Question about Model Loops Option in Chimera
Hello, I have a protein crystal structure where some regions are not resolved in the crystal. Therefore, I want to predict and fill these regions, without touching the rest, by using the Model Loops/Refine Structure option of Modeller in Chimera. When I select the non-terminal missing structure as a model and standard as a loop modeling protocol, some atoms in the rest of the structure are shifted from the original positions. However, I want to keep them as they are. Here I would like to ask what standard loop modeling protocol corresponds to. Does it also refine or optimize the structure as a whole? What does it actually correspond to? Any help would be appreciated. Thanks. Ece
Hi Ece, There is an option "Allow this many residues adjacent to missing regions to move" which you can change to zero instead of the default of one. Then it should not change any atom positions. See the documentation: <https://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/multalignviewer/modeller.html#building> I don't know the details of the "standard" option, maybe others can elaborate. As mentioned in the documentation, if you choose the DOPE option instead, it may produce higher-quality results, but it takes longer and is also more prone to calculation failure. I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Dec 12, 2022, at 5:38 AM, Ecem Güngör via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Hello,
I have a protein crystal structure where some regions are not resolved in the crystal. Therefore, I want to predict and fill these regions, without touching the rest, by using the Model Loops/Refine Structure option of Modeller in Chimera. When I select the non-terminal missing structure as a model and standard as a loop modeling protocol, some atoms in the rest of the structure are shifted from the original positions. However, I want to keep them as they are. Here I would like to ask what standard loop modeling protocol corresponds to. Does it also refine or optimize the structure as a whole? What does it actually correspond to?
Any help would be appreciated. Thanks.
Ece
When using the "standard" protocol, Modeller's LoopModel class is used, where as for DOPE it uses DOPELoopModel and for DOPE-HR it uses DOPEHRLoopModel. See DOPELoopModel reference <https://salilab.org/modeller/manual/node97.html> --Eric Eric Pettersen UCSF Computer Graphics Lab
On Dec 12, 2022, at 8:39 AM, Elaine Meng via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Hi Ece, There is an option "Allow this many residues adjacent to missing regions to move" which you can change to zero instead of the default of one. Then it should not change any atom positions.
See the documentation: <https://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/multalignviewer/modeller.html#building>
I don't know the details of the "standard" option, maybe others can elaborate. As mentioned in the documentation, if you choose the DOPE option instead, it may produce higher-quality results, but it takes longer and is also more prone to calculation failure.
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Dec 12, 2022, at 5:38 AM, Ecem Güngör via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Hello,
I have a protein crystal structure where some regions are not resolved in the crystal. Therefore, I want to predict and fill these regions, without touching the rest, by using the Model Loops/Refine Structure option of Modeller in Chimera. When I select the non-terminal missing structure as a model and standard as a loop modeling protocol, some atoms in the rest of the structure are shifted from the original positions. However, I want to keep them as they are. Here I would like to ask what standard loop modeling protocol corresponds to. Does it also refine or optimize the structure as a whole? What does it actually correspond to?
Any help would be appreciated. Thanks.
Ece
_______________________________________________ Chimera-users mailing list: Chimera-users@cgl.ucsf.edu Manage subscription: https://www.rbvi.ucsf.edu/mailman/listinfo/chimera-users
Dear Elanie, Thanks for your answer. Even if I set the option you mentioned to 0, it still moves some positions of the atoms. That's why I could not comprehend the reason. Ece Elaine Meng <meng@cgl.ucsf.edu>, 12 Ara 2022 Pzt, 17:39 tarihinde şunu yazdı:
Hi Ece, There is an option "Allow this many residues adjacent to missing regions to move" which you can change to zero instead of the default of one. Then it should not change any atom positions.
See the documentation: < https://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/multalignviewer/m...
I don't know the details of the "standard" option, maybe others can elaborate. As mentioned in the documentation, if you choose the DOPE option instead, it may produce higher-quality results, but it takes longer and is also more prone to calculation failure.
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Dec 12, 2022, at 5:38 AM, Ecem Güngör via Chimera-users < chimera-users@cgl.ucsf.edu> wrote:
Hello,
I have a protein crystal structure where some regions are not resolved in the crystal. Therefore, I want to predict and fill these regions, without touching the rest, by using the Model Loops/Refine Structure option of Modeller in Chimera. When I select the non-terminal missing structure as a model and standard as a loop modeling protocol, some atoms in the rest of the structure are shifted from the original positions. However, I want to keep them as they are. Here I would like to ask what standard loop modeling protocol corresponds to. Does it also refine or optimize the structure as a whole? What does it actually correspond to?
Any help would be appreciated. Thanks.
Ece
Hi Ece, As far as I know they should not move. If you are showing ribbons (not just atomic coordinates) it might seem like there are changes because the ribbon is interpolated over multiple positions, so the ribbon might be in a slightly different place even at residues where the atoms have not moved, just because the ribbon path calculation now has more atoms to use. Otherwise I would not expect anything to move. Hide ribbons, show all atoms... if you still see atoms in different positions then you could use Help... Report a Bug and attach data like the session, if you are willing to share the data. Elaine
On Dec 12, 2022, at 11:08 AM, Ecem Güngör via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Dear Elanie,
Thanks for your answer.
Even if I set the option you mentioned to 0, it still moves some positions of the atoms. That's why I could not comprehend the reason.
Ece
Elaine Meng <meng@cgl.ucsf.edu>, 12 Ara 2022 Pzt, 17:39 tarihinde şunu yazdı: Hi Ece, There is an option "Allow this many residues adjacent to missing regions to move" which you can change to zero instead of the default of one. Then it should not change any atom positions.
See the documentation: <https://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/multalignviewer/modeller.html#building>
I don't know the details of the "standard" option, maybe others can elaborate. As mentioned in the documentation, if you choose the DOPE option instead, it may produce higher-quality results, but it takes longer and is also more prone to calculation failure.
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Dec 12, 2022, at 5:38 AM, Ecem Güngör via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Hello,
I have a protein crystal structure where some regions are not resolved in the crystal. Therefore, I want to predict and fill these regions, without touching the rest, by using the Model Loops/Refine Structure option of Modeller in Chimera. When I select the non-terminal missing structure as a model and standard as a loop modeling protocol, some atoms in the rest of the structure are shifted from the original positions. However, I want to keep them as they are. Here I would like to ask what standard loop modeling protocol corresponds to. Does it also refine or optimize the structure as a whole? What does it actually correspond to?
Any help would be appreciated. Thanks.
Ece
_______________________________________________ Chimera-users mailing list: Chimera-users@cgl.ucsf.edu Manage subscription: https://www.rbvi.ucsf.edu/mailman/listinfo/chimera-users
This seems like a bug in both Chimera and ChimeraX. I will open a ticket in our bug-tracking database with you cc'ed so that you will be notified when it gets fixed. --Eric Eric Pettersen UCSF Computer Graphics Lab
On Dec 12, 2022, at 11:15 AM, Elaine Meng via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Hi Ece, As far as I know they should not move.
If you are showing ribbons (not just atomic coordinates) it might seem like there are changes because the ribbon is interpolated over multiple positions, so the ribbon might be in a slightly different place even at residues where the atoms have not moved, just because the ribbon path calculation now has more atoms to use.
Otherwise I would not expect anything to move. Hide ribbons, show all atoms... if you still see atoms in different positions then you could use Help... Report a Bug and attach data like the session, if you are willing to share the data. Elaine
On Dec 12, 2022, at 11:08 AM, Ecem Güngör via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Dear Elanie,
Thanks for your answer.
Even if I set the option you mentioned to 0, it still moves some positions of the atoms. That's why I could not comprehend the reason.
Ece
Elaine Meng <meng@cgl.ucsf.edu>, 12 Ara 2022 Pzt, 17:39 tarihinde şunu yazdı: Hi Ece, There is an option "Allow this many residues adjacent to missing regions to move" which you can change to zero instead of the default of one. Then it should not change any atom positions.
See the documentation: <https://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/multalignviewer/modeller.html#building>
I don't know the details of the "standard" option, maybe others can elaborate. As mentioned in the documentation, if you choose the DOPE option instead, it may produce higher-quality results, but it takes longer and is also more prone to calculation failure.
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Dec 12, 2022, at 5:38 AM, Ecem Güngör via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Hello,
I have a protein crystal structure where some regions are not resolved in the crystal. Therefore, I want to predict and fill these regions, without touching the rest, by using the Model Loops/Refine Structure option of Modeller in Chimera. When I select the non-terminal missing structure as a model and standard as a loop modeling protocol, some atoms in the rest of the structure are shifted from the original positions. However, I want to keep them as they are. Here I would like to ask what standard loop modeling protocol corresponds to. Does it also refine or optimize the structure as a whole? What does it actually correspond to?
Any help would be appreciated. Thanks.
Ece
_______________________________________________ Chimera-users mailing list: Chimera-users@cgl.ucsf.edu Manage subscription: https://www.rbvi.ucsf.edu/mailman/listinfo/chimera-users
_______________________________________________ Chimera-users mailing list: Chimera-users@cgl.ucsf.edu Manage subscription: https://www.rbvi.ucsf.edu/mailman/listinfo/chimera-users
Thanks a lot. But even I compare the atomic positions of the model and the original structures, it still shows a difference. By the way, this difference is not big but not understanding the protocol through Chimera documentation made me a bit annoyed. But it is still okay. What I found on the Modeller website is that potentially all of the atoms in the final model could move since either AutoLoop or LoopModel will try to build a comparative model using the input PDB as a template. You can check this site <https://salilab.org/modeller/wiki/Missing_residues>for more. If this is the case, it might be good to explain what the standard protocol means somewhere in the Chimera documentation. Otherwise, it might be a bit confusing. I will keep my eyes on the updates. Thanks a lot! Eric Pettersen <pett@cgl.ucsf.edu>, 12 Ara 2022 Pzt, 20:27 tarihinde şunu yazdı:
This *seems* like a bug in both Chimera and ChimeraX. I will open a ticket in our bug-tracking database with you cc'ed so that you will be notified when it gets fixed.
--Eric
Eric Pettersen UCSF Computer Graphics Lab
On Dec 12, 2022, at 11:15 AM, Elaine Meng via Chimera-users < chimera-users@cgl.ucsf.edu> wrote:
Hi Ece, As far as I know they should not move.
If you are showing ribbons (not just atomic coordinates) it might seem like there are changes because the ribbon is interpolated over multiple positions, so the ribbon might be in a slightly different place even at residues where the atoms have not moved, just because the ribbon path calculation now has more atoms to use.
Otherwise I would not expect anything to move. Hide ribbons, show all atoms... if you still see atoms in different positions then you could use Help... Report a Bug and attach data like the session, if you are willing to share the data. Elaine
On Dec 12, 2022, at 11:08 AM, Ecem Güngör via Chimera-users < chimera-users@cgl.ucsf.edu> wrote:
Dear Elanie,
Thanks for your answer.
Even if I set the option you mentioned to 0, it still moves some positions of the atoms. That's why I could not comprehend the reason.
Ece
Elaine Meng <meng@cgl.ucsf.edu>, 12 Ara 2022 Pzt, 17:39 tarihinde şunu yazdı: Hi Ece, There is an option "Allow this many residues adjacent to missing regions to move" which you can change to zero instead of the default of one. Then it should not change any atom positions.
See the documentation: < https://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/multalignviewer/m...
I don't know the details of the "standard" option, maybe others can elaborate. As mentioned in the documentation, if you choose the DOPE option instead, it may produce higher-quality results, but it takes longer and is also more prone to calculation failure.
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Dec 12, 2022, at 5:38 AM, Ecem Güngör via Chimera-users < chimera-users@cgl.ucsf.edu> wrote:
Hello,
I have a protein crystal structure where some regions are not resolved in the crystal. Therefore, I want to predict and fill these regions, without touching the rest, by using the Model Loops/Refine Structure option of Modeller in Chimera. When I select the non-terminal missing structure as a model and standard as a loop modeling protocol, some atoms in the rest of the structure are shifted from the original positions. However, I want to keep them as they are. Here I would like to ask what standard loop modeling protocol corresponds to. Does it also refine or optimize the structure as a whole? What does it actually correspond to?
Any help would be appreciated. Thanks.
Ece
_______________________________________________ Chimera-users mailing list: Chimera-users@cgl.ucsf.edu Manage subscription: https://www.rbvi.ucsf.edu/mailman/listinfo/chimera-users
_______________________________________________ Chimera-users mailing list: Chimera-users@cgl.ucsf.edu Manage subscription: https://www.rbvi.ucsf.edu/mailman/listinfo/chimera-users
On 12/12/22 9:03 PM, Ecem Güngör via Chimera-users wrote:
What I found on the Modeller website is that potentially all of the atoms in the final model could move since either AutoLoop or LoopModel will try to build a comparative model using the input PDB as a template. You can check this site <https://salilab.org/modeller/wiki/Missing_residues>for more.
You are misinterpreting that text. Comparative modeling (which is what is described at that URL) will optimize the entire model by default (although you can ask it not to). Loop modeling of a given input structure will only move the loop atoms: https://salilab.org/modeller/10.4/manual/node36.html Of course, if you do comparative modeling followed by loop refinement, all the atoms will move because of the comparative modeling step. (I had thought my answer when you asked about this a week ago on the Modeller mailing list addressed your question, but if anything wasn't clear, please let me know.) Ben -- ben@salilab.org https://salilab.org/~ben/ "It is a capital mistake to theorize before one has data." - Sir Arthur Conan Doyle
Hi Ece (and others), The bug in Chimera and ChimeraX, now fixed in both programs, was that when you choose "0" residues allowed to move, it was still allowing 1 residue on each side of the loop to move. To get the fix, download the current daily build: <https://www.rbvi.ucsf.edu/chimera/download.html#daily> In the fixed version, if you model loops and specify "0" residues adjacent to the loop allowed to move, then it will follow those instructions: all the atoms except for the loops/segments you are modeling will stay exactly in the same place. The protocol options (standard, DOPE, ...) are just which kind of scoring Modeller uses to choose the loop conformation, with DOPE being fancier/slower than the standard. I think that is already mentioned in the Chimera help. It doesn't say much because I couldn't find much at the Modeller website, but I will try to make it better. I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Dec 12, 2022, at 9:29 PM, Ben Webb via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
On 12/12/22 9:03 PM, Ecem Güngör via Chimera-users wrote:
What I found on the Modeller website is that potentially all of the atoms in the final model could move since either AutoLoop or LoopModel will try to build a comparative model using the input PDB as a template. You can check this site <https://salilab.org/modeller/wiki/Missing_residues>for more.
You are misinterpreting that text. Comparative modeling (which is what is described at that URL) will optimize the entire model by default (although you can ask it not to). Loop modeling of a given input structure will only move the loop atoms: https://salilab.org/modeller/10.4/manual/node36.html Of course, if you do comparative modeling followed by loop refinement, all the atoms will move because of the comparative modeling step.
(I had thought my answer when you asked about this a week ago on the Modeller mailing list addressed your question, but if anything wasn't clear, please let me know.)
Ben -- ben@salilab.org https://salilab.org/~ben/ "It is a capital mistake to theorize before one has data." - Sir Arthur Conan Doyle _______________________________________________ Chimera-users mailing list: Chimera-users@cgl.ucsf.edu Manage subscription: https://www.rbvi.ucsf.edu/mailman/listinfo/chimera-users
Thanks Elanie and others! Elaine Meng <meng@cgl.ucsf.edu>, 13 Ara 2022 Sal, 17:33 tarihinde şunu yazdı:
Hi Ece (and others), The bug in Chimera and ChimeraX, now fixed in both programs, was that when you choose "0" residues allowed to move, it was still allowing 1 residue on each side of the loop to move. To get the fix, download the current daily build: <https://www.rbvi.ucsf.edu/chimera/download.html#daily>
In the fixed version, if you model loops and specify "0" residues adjacent to the loop allowed to move, then it will follow those instructions: all the atoms except for the loops/segments you are modeling will stay exactly in the same place.
The protocol options (standard, DOPE, ...) are just which kind of scoring Modeller uses to choose the loop conformation, with DOPE being fancier/slower than the standard. I think that is already mentioned in the Chimera help. It doesn't say much because I couldn't find much at the Modeller website, but I will try to make it better.
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Dec 12, 2022, at 9:29 PM, Ben Webb via Chimera-users < chimera-users@cgl.ucsf.edu> wrote:
What I found on the Modeller website is that potentially all of the atoms in the final model could move since either AutoLoop or LoopModel will
On 12/12/22 9:03 PM, Ecem Güngör via Chimera-users wrote: try to build a comparative model using the input PDB as a template. You can check this site <https://salilab.org/modeller/wiki/Missing_residues>for more.
You are misinterpreting that text. Comparative modeling (which is what is described at that URL) will optimize the entire model by default (although you can ask it not to). Loop modeling of a given input structure will only move the loop atoms: https://salilab.org/modeller/10.4/manual/node36.html Of course, if you do comparative modeling followed by loop refinement, all the atoms will move because of the comparative modeling step.
(I had thought my answer when you asked about this a week ago on the Modeller mailing list addressed your question, but if anything wasn't clear, please let me know.)
Ben -- ben@salilab.org https://salilab.org/~ben/ "It is a capital mistake to theorize before one has data." - Sir Arthur Conan Doyle _______________________________________________ Chimera-users mailing list: Chimera-users@cgl.ucsf.edu Manage subscription: https://www.rbvi.ucsf.edu/mailman/listinfo/chimera-users
participants (4)
-
Ben Webb -
Ecem Güngör -
Elaine Meng -
Eric Pettersen