Selection of residues and superposition
Hi, I am trying to use Chimera to do superposition of multiple 3D structures. What I would really like to do is: 1. to select some residues for each of the structure and then 2. to superimpose the structures telling to chimera that the residues I selected have necessarily to be superimposed. To align "only" the protein sequences in a similar way for example I use COBALT that allows to specify the position of residues that have to be aligned necessarily together. Can I do this in chimera with 3D structures? thank so much for any help! -- Alessandra Gastaldello SIB | Swiss Institute of Bioinformatics CMU - 1, rue Michel Servet - 1211 Geneva 4 t: +41 22 379 50 50 - f: ... Alessandra.Gastaldello@sib.swiss - http://www.sib.swiss
Hi Allessandra, (A) Instead of selecting, you can use the “match” command and directly specify the residues to match. It will be a pairwise match, but if you choose one as the reference structure and match the others to it, you will end up with all of them superimposed. For example, match structure 1 to structure 3, and then match structure 2 to structure 3. The main difficulty with match is that you have to be careful to give exactly the same total number of atoms for each of the two structures. It can also be a lot of typing to enter many residue numbers, but you can give ranges. Example: match #0:25-35.A,68.A,123.A@n,ca,c,o #1:28-38.A,80.A,166.A@n,ca,c,o …. means superimpose model 0 residues 25-35,68,123 of chain A backbone atoms N,CA,C,O onto model 1 residues 28-38,80,166 of chain A backbone atoms N,CA,C,O. Many variations are possible, the chain IDs in the two structures could be different, and you could use different atoms, maybe only CA. You can see which structures are which model number in the Model Panel (in menu under Favorites). “match” command help: <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/match.html> For more examples, see the “Matching” section in the hydrolases image tutorial: <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/tutorials/frameimages.html> (B) Alternatively, if it is something like whole domains or other long residue ranges, it may be easier to still use MatchMaker with the option "Further restrict matching to current selection” (after selecting those parts). There is a similar capability in the “matchmaker” command but then instead of selecting you would again need to type in those ranges. <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/matchmaker/matchmaker.html> <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/mmaker.html> Summary of the different ways of superimposing structures in Chimera and when you might want to use one or the other: <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/superposition.html> 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
On Jun 23, 2016, at 9:15 AM, Alessandra Gastaldello <Alessandra.Gastaldello@sib.swiss> wrote:
Hi, I am trying to use Chimera to do superposition of multiple 3D structures. What I would really like to do is: 1. to select some residues for each of the structure and then 2. to superimpose the structures telling to chimera that the residues I selected have necessarily to be superimposed.
To align "only" the protein sequences in a similar way for example I use COBALT that allows to specify the position of residues that have to be aligned necessarily together.
Can I do this in chimera with 3D structures?
thank so much for any help!
Il 2016-06-23 19:37 Elaine Meng ha scritto:
Hi Allessandra, (A) Instead of selecting, you can use the “match” command and directly specify the residues to match. It will be a pairwise match, but if you choose one as the reference structure and match the others to it, you will end up with all of them superimposed. For example, match structure 1 to structure 3, and then match structure 2 to structure 3. The main difficulty with match is that you have to be careful to give exactly the same total number of atoms for each of the two structures. It can also be a lot of typing to enter many residue numbers, but you can give ranges. Example:
match #0:25-35.A,68.A,123.A@n,ca,c,o #1:28-38.A,80.A,166.A@n,ca,c,o
…. means superimpose model 0 residues 25-35,68,123 of chain A backbone atoms N,CA,C,O onto model 1 residues 28-38,80,166 of chain A backbone atoms N,CA,C,O. Many variations are possible, the chain IDs in the two structures could be different, and you could use different atoms, maybe only CA. You can see which structures are which model number in the Model Panel (in menu under Favorites).
“match” command help: <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/match.html>
For more examples, see the “Matching” section in the hydrolases image tutorial: <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/tutorials/frameimages.html>
(B) Alternatively, if it is something like whole domains or other long residue ranges, it may be easier to still use MatchMaker with the option "Further restrict matching to current selection” (after selecting those parts). There is a similar capability in the “matchmaker” command but then instead of selecting you would again need to type in those ranges. <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/matchmaker/matchmaker.html> <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/mmaker.html>
Summary of the different ways of superimposing structures in Chimera and when you might want to use one or the other: <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/superposition.html>
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
On Jun 23, 2016, at 9:15 AM, Alessandra Gastaldello <Alessandra.Gastaldello@sib.swiss> wrote:
Hi, I am trying to use Chimera to do superposition of multiple 3D structures. What I would really like to do is: 1. to select some residues for each of the structure and then 2. to superimpose the structures telling to chimera that the residues I selected have necessarily to be superimposed.
To align "only" the protein sequences in a similar way for example I use COBALT that allows to specify the position of residues that have to be aligned necessarily together.
Can I do this in chimera with 3D structures?
thank so much for any help!
Thank you so much Elaine! Very helpfull explanation about how to use match! Despite that, I have a problem with some match and selection. For some protein when I type for example: match #0:346@n,ca,o,c #1:23@n,ca,o,c to match those atoms in position 346 and 23, chimera tells me that the number of atoms is not the same. How this is possible? Thank again. -- Alessandra Gastaldello SIB | Swiss Institute of Bioinformatics CMU - 1, rue Michel Servet - 1211 Geneva 4 t: +41 22 379 50 50 - f: ... Alessandra.Gastaldello@sib.swiss - http://www.sib.swiss
Thank you so much Elaine! Very helpfull explanation about how to use match! Despite that, I have a problem with some match and selection. For some protein when I type for example: match #0:346@n,ca,o,c #1:23@n,ca,o,c to match those atoms in position 346 and 23, chimera tells me that the number of atoms is not the same. How this is possible? Thank again. -- Alessandra Gastaldello
Hi Allessandra, You’re welcome! You would have to check specifically, but some reasons for different or unexpected numbers of atoms are: (1) a model has multiple chains, so that if you just say residue 23, there might be residue 23 in chain A, residue 23 in chain B, … to avoid this, just specify the chain directly. You can see which chains you have by putting the mouse over the structure and it will be shown in the pop-up information. For residue 23 in chain A only: #1.23.A@n,ca,c,o (2) one or more of the atoms is missing from a structure (3) structure has alternate locations of atoms, e.g. two positions of the same atom. If there were locations A and B, you could first delete all the “B” locations in all structures with: delete @.b (4) residue number is incorrect so it did not specify any atoms Some ways to check are by trying to display all those atoms, for example: ~ribbon disp #0:346@n,ca,c,o #1:23@n,ca,c,o (hiding ribbon allows showing backbone atoms) You can also try selecting either set of atoms, for example: select #0:346@n,ca,c,o … and then see how many atoms are selected, such as by clicking the green magnifying glass in the bottom right corner to open the Selection Inspector. 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
Il 2016-06-24 20:21 Elaine Meng ha scritto:
Thank you so much Elaine! Very helpfull explanation about how to use match! Despite that, I have a problem with some match and selection. For some protein when I type for example: match #0:346@n,ca,o,c #1:23@n,ca,o,c to match those atoms in position 346 and 23, chimera tells me that the number of atoms is not the same. How this is possible? Thank again. -- Alessandra Gastaldello
Hi Allessandra, You’re welcome! You would have to check specifically, but some reasons for different or unexpected numbers of atoms are:
(1) a model has multiple chains, so that if you just say residue 23, there might be residue 23 in chain A, residue 23 in chain B, … to avoid this, just specify the chain directly. You can see which chains you have by putting the mouse over the structure and it will be shown in the pop-up information. For residue 23 in chain A only: #1.23.A@n,ca,c,o
(2) one or more of the atoms is missing from a structure
(3) structure has alternate locations of atoms, e.g. two positions of the same atom. If there were locations A and B, you could first delete all the “B” locations in all structures with: delete @.b
(4) residue number is incorrect so it did not specify any atoms
Some ways to check are by trying to display all those atoms, for example:
~ribbon disp #0:346@n,ca,c,o #1:23@n,ca,c,o
(hiding ribbon allows showing backbone atoms) You can also try selecting either set of atoms, for example:
select #0:346@n,ca,c,o … and then see how many atoms are selected, such as by clicking the green magnifying glass in the bottom right corner to open the Selection Inspector.
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
Thank you again Elaine! Very very useful. Unfortunately, I have another question. I would like to define a plane using the coordinates of some atoms in a structure and then to calculate the coordinates of the atoms that "surround" that plane. I know there is the "define plane" function to do that. But how can I get the coordinates of all atoms around that plane? Moreover I would like to compare that coordinates with the coordinates of other atoms in other structures calculated in the same way (defining a plane). In summary, I would like to see the atom distribution around some atom positions in different structures and to compare it. This means I would have to have a same "reference" plane for all atoms even if they are in different structures. Can I consider the different planes defined in the different structures as identical planes? I mean, is each of them as an "origin" plane, spatially identical to other calculated in the same way? Can I compare the coordinates of all the atoms in the different structures ? Or should I define a big unique plane specifying coordinates from all the structures and after calculate the distance of the atoms from that unique plane? Thank you very much in advance! -- Alessandra Gastaldello SIB | Swiss Institute of Bioinformatics CMU - 1, rue Michel Servet - 1211 Geneva 4 t: +41 22 379 50 50 - f: ... Alessandra.Gastaldello@sib.swiss - http://www.sib.swiss
Hi Allessandra, Sorry, there is no way to automatically find atoms that are close to a plane. It would require scripting (“define” command to create planes, “distance” to measure distance between each atom and a specific plane, python to loop over all atoms and to select only the ones with short distances). Distance measurements are also reported in the Reply Log and could be saved from there. You can save the set of selected atoms to a file (menu: File… Save PDB or command “write” with the option to save only selected atoms). <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/define.html> <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/distance.html> <http://www.rbvi.ucsf.edu/chimera/docs/ProgrammersGuide/basicPrimer.html> The planes are not all the same. Each will have a different size, orientation, and location because it is simply derived from the coordinates of the atoms used to define it. There is one way that might work to superimpose the structures using their individual planes, but I don’t know how well it would work (so maybe you should just skip to the next paragraph): use “align” commands to position each structure with its plane in the middle of the window facing the viewer, and then because some of them will be facing the wrong way (no way to distinguish the two sides of the plane), flip them using “turn y 180 center view” plus the “models” option to specify only the ones that need flipping. It would probably take a few separate attempts because you should not move anything with the mouse until after all of the align and turn commands. So the first time I’d just use all the align commands, one for each plane/structure, and then use the mouse to see which ones need flipping, and then I’d repeat all the align commands followed by turn as needed. <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/align.html> <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/turn.html> However, instead of trying to superimpose the individual planes, it may work better to first superimpose all your structures and then define a single master plane. 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 On Jun 29, 2016, at 7:43 AM, Alessandra Gastaldello <Alessandra.Gastaldello@sib.swiss> wrote:
Thank you again Elaine! Very very useful. Unfortunately, I have another question. I would like to define a plane using the coordinates of some atoms in a structure and then to calculate the coordinates of the atoms that "surround" that plane. I know there is the "define plane" function to do that. But how can I get the coordinates of all atoms around that plane?
Moreover I would like to compare that coordinates with the coordinates of other atoms in other structures calculated in the same way (defining a plane). In summary, I would like to see the atom distribution around some atom positions in different structures and to compare it. This means I would have to have a same "reference" plane for all atoms even if they are in different structures.
Can I consider the different planes defined in the different structures as identical planes? I mean, is each of them as an "origin" plane, spatially identical to other calculated in the same way? Can I compare the coordinates of all the atoms in the different structures ? Or should I define a big unique plane specifying coordinates from all the structures and after calculate the distance of the atoms from that unique plane?
Thank you very much in advance!
-- Alessandra Gastaldello
participants (2)
-
Alessandra Gastaldello -
Elaine Meng