD among residues at each position, and it is also available as the residue attribute mavRMSD for display with colors and "worms" using Render by Attribute.
(1) showing sequence. Choose Favorites... Sequence, pick any one of the RNA chains. Remember which one you chose. (2) associating. In the sequence window menu, choose Structures... Associations and set all of the other structure models to also associate with that same sequence, OK. This could also be done if the sequences were not identical but highly similar. (3) superimposing. In the sequence window menu, choose Structure... Match and superimpose everything. Don't turn on any of the options, just click OK. (4) showing RMSD header. In the sequence window menu, FIRST make the sequence area larger vertically by resizing the window (necessary to avoid a technical problem) and then choose Headers... RMSD. (5) showing RMSD values on a structure with color and/or "worms." You would probably hide all the structures except one (e.g. command: ~modeldisp #.2-end), then choose Tools... Depiction... Render by Attribute, and in that dialog, use the attribute of "residues" named "mavRMSD" of the same model as you are displaying, etc. You can adjust the mapping of values to color or worm radius however you like, as explained here: <http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/render/render.html#...>
(alternatively, save sequence window as EPS image and/or save mavRMSD values to file)
participants (2)
-
Bala subramanian -
Elaine Meng