Issue in using vop maximum to make composite map
Dear ChimeraX developers, I have an issue when using vop maximum command to combine two local maps. Map A and Map B have an overlapping area with each other. The hinge point of structural flexibility is at the center of the overlapping area. Therefore, the left part of Map A and the right part of Map B have a better local density quality, which I want to keep in the composite map. But vop maximum did it on the opposite, resulting in a worse local resolution at the edges of the overlapping area (e.g., Region 1 and 2). Technically, this is not an issue of vop maximium itself. In Region 1, Map A has a better local density quality, while Map B has higher values in some voxels. So the composite map has more voxels derived from Map B in Region 1. Vice versa for Region 2. The only solution in my mind is to split both Map A and B at the hinge point, then do vop maximum for only the left part of Map A and the right part of Map B. But this is time-consuming since I have many local maps to combine. I wonder if you have a better solution? I would appreciate it if you could give advices. Thanks! Best regards, Heng  Heng Zhou, PhD | Postdoctoral research fellow Stefan Raunser lab | Structural Biochemistry Max Planck Institute of Molecular Physiology Otto-Hahn-Straße 11, 44227 Dortmund, Germany Phone: +49 231-133-2326
Hi Heng, I don't have a good idea of how to combine to focused refinement maps so that the overlap region is of best quality. But here are some ideas about what is causing the degradation in quality when you take the maximum at each grid point. If one map has low resolution or has high enough intensity noise where the other map has sharp well-defined density you would hope that the second map has higher intensity at each grid point so that taking a maximum value would ignore the poor quality map. But the low resolution noisy map has broader blobs than the sharp map or blobs of density that don't even overlap the sharp map structure so when you take the maximum all that noise gets added outside the contours of the sharp map density and degrades the quality. I think the only way to remedy that is you can't combine noisy poor quality density when it extends beyond the sharp density of the other map. But the low quality density is probably always going to extend beyond the sharp density. So my conclusion is you to get rid of that poor quality density before you combine the two with maximum, or you combine them using some other function for instance that only puts high density where both maps have high density. The most obvious way to achieve this is to carefully by hand erase the low quality density in both maps near the overlap reagions for example with the ChimeraX Map Eraser tool. But you said you want a more automated way since you have many maps to combine. Possibly if you have local resolution estimates for both maps you could use that to combine them more sensibily than using maximum, but that would require some custom code and I think it is harder than it looks. Another vague idea that would need special purpose code is to define a plane at the hinge and only blend the two maps within a small region (10A either side) near that plane. Combining focused refinement maps has been a common problem for several years now, so perhaps there is literature that describes better techniques, or exisiting software which does it in better ways than per-grid-point maximum. Tom
On May 27, 2025, at 4:50 AM, Dr. Heng Zhou via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:
Dear ChimeraX developers,
I have an issue when using vop maximum command to combine two local maps. Map A and Map B have an overlapping area with each other. The hinge point of structural flexibility is at the center of the overlapping area. Therefore, the left part of Map A and the right part of Map B have a better local density quality, which I want to keep in the composite map. But vop maximum did it on the opposite, resulting in a worse local resolution at the edges of the overlapping area (e.g., Region 1 and 2).
Technically, this is not an issue of vop maximium itself. In Region 1, Map A has a better local density quality, while Map B has higher values in some voxels. So the composite map has more voxels derived from Map B in Region 1. Vice versa for Region 2.
The only solution in my mind is to split both Map A and B at the hinge point, then do vop maximum for only the left part of Map A and the right part of Map B. But this is time-consuming since I have many local maps to combine. I wonder if you have a better solution? I would appreciate it if you could give advices. Thanks!
Best regards, Heng <PastedGraphic-1.png>
Heng Zhou, PhD | Postdoctoral research fellow Stefan Raunser lab | Structural Biochemistry Max Planck Institute of Molecular Physiology Otto-Hahn-Straße 11, 44227 Dortmund, Germany Phone: +49 231-133-2326
_______________________________________________ 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/
Hello Heng, I am not sure there is any consensus in the cryoEM community about composite maps. Yes, local refinement can produce better maps for individual domains, and yes these maps are fair game to use to build and refine better atomic models of the corresponding domains separately. But it is difficult to say what it the best way to model the entire particle. There are several problems with a composite map obtained by stitching locally refined maps: 1. The composite map does not really represents the underlying particles. If the domains are moving relative to each other around the hinge, then the composite map with a single relative orientation between the two domains only correctly represents very few particles in the dataset. And the resulting atomic model refined against this map will not capture the conformational variability. So, with such a model, you are in effect hiding the fact that the particle is dynamic. One could argue that, in such a case, a more faithful model would be to rigidly place the atomic models of the individual domains obtained from the locally refined maps into the "consensus" map, and refine only their atomic b-factors (or maybe their coordinates too, depending on the resolution of the consensus map, but then likely with tight restraints: either secondary structure restraints or reference model restraints set to the initial atomic models). 2. As you experience first hand, it is very difficult to produce a composite map without artifacts at the "seams". Unphysical density at the seams will cause problems when refining an atomic model against this map. 3. No half-maps exist for the composite map, therefore you cannot determine a global resolution for this map. In a way this is good, because it forces you to make it explicit that it is a composite map. Tempy-Reff has a feature to produce composite maps without "seam artifacts". I think you first need to refine an atomic model with this program to be able to use this feature. This only addresses problem 2. You would still have no half-maps for the composite maps, and the map and model would still not capture the conformational heterogeneity in the set of particles. https://doi.org/10.1038/s41467-023-44593-1 https://www.topf-group.org/tempy-reff I hope this helps, Guillaume On 27 May 2025, at 19:26, Tom Goddard via ChimeraX-users <chimerax-users@cgl.ucsf.edu<mailto:chimerax-users@cgl.ucsf.edu>> wrote: Hi Heng, I don't have a good idea of how to combine to focused refinement maps so that the overlap region is of best quality. But here are some ideas about what is causing the degradation in quality when you take the maximum at each grid point. If one map has low resolution or has high enough intensity noise where the other map has sharp well-defined density you would hope that the second map has higher intensity at each grid point so that taking a maximum value would ignore the poor quality map. But the low resolution noisy map has broader blobs than the sharp map or blobs of density that don't even overlap the sharp map structure so when you take the maximum all that noise gets added outside the contours of the sharp map density and degrades the quality. I think the only way to remedy that is you can't combine noisy poor quality density when it extends beyond the sharp density of the other map. But the low quality density is probably always going to extend beyond the sharp density. So my conclusion is you to get rid of that poor quality density before you combine the two with maximum, or you combine them using some other function for instance that only puts high density where both maps have high density. The most obvious way to achieve this is to carefully by hand erase the low quality density in both maps near the overlap reagions for example with the ChimeraX Map Eraser tool. But you said you want a more automated way since you have many maps to combine. Possibly if you have local resolution estimates for both maps you could use that to combine them more sensibily than using maximum, but that would require some custom code and I think it is harder than it looks. Another vague idea that would need special purpose code is to define a plane at the hinge and only blend the two maps within a small region (10A either side) near that plane. Combining focused refinement maps has been a common problem for several years now, so perhaps there is literature that describes better techniques, or exisiting software which does it in better ways than per-grid-point maximum. Tom On May 27, 2025, at 4:50 AM, Dr. Heng Zhou via ChimeraX-users <chimerax-users@cgl.ucsf.edu<mailto:chimerax-users@cgl.ucsf.edu>> wrote: Dear ChimeraX developers, I have an issue when using vop maximum command to combine two local maps. Map A and Map B have an overlapping area with each other. The hinge point of structural flexibility is at the center of the overlapping area. Therefore, the left part of Map A and the right part of Map B have a better local density quality, which I want to keep in the composite map. But vop maximum did it on the opposite, resulting in a worse local resolution at the edges of the overlapping area (e.g., Region 1 and 2). Technically, this is not an issue of vop maximium itself. In Region 1, Map A has a better local density quality, while Map B has higher values in some voxels. So the composite map has more voxels derived from Map B in Region 1. Vice versa for Region 2. The only solution in my mind is to split both Map A and B at the hinge point, then do vop maximum for only the left part of Map A and the right part of Map B. But this is time-consuming since I have many local maps to combine. I wonder if you have a better solution? I would appreciate it if you could give advices. Thanks! Best regards, Heng <PastedGraphic-1.png> Heng Zhou, PhD | Postdoctoral research fellow Stefan Raunser lab | Structural Biochemistry Max Planck Institute of Molecular Physiology Otto-Hahn-Straße 11, 44227 Dortmund, Germany Phone: +49 231-133-2326 _______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu<mailto:chimerax-users@cgl.ucsf.edu> To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu<mailto:chimerax-users-leave@cgl.ucsf.edu> Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/ VARNING: Klicka inte på länkar och öppna inte bilagor om du inte känner igen avsändaren och vet att innehållet är säkert. CAUTION: Do not click on links or open attachments unless you recognise the sender and know the content is safe. _______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu<mailto:chimerax-users@cgl.ucsf.edu> To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu<mailto:chimerax-users-leave@cgl.ucsf.edu> Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/ När du har kontakt med oss på Uppsala universitet med e-post så innebär det att vi behandlar dina personuppgifter. För att läsa mer om hur vi gör det kan du läsa här: http://www.uu.se/om-uu/dataskydd-personuppgifter/ E-mailing Uppsala University means that we will process your personal data. For more information on how this is performed, please read here: http://www.uu.se/en/about-uu/data-protection-policy
Hi all, You may be interested in this phenix program: https://phenix-online.org/documentation/reference/combine_focused_maps.html Cheers, Alexis On Tue, May 27, 2025 at 11:42 AM Guillaume Gaullier via ChimeraX-users < chimerax-users@cgl.ucsf.edu> wrote:
Hello Heng,
I am not sure there is any consensus in the cryoEM community about composite maps.
Yes, local refinement can produce better maps for individual domains, and yes these maps are fair game to use to build and refine better atomic models of the corresponding domains separately. But it is difficult to say what it the best way to model the entire particle. There are several problems with a composite map obtained by stitching locally refined maps:
1. The composite map does not really represents the underlying particles. If the domains are moving relative to each other around the hinge, then the composite map with a single relative orientation between the two domains only correctly represents very few particles in the dataset. And the resulting atomic model refined against this map will not capture the conformational variability. So, with such a model, you are in effect hiding the fact that the particle is dynamic. One could argue that, in such a case, a more faithful model would be to rigidly place the atomic models of the individual domains obtained from the locally refined maps into the "consensus" map, and refine only their atomic b-factors (or maybe their coordinates too, depending on the resolution of the consensus map, but then likely with tight restraints: either secondary structure restraints or reference model restraints set to the initial atomic models).
2. As you experience first hand, it is very difficult to produce a composite map without artifacts at the "seams". Unphysical density at the seams will cause problems when refining an atomic model against this map.
3. No half-maps exist for the composite map, therefore you cannot determine a global resolution for this map. In a way this is good, because it forces you to make it explicit that it is a composite map.
Tempy-Reff has a feature to produce composite maps without "seam artifacts". I think you first need to refine an atomic model with this program to be able to use this feature. This only addresses problem 2. You would still have no half-maps for the composite maps, and the map and model would still not capture the conformational heterogeneity in the set of particles. https://doi.org/10.1038/s41467-023-44593-1 https://www.topf-group.org/tempy-reff
I hope this helps,
Guillaume
On 27 May 2025, at 19:26, Tom Goddard via ChimeraX-users < chimerax-users@cgl.ucsf.edu> wrote:
Hi Heng,
I don't have a good idea of how to combine to focused refinement maps so that the overlap region is of best quality. But here are some ideas about what is causing the degradation in quality when you take the maximum at each grid point. If one map has low resolution or has high enough intensity noise where the other map has sharp well-defined density you would hope that the second map has higher intensity at each grid point so that taking a maximum value would ignore the poor quality map. But the low resolution noisy map has broader blobs than the sharp map or blobs of density that don't even overlap the sharp map structure so when you take the maximum all that noise gets added outside the contours of the sharp map density and degrades the quality. I think the only way to remedy that is you can't combine noisy poor quality density when it extends beyond the sharp density of the other map. But the low quality density is probably always going to extend beyond the sharp density. So my conclusion is you to get rid of that poor quality density before you combine the two with maximum, or you combine them using some other function for instance that only puts high density where both maps have high density. The most obvious way to achieve this is to carefully by hand erase the low quality density in both maps near the overlap reagions for example with the ChimeraX Map Eraser tool. But you said you want a more automated way since you have many maps to combine. Possibly if you have local resolution estimates for both maps you could use that to combine them more sensibily than using maximum, but that would require some custom code and I think it is harder than it looks. Another vague idea that would need special purpose code is to define a plane at the hinge and only blend the two maps within a small region (10A either side) near that plane.
Combining focused refinement maps has been a common problem for several years now, so perhaps there is literature that describes better techniques, or exisiting software which does it in better ways than per-grid-point maximum.
Tom
On May 27, 2025, at 4:50 AM, Dr. Heng Zhou via ChimeraX-users < chimerax-users@cgl.ucsf.edu> wrote:
Dear ChimeraX developers,
I have an issue when using vop maximum command to combine two local maps. Map A and Map B have an overlapping area with each other. The hinge point of structural flexibility is at the center of the overlapping area. Therefore, the left part of Map A and the right part of Map B have a better local density quality, which I want to keep in the composite map. But vop maximum did it on the opposite, resulting in a worse local resolution at the edges of the overlapping area (e.g., Region 1 and 2).
Technically, this is not an issue of vop maximium itself. In Region 1, Map A has a better local density quality, while Map B has higher values in some voxels. So the composite map has more voxels derived from Map B in Region 1. Vice versa for Region 2.
The only solution in my mind is to split both Map A and B at the hinge point, then do vop maximum for only the left part of Map A and the right part of Map B. But this is time-consuming since I have many local maps to combine. I wonder if you have a better solution? I would appreciate it if you could give advices. Thanks!
Best regards, Heng <PastedGraphic-1.png>
Heng Zhou, PhD | Postdoctoral research fellow Stefan Raunser lab | Structural Biochemistry Max Planck Institute of Molecular Physiology Otto-Hahn-Straße 11, 44227 Dortmund, Germany Phone: +49 231-133-2326
_______________________________________________ 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/
VARNING: Klicka inte på länkar och öppna inte bilagor om du inte känner igen avsändaren och vet att innehållet är säkert. CAUTION: Do not click on links or open attachments unless you recognise the sender and know the content is safe.
_______________________________________________ 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/
När du har kontakt med oss på Uppsala universitet med e-post så innebär det att vi behandlar dina personuppgifter. För att läsa mer om hur vi gör det kan du läsa här: http://www.uu.se/om-uu/dataskydd-personuppgifter/
E-mailing Uppsala University means that we will process your personal data. For more information on how this is performed, please read here: http://www.uu.se/en/about-uu/data-protection-policy _______________________________________________ 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/
Hi Tom, I fully agree with your idea about the reason for quality degradation. And I agree that manually erasing the low-resolution density before combining maps would work. Actually, my thoughts were quite similar to yours. First, build local models only using regions with good quality of focused maps. Then, use the color zone tool in ChimeraX to split the focused maps, and only combine those split regions with good quality to make the composite map. But this is a bit more complicated than volume erasing. I will try your approach instead. Thank you so much for your advice! The other two suggestions also sound feasible, but custom coding is beyond my capability. Fortunately, there’s already a similar solution, as mentioned by Guillaume (https://doi.org/10.1038/s41467-023-44593-1 the idea that only blend the two maps within a small region near a defined plane, it may only work for simple hinge. I also have other cases where the hinge is a curved surface, I imagine defining that surface could already be an issue. Best, Heng
On 27. May 2025, at 19:26, Tom Goddard <goddard@sonic.net> wrote:
Hi Heng,
I don't have a good idea of how to combine to focused refinement maps so that the overlap region is of best quality. But here are some ideas about what is causing the degradation in quality when you take the maximum at each grid point. If one map has low resolution or has high enough intensity noise where the other map has sharp well-defined density you would hope that the second map has higher intensity at each grid point so that taking a maximum value would ignore the poor quality map. But the low resolution noisy map has broader blobs than the sharp map or blobs of density that don't even overlap the sharp map structure so when you take the maximum all that noise gets added outside the contours of the sharp map density and degrades the quality. I think the only way to remedy that is you can't combine noisy poor quality density when it extends beyond the sharp density of the other map. But the low quality density is probably always going to extend beyond the sharp density. So my conclusion is you to get rid of that poor quality density before you combine the two with maximum, or you combine them using some other function for instance that only puts high density where both maps have high density. The most obvious way to achieve this is to carefully by hand erase the low quality density in both maps near the overlap reagions for example with the ChimeraX Map Eraser tool. But you said you want a more automated way since you have many maps to combine. Possibly if you have local resolution estimates for both maps you could use that to combine them more sensibily than using maximum, but that would require some custom code and I think it is harder than it looks. Another vague idea that would need special purpose code is to define a plane at the hinge and only blend the two maps within a small region (10A either side) near that plane.
Combining focused refinement maps has been a common problem for several years now, so perhaps there is literature that describes better techniques, or exisiting software which does it in better ways than per-grid-point maximum.
Tom
On May 27, 2025, at 4:50 AM, Dr. Heng Zhou via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:
Dear ChimeraX developers,
I have an issue when using vop maximum command to combine two local maps. Map A and Map B have an overlapping area with each other. The hinge point of structural flexibility is at the center of the overlapping area. Therefore, the left part of Map A and the right part of Map B have a better local density quality, which I want to keep in the composite map. But vop maximum did it on the opposite, resulting in a worse local resolution at the edges of the overlapping area (e.g., Region 1 and 2).
Technically, this is not an issue of vop maximium itself. In Region 1, Map A has a better local density quality, while Map B has higher values in some voxels. So the composite map has more voxels derived from Map B in Region 1. Vice versa for Region 2.
The only solution in my mind is to split both Map A and B at the hinge point, then do vop maximum for only the left part of Map A and the right part of Map B. But this is time-consuming since I have many local maps to combine. I wonder if you have a better solution? I would appreciate it if you could give advices. Thanks!
Best regards, Heng <http://PastedGraphic-1.png>
Heng Zhou, PhD | Postdoctoral research fellow Stefan Raunser lab | Structural Biochemistry Max Planck Institute of Molecular Physiology Otto-Hahn-Straße 11, 44227 Dortmund, Germany Phone: +49 231-133-2326
_______________________________________________ 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/
Hi Guillaume, Yes, I agree with you, the composite map only represents very few particles in the dataset, and can’t faithfully reflect structural flexibility. So I’m also going to show the "consensus map” and maps from 3D classification in the manuscript (and upload to PDB) to analyze conformational viability. The protein I’m working on is a pentamer, with 3 flexible regions in each protomer. All these flexible regions move independently, so more than half of the areas in the consensus map are blurred or missing. Without a composite map, one can hardly get an impression of what the whole complex looks like. I think the composite map is a plus, as long as it is properly used together with the “real” maps, and with a clear statement that it is a composite map. The software you mentioned for producing seamless composite maps sounds like a wonderful solution. I will try it out. Thanks a lot for your suggestions. And glad to discuss with you! Best, Heng
On 27. May 2025, at 20:38, Guillaume Gaullier <http://guillaume.gaullier@kemi.uu.se> wrote:
Hello Heng,
I am not sure there is any consensus in the cryoEM community about composite maps.
Yes, local refinement can produce better maps for individual domains, and yes these maps are fair game to use to build and refine better atomic models of the corresponding domains separately. But it is difficult to say what it the best way to model the entire particle. There are several problems with a composite map obtained by stitching locally refined maps:
1. The composite map does not really represents the underlying particles. If the domains are moving relative to each other around the hinge, then the composite map with a single relative orientation between the two domains only correctly represents very few particles in the dataset. And the resulting atomic model refined against this map will not capture the conformational variability. So, with such a model, you are in effect hiding the fact that the particle is dynamic. One could argue that, in such a case, a more faithful model would be to rigidly place the atomic models of the individual domains obtained from the locally refined maps into the "consensus" map, and refine only their atomic b-factors (or maybe their coordinates too, depending on the resolution of the consensus map, but then likely with tight restraints: either secondary structure restraints or reference model restraints set to the initial atomic models).
2. As you experience first hand, it is very difficult to produce a composite map without artifacts at the "seams". Unphysical density at the seams will cause problems when refining an atomic model against this map.
3. No half-maps exist for the composite map, therefore you cannot determine a global resolution for this map. In a way this is good, because it forces you to make it explicit that it is a composite map.
Tempy-Reff has a feature to produce composite maps without "seam artifacts". I think you first need to refine an atomic model with this program to be able to use this feature. This only addresses problem 2. You would still have no half-maps for the composite maps, and the map and model would still not capture the conformational heterogeneity in the set of particles. https://doi.org/10.1038/s41467-023-44593-1> https://www.topf-group.org/tempy-reff> I hope this helps,
Guillaume
On 27 May 2025, at 19:26, Tom Goddard via ChimeraX-users <chimerax-users@cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu>> wrote:
Hi Heng,
I don't have a good idea of how to combine to focused refinement maps so that the overlap region is of best quality. But here are some ideas about what is causing the degradation in quality when you take the maximum at each grid point. If one map has low resolution or has high enough intensity noise where the other map has sharp well-defined density you would hope that the second map has higher intensity at each grid point so that taking a maximum value would ignore the poor quality map. But the low resolution noisy map has broader blobs than the sharp map or blobs of density that don't even overlap the sharp map structure so when you take the maximum all that noise gets added outside the contours of the sharp map density and degrades the quality. I think the only way to remedy that is you can't combine noisy poor quality density when it extends beyond the sharp density of the other map. But the low quality density is probably always going to extend beyond the sharp density. So my conclusion is you to get rid of that poor quality density before you combine the two with maximum, or you combine them using some other function for instance that only puts high density where both maps have high density. The most obvious way to achieve this is to carefully by hand erase the low quality density in both maps near the overlap reagions for example with the ChimeraX Map Eraser tool. But you said you want a more automated way since you have many maps to combine. Possibly if you have local resolution estimates for both maps you could use that to combine them more sensibily than using maximum, but that would require some custom code and I think it is harder than it looks. Another vague idea that would need special purpose code is to define a plane at the hinge and only blend the two maps within a small region (10A either side) near that plane.
Combining focused refinement maps has been a common problem for several years now, so perhaps there is literature that describes better techniques, or exisiting software which does it in better ways than per-grid-point maximum.
Tom
On May 27, 2025, at 4:50 AM, Dr. Heng Zhou via ChimeraX-users <chimerax-users@cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu>> wrote:
Dear ChimeraX developers,
I have an issue when using vop maximum command to combine two local maps. Map A and Map B have an overlapping area with each other. The hinge point of structural flexibility is at the center of the overlapping area. Therefore, the left part of Map A and the right part of Map B have a better local density quality, which I want to keep in the composite map. But vop maximum did it on the opposite, resulting in a worse local resolution at the edges of the overlapping area (e.g., Region 1 and 2).
Technically, this is not an issue of vop maximium itself. In Region 1, Map A has a better local density quality, while Map B has higher values in some voxels. So the composite map has more voxels derived from Map B in Region 1. Vice versa for Region 2.
The only solution in my mind is to split both Map A and B at the hinge point, then do vop maximum for only the left part of Map A and the right part of Map B. But this is time-consuming since I have many local maps to combine. I wonder if you have a better solution? I would appreciate it if you could give advices. Thanks!
Best regards, Heng <http://PastedGraphic-1.png>
Heng Zhou, PhD | Postdoctoral research fellow Stefan Raunser lab | Structural Biochemistry Max Planck Institute of Molecular Physiology Otto-Hahn-Straße 11, 44227 Dortmund, Germany Phone: +49 231-133-2326
_______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu> To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu <mailto:chimerax-users-leave@cgl.ucsf.edu> Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/
VARNING: Klicka inte på länkar och öppna inte bilagor om du inte känner igen avsändaren och vet att innehållet är säkert. CAUTION: Do not click on links or open attachments unless you recognise the sender and know the content is safe.
_______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu> To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu <mailto:chimerax-users-leave@cgl.ucsf.edu> Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/
När du har kontakt med oss på Uppsala universitet med e-post så innebär det att vi behandlar dina personuppgifter. För att läsa mer om hur vi gör det kan du läsa här: http://www.uu.se/om-uu/dataskydd-personuppgifter/
E-mailing Uppsala University means that we will process your personal data. For more information on how this is performed, please read here: http://www.uu.se/en/about-uu/data-protection-policy
Hi Alexis, I used to make composite maps by Phenix combine focused maps. This program requires an atomic model split into chains at the seams. If the seams are not at the contact surfaces of two subunits, I have to manually split one chain into several by COOT. And I also need to look very closely at the densities to determine where I make the seam, which may only keep the best density for the map combination. It’s a bit time-consuming, especially for my case with many seams in one structure. So I’m looking for an alternative approach. But thank you anyway. Phenix combine focused maps is indeed a slow but reliable approach. Best regards, Heng
On 27. May 2025, at 21:13, Alexis Rohou <http://a.rohou@gmail.com> wrote:
Hi all,
You may be interested in this phenix program: https://phenix-online.org/documentation/reference/combine_focused_maps.html> Cheers, Alexis
On Tue, May 27, 2025 at 11:42 AM Guillaume Gaullier via ChimeraX-users <chimerax-users@cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu>> wrote:
Hello Heng,
I am not sure there is any consensus in the cryoEM community about composite maps.
Yes, local refinement can produce better maps for individual domains, and yes these maps are fair game to use to build and refine better atomic models of the corresponding domains separately. But it is difficult to say what it the best way to model the entire particle. There are several problems with a composite map obtained by stitching locally refined maps:
1. The composite map does not really represents the underlying particles. If the domains are moving relative to each other around the hinge, then the composite map with a single relative orientation between the two domains only correctly represents very few particles in the dataset. And the resulting atomic model refined against this map will not capture the conformational variability. So, with such a model, you are in effect hiding the fact that the particle is dynamic. One could argue that, in such a case, a more faithful model would be to rigidly place the atomic models of the individual domains obtained from the locally refined maps into the "consensus" map, and refine only their atomic b-factors (or maybe their coordinates too, depending on the resolution of the consensus map, but then likely with tight restraints: either secondary structure restraints or reference model restraints set to the initial atomic models).
2. As you experience first hand, it is very difficult to produce a composite map without artifacts at the "seams". Unphysical density at the seams will cause problems when refining an atomic model against this map.
3. No half-maps exist for the composite map, therefore you cannot determine a global resolution for this map. In a way this is good, because it forces you to make it explicit that it is a composite map.
Tempy-Reff has a feature to produce composite maps without "seam artifacts". I think you first need to refine an atomic model with this program to be able to use this feature. This only addresses problem 2. You would still have no half-maps for the composite maps, and the map and model would still not capture the conformational heterogeneity in the set of particles. https://doi.org/10.1038/s41467-023-44593-1>> https://www.topf-group.org/tempy-reff>> I hope this helps,
Guillaume
On 27 May 2025, at 19:26, Tom Goddard via ChimeraX-users <chimerax-users@cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu>> wrote:
Hi Heng,
I don't have a good idea of how to combine to focused refinement maps so that the overlap region is of best quality. But here are some ideas about what is causing the degradation in quality when you take the maximum at each grid point. If one map has low resolution or has high enough intensity noise where the other map has sharp well-defined density you would hope that the second map has higher intensity at each grid point so that taking a maximum value would ignore the poor quality map. But the low resolution noisy map has broader blobs than the sharp map or blobs of density that don't even overlap the sharp map structure so when you take the maximum all that noise gets added outside the contours of the sharp map density and degrades the quality. I think the only way to remedy that is you can't combine noisy poor quality density when it extends beyond the sharp density of the other map. But the low quality density is probably always going to extend beyond the sharp density. So my conclusion is you to get rid of that poor quality density before you combine the two with maximum, or you combine them using some other function for instance that only puts high density where both maps have high density. The most obvious way to achieve this is to carefully by hand erase the low quality density in both maps near the overlap reagions for example with the ChimeraX Map Eraser tool. But you said you want a more automated way since you have many maps to combine. Possibly if you have local resolution estimates for both maps you could use that to combine them more sensibily than using maximum, but that would require some custom code and I think it is harder than it looks. Another vague idea that would need special purpose code is to define a plane at the hinge and only blend the two maps within a small region (10A either side) near that plane.
Combining focused refinement maps has been a common problem for several years now, so perhaps there is literature that describes better techniques, or exisiting software which does it in better ways than per-grid-point maximum.
Tom
On May 27, 2025, at 4:50 AM, Dr. Heng Zhou via ChimeraX-users <chimerax-users@cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu>> wrote:
Dear ChimeraX developers,
I have an issue when using vop maximum command to combine two local maps. Map A and Map B have an overlapping area with each other. The hinge point of structural flexibility is at the center of the overlapping area. Therefore, the left part of Map A and the right part of Map B have a better local density quality, which I want to keep in the composite map. But vop maximum did it on the opposite, resulting in a worse local resolution at the edges of the overlapping area (e.g., Region 1 and 2).
Technically, this is not an issue of vop maximium itself. In Region 1, Map A has a better local density quality, while Map B has higher values in some voxels. So the composite map has more voxels derived from Map B in Region 1. Vice versa for Region 2.
The only solution in my mind is to split both Map A and B at the hinge point, then do vop maximum for only the left part of Map A and the right part of Map B. But this is time-consuming since I have many local maps to combine. I wonder if you have a better solution? I would appreciate it if you could give advices. Thanks!
Best regards, Heng <http://PastedGraphic-1.png>
Heng Zhou, PhD | Postdoctoral research fellow Stefan Raunser lab | Structural Biochemistry Max Planck Institute of Molecular Physiology Otto-Hahn-Straße 11, 44227 Dortmund, Germany Phone: +49 231-133-2326
_______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu> To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu <mailto:chimerax-users-leave@cgl.ucsf.edu> Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/
VARNING: Klicka inte på länkar och öppna inte bilagor om du inte känner igen avsändaren och vet att innehållet är säkert. CAUTION: Do not click on links or open attachments unless you recognise the sender and know the content is safe.
_______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu> To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu <mailto:chimerax-users-leave@cgl.ucsf.edu> Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/
När du har kontakt med oss på Uppsala universitet med e-post så innebär det att vi behandlar dina personuppgifter. För att läsa mer om hur vi gör det kan du läsa här: http://www.uu.se/om-uu/dataskydd-personuppgifter/
E-mailing Uppsala University means that we will process your personal data. For more information on how this is performed, please read here: http://www.uu.se/en/about-uu/data-protection-policy>> _______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu> To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu <mailto:chimerax-users-leave@cgl.ucsf.edu> Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/
Hi Heng, I am one of the TEMPy-ReFF developers (thanks for suggesting this Guillaume). Indeed, TEMPy-ReFF comes with a script for producing composite maps. We never got round to writing tutorials for using this composite map generation functionality, because compared to refinement, its a bit of a niche use case. Do get in touch if you have any problems. In the meantime I had better write a little tutorial. Best, Tom — Tom Mulvaney Postdoctoral Researcher Centre for Structural Systems Biology (CSSB) Research Group of Prof. Dr. Maya Topf c/o DESY, Building 15 Notkestraße 85, 22607 Hamburg Germany
On 28. May 2025, at 12:21, Dr. Heng Zhou via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:
Hi Guillaume,
Yes, I agree with you, the composite map only represents very few particles in the dataset, and can’t faithfully reflect structural flexibility. So I’m also going to show the "consensus map” and maps from 3D classification in the manuscript (and upload to PDB) to analyze conformational viability. The protein I’m working on is a pentamer, with 3 flexible regions in each protomer. All these flexible regions move independently, so more than half of the areas in the consensus map are blurred or missing. Without a composite map, one can hardly get an impression of what the whole complex looks like. I think the composite map is a plus, as long as it is properly used together with the “real” maps, and with a clear statement that it is a composite map.
The software you mentioned for producing seamless composite maps sounds like a wonderful solution. I will try it out.
Thanks a lot for your suggestions. And glad to discuss with you!
Best, Heng
On 27. May 2025, at 20:38, Guillaume Gaullier <guillaume.gaullier@kemi.uu.se> wrote:
Hello Heng,
I am not sure there is any consensus in the cryoEM community about composite maps.
Yes, local refinement can produce better maps for individual domains, and yes these maps are fair game to use to build and refine better atomic models of the corresponding domains separately. But it is difficult to say what it the best way to model the entire particle. There are several problems with a composite map obtained by stitching locally refined maps:
1. The composite map does not really represents the underlying particles. If the domains are moving relative to each other around the hinge, then the composite map with a single relative orientation between the two domains only correctly represents very few particles in the dataset. And the resulting atomic model refined against this map will not capture the conformational variability. So, with such a model, you are in effect hiding the fact that the particle is dynamic. One could argue that, in such a case, a more faithful model would be to rigidly place the atomic models of the individual domains obtained from the locally refined maps into the "consensus" map, and refine only their atomic b-factors (or maybe their coordinates too, depending on the resolution of the consensus map, but then likely with tight restraints: either secondary structure restraints or reference model restraints set to the initial atomic models).
2. As you experience first hand, it is very difficult to produce a composite map without artifacts at the "seams". Unphysical density at the seams will cause problems when refining an atomic model against this map.
3. No half-maps exist for the composite map, therefore you cannot determine a global resolution for this map. In a way this is good, because it forces you to make it explicit that it is a composite map.
Tempy-Reff has a feature to produce composite maps without "seam artifacts". I think you first need to refine an atomic model with this program to be able to use this feature. This only addresses problem 2. You would still have no half-maps for the composite maps, and the map and model would still not capture the conformational heterogeneity in the set of particles. https://doi.org/10.1038/s41467-023-44593-1 https://www.topf-group.org/tempy-reff
I hope this helps,
Guillaume
On 27 May 2025, at 19:26, Tom Goddard via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:
Hi Heng,
I don't have a good idea of how to combine to focused refinement maps so that the overlap region is of best quality. But here are some ideas about what is causing the degradation in quality when you take the maximum at each grid point. If one map has low resolution or has high enough intensity noise where the other map has sharp well-defined density you would hope that the second map has higher intensity at each grid point so that taking a maximum value would ignore the poor quality map. But the low resolution noisy map has broader blobs than the sharp map or blobs of density that don't even overlap the sharp map structure so when you take the maximum all that noise gets added outside the contours of the sharp map density and degrades the quality. I think the only way to remedy that is you can't combine noisy poor quality density when it extends beyond the sharp density of the other map. But the low quality density is probably always going to extend beyond the sh! arp density. So my conclusion is you to get rid of that poor quality density before you combine the two with maximum, or you combine them using some other function for instance that only puts high density where both maps have high density. The most obvious way to achieve this is to carefully by hand erase the low quality density in both maps near the overlap reagions for example with the ChimeraX Map Eraser tool. But you said you want a more automated way since you have many maps to combine. Possibly if you have local resolution estimates for both maps you could use that to combine them more sensibily than using maximum, but that would require some custom code and I think it is harder than it looks. Another vague idea that would need special purpose code is to define a plane at the hinge and only blend the two maps within a small region (10A either side) near that plane.
Combining focused refinement maps has been a common problem for several years now, so perhaps there is literature that describes better techniques, or exisiting software which does it in better ways than per-grid-point maximum.
Tom
On May 27, 2025, at 4:50 AM, Dr. Heng Zhou via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:
Dear ChimeraX developers,
I have an issue when using vop maximum command to combine two local maps. Map A and Map B have an overlapping area with each other. The hinge point of structural flexibility is at the center of the overlapping area. Therefore, the left part of Map A and the right part of Map B have a better local density quality, which I want to keep in the composite map. But vop maximum did it on the opposite, resulting in a worse local resolution at the edges of the overlapping area (e.g., Region 1 and 2).
Technically, this is not an issue of vop maximium itself. In Region 1, Map A has a better local density quality, while Map B has higher values in some voxels. So the composite map has more voxels derived from Map B in Region 1. Vice versa for Region 2.
The only solution in my mind is to split both Map A and B at the hinge point, then do vop maximum for only the left part of Map A and the right part of Map B. But this is time-consuming since I have many local maps to combine. I wonder if you have a better solution? I would appreciate it if you could give advices. Thanks!
Best regards, Heng <PastedGraphic-1.png>
Heng Zhou, PhD | Postdoctoral research fellow Stefan Raunser lab | Structural Biochemistry Max Planck Institute of Molecular Physiology Otto-Hahn-Straße 11, 44227 Dortmund, Germany Phone: +49 231-133-2326
_______________________________________________ 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/
VARNING: Klicka inte på länkar och öppna inte bilagor om du inte känner igen avsändaren och vet att innehållet är säkert. CAUTION: Do not click on links or open attachments unless you recognise the sender and know the content is safe.
_______________________________________________ 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/
Page Title
När du har kontakt med oss på Uppsala universitet med e-post så innebär det att vi behandlar dina personuppgifter. För att läsa mer om hur vi gör det kan du läsa här: http://www.uu.se/om-uu/dataskydd-personuppgifter/
E-mailing Uppsala University means that we will process your personal data. For more information on how this is performed, please read here: http://www.uu.se/en/about-uu/data-protection-policy
_______________________________________________ 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 (5)
-
Alexis Rohou -
Dr. Heng Zhou -
Guillaume Gaullier -
Thomas Mulvaney -
Tom Goddard