How to set the contour level correctly?
Hello, my apologies if this is a recurring (and potentially trivial) question that does not directly concern Chimeras functionalities. I am wondering how to properly set the value of the contour level to visualize a Coulomb potential map. When I use the approximation of 1 Da = 1.2 [cid:image003.png@01D7C06B.AA8F9850] ³ to adjust the displayed volume in the Volume viewer dialog almost no secondary structure, but a lot of dust is visible. The same problem occurs when I’d like to visualize, for example, an α-helix of a subunit rigid-body fitted to the density. I use the “zone” command to trim the density at a pre-defined 2 Å around the α-helix. But is there any general guideline on how to properly set the contour level? I often find values between 1 σ to 3 σ. I assume this is the standard deviation of the map. If so, can it be measured via the “Volume Mean, SD, RMS” command? Kind regards Phil
Hi Philipp, I can't tell if you are talking about electron density like from cryoEM, or a Coulombic electrostatic potential (ESP) map. Either way, the units shown in the Volume Viewer are not sigma, they are the raw values from the map file. The "Volume Mean, SD, RMS" tool (or command "measure mapStats") will tell you the mean, standard deviation (SD) from the mean, and the root-mean-square (RMS) deviation from zero but won't affect what is shown in the Volume Viewer. You'd have to use the stats to back-calculate the contour value that you want, or use command "volume" with "sdLevel" or "rmsLevel" options to do the conversion and set the contour level directly: <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/volume.html#general> If electron density: I don't think there is any universally recommended value. It depends on the resolution, noise, and other specifics of a given experiment. If ESP: the Coulombic ESP calculation is mainly for coloring the molecular surface to see relatively positive or negative regions, rather than for display as isopotential contour surfaces. The values are in kcal/(mol·e) at 298 K. However, the absolute magnitudes of the values are not that important as you can clearly halve them by doubling the dieletric constant, or double them by halving the dielectric constant, etc. Thus it is not meant for quantitative calculations, and there is specific recommended contour level for isosurface display. <https://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/coulombic/coulomb...> I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Oct 13, 2021, at 11:31 AM, Selenschik, Philipp via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Hello, my apologies if this is a recurring (and potentially trivial) question that does not directly concern Chimeras functionalities. I am wondering how to properly set the value of the contour level to visualize a Coulomb potential map. When I use the approximation of 1 Da = 1.2 <image003.png>³ to adjust the displayed volume in the Volume viewer dialog almost no secondary structure, but a lot of dust is visible. The same problem occurs when I’d like to visualize, for example, an α-helix of a subunit rigid-body fitted to the density. I use the “zone” command to trim the density at a pre-defined 2 Å around the α-helix. But is there any general guideline on how to properly set the contour level? I often find values between 1 σ to 3 σ. I assume this is the standard deviation of the map. If so, can it be measured via the “Volume Mean, SD, RMS” command? Kind regards Phil
Hi Elaine, thanks for your quick and helpful reply! I should have been more specific, sorry. My question was about an electron density map of a single particle cryo-EM reconstruction. Concerning the back-calculation, I found an older mailing list entry: https://www.cgl.ucsf.edu/pipermail/chimera-users/2008-May/002651.html If I understand things correctly: Depending on which normalization is used, the value of the volume dialog to obtain a given (e.g. 3) sigma value can be determined based on the measured SD, RMS and mean by using the relations given above. Is this what “sdLevel” or “rmsLevel” options of the “Volume” command do? I would like to show one representative helix of a subunit with its surrounding density (the subunit is fitted into the map). To do this, I delete all residues except the ones of interest, select the atoms of the helix and use the zone function in the Volume Viewer to display the electron density locally. But (how) does the radius at which the density is shown relate to/affect the contour level? Are these dependent parameters? Ideally, I would like to crop the residues of interest at the previously set contour level of the map. Probably there is a more elegant approach to achieve this. Kind regards Philip ________________________________ Von: Elaine Meng <meng@cgl.ucsf.edu> Gesendet: Mittwoch, 13. Oktober 2021 21:29:49 An: Selenschik, Philipp Cc: chimera-users@cgl.ucsf.edu Betreff: Re: [Chimera-users] How to set the contour level correctly? Hi Philipp, I can't tell if you are talking about electron density like from cryoEM, or a Coulombic electrostatic potential (ESP) map. Either way, the units shown in the Volume Viewer are not sigma, they are the raw values from the map file. The "Volume Mean, SD, RMS" tool (or command "measure mapStats") will tell you the mean, standard deviation (SD) from the mean, and the root-mean-square (RMS) deviation from zero but won't affect what is shown in the Volume Viewer. You'd have to use the stats to back-calculate the contour value that you want, or use command "volume" with "sdLevel" or "rmsLevel" options to do the conversion and set the contour level directly: <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/volume.html#general> If electron density: I don't think there is any universally recommended value. It depends on the resolution, noise, and other specifics of a given experiment. If ESP: the Coulombic ESP calculation is mainly for coloring the molecular surface to see relatively positive or negative regions, rather than for display as isopotential contour surfaces. The values are in kcal/(mol·e) at 298 K. However, the absolute magnitudes of the values are not that important as you can clearly halve them by doubling the dieletric constant, or double them by halving the dielectric constant, etc. Thus it is not meant for quantitative calculations, and there is specific recommended contour level for isosurface display. <https://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/coulombic/coulomb...> I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Oct 13, 2021, at 11:31 AM, Selenschik, Philipp via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Hello, my apologies if this is a recurring (and potentially trivial) question that does not directly concern Chimeras functionalities. I am wondering how to properly set the value of the contour level to visualize a Coulomb potential map. When I use the approximation of 1 Da = 1.2 <image003.png>³ to adjust the displayed volume in the Volume viewer dialog almost no secondary structure, but a lot of dust is visible. The same problem occurs when I’d like to visualize, for example, an α-helix of a subunit rigid-body fitted to the density. I use the “zone” command to trim the density at a pre-defined 2 Å around the α-helix. But is there any general guideline on how to properly set the contour level? I often find values between 1 σ to 3 σ. I assume this is the standard deviation of the map. If so, can it be measured via the “Volume Mean, SD, RMS” command? Kind regards Phil
Hi Philipp, The zone radius is independent of the contour level. Think of it in two stages: first the contour level is applied, and then all the stuff beyond the zone radius is hidden. If you want to mask the map by the contour surface, see the "mask" command instead. <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/mask.html> The sdLevel and rmsLevel options are in the help link below. If either one fits your definition of sigma, then using the option with 3 would give contour at 3 sigma. <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/volume.html#general> I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Oct 16, 2021, at 6:43 AM, Selenschik, Philipp via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Hi Elaine, thanks for your quick and helpful reply! I should have been more specific, sorry. My question was about an electron density map of a single particle cryo-EM reconstruction. Concerning the back-calculation, I found an older mailing list entry: https://www.cgl.ucsf.edu/pipermail/chimera-users/2008-May/002651.html If I understand things correctly: Depending on which normalization is used, the value of the volume dialog to obtain a given (e.g. 3) sigma value can be determined based on the measured SD, RMS and mean by using the relations given above. Is this what “sdLevel” or “rmsLevel” options of the “Volume” command do? I would like to show one representative helix of a subunit with its surrounding density (the subunit is fitted into the map). To do this, I delete all residues except the ones of interest, select the atoms of the helix and use the zone function in the Volume Viewer to display the electron density locally. But (how) does the radius at which the density is shown relate to/affect the contour level? Are these dependent parameters? Ideally, I would like to crop the residues of interest at the previously set contour level of the map. Probably there is a more elegant approach to achieve this.
Kind regards
Philip Von: Elaine Meng <meng@cgl.ucsf.edu> Gesendet: Mittwoch, 13. Oktober 2021 21:29:49 An: Selenschik, Philipp Cc: chimera-users@cgl.ucsf.edu Betreff: Re: [Chimera-users] How to set the contour level correctly?
Hi Philipp, I can't tell if you are talking about electron density like from cryoEM, or a Coulombic electrostatic potential (ESP) map. Either way, the units shown in the Volume Viewer are not sigma, they are the raw values from the map file. The "Volume Mean, SD, RMS" tool (or command "measure mapStats") will tell you the mean, standard deviation (SD) from the mean, and the root-mean-square (RMS) deviation from zero but won't affect what is shown in the Volume Viewer. You'd have to use the stats to back-calculate the contour value that you want, or use command "volume" with "sdLevel" or "rmsLevel" options to do the conversion and set the contour level directly: <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/volume.html#general>
If electron density: I don't think there is any universally recommended value. It depends on the resolution, noise, and other specifics of a given experiment.
If ESP: the Coulombic ESP calculation is mainly for coloring the molecular surface to see relatively positive or negative regions, rather than for display as isopotential contour surfaces. The values are in kcal/(mol·e) at 298 K. However, the absolute magnitudes of the values are not that important as you can clearly halve them by doubling the dieletric constant, or double them by halving the dielectric constant, etc. Thus it is not meant for quantitative calculations, and there is specific recommended contour level for isosurface display. <https://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/coulombic/coulomb...>
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Oct 13, 2021, at 11:31 AM, Selenschik, Philipp via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Hello, my apologies if this is a recurring (and potentially trivial) question that does not directly concern Chimeras functionalities. I am wondering how to properly set the value of the contour level to visualize a Coulomb potential map. When I use the approximation of 1 Da = 1.2 <image003.png>³ to adjust the displayed volume in the Volume viewer dialog almost no secondary structure, but a lot of dust is visible. The same problem occurs when I’d like to visualize, for example, an α-helix of a subunit rigid-body fitted to the density. I use the “zone” command to trim the density at a pre-defined 2 Å around the α-helix. But is there any general guideline on how to properly set the contour level? I often find values between 1 σ to 3 σ. I assume this is the standard deviation of the map. If so, can it be measured via the “Volume Mean, SD, RMS” command? Kind regards Phil
Hi Elaine, thanks for suggesting the mask option! For the latter, I use the estimated 0.143 cutoff FSC resolution of the density map (2.5 Å) to create a molmap of the helix: molmap #1 2.5 and then use that map to mask the electron density: mask #0 #1.1 However, the masked volume seems to depend on the contour level of the (intermediate) molmap. I've set it to 3 σ, just like for the original density map. Concerning the "zone" function, I think that although the contour level is not directly affected by the zone radius, the density is still clipped below a certain threshold. I wonder which approach (zone or mask) leads to least artifacts when extracting the original density? I haven't tried "Segger" yet. Best wishes Philipp -----Ursprüngliche Nachricht----- Von: Elaine Meng <meng@cgl.ucsf.edu> Gesendet: Samstag, 16. Oktober 2021 17:50 An: Selenschik, Philipp <pselensc@uni-muenster.de> Cc: chimera-users@cgl.ucsf.edu List <chimera-users@cgl.ucsf.edu> Betreff: Re: [Chimera-users] How to set the contour level correctly? Hi Philipp, The zone radius is independent of the contour level. Think of it in two stages: first the contour level is applied, and then all the stuff beyond the zone radius is hidden. If you want to mask the map by the contour surface, see the "mask" command instead. <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/mask.html> The sdLevel and rmsLevel options are in the help link below. If either one fits your definition of sigma, then using the option with 3 would give contour at 3 sigma. <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/volume.html#general> I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
Hi Philipp, The "mask" command just masks the first map by a surface (not caring that the surface is for another map), so yes, it does depend exactly on that contour surface. Using some specific sigma may not be that meaningful, instead you may want to contour the molmap far enough away from the helix to include whatever density from the first map that you think should be included. I don't know what you mean by "artifacts." "mask" command <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/mask.html> As I understand it, zoning is not for editing the data so much as for visualization; it just hides the density rather than deleting it. For deleting/zeroing density see masking, cropping, and/or volume eraser. I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
On Oct 18, 2021, at 6:14 AM, Philip via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Hi Elaine,
thanks for suggesting the mask option! For the latter, I use the estimated 0.143 cutoff FSC resolution of the density map (2.5 Å) to create a molmap of the helix: molmap #1 2.5 and then use that map to mask the electron density: mask #0 #1.1 However, the masked volume seems to depend on the contour level of the (intermediate) molmap. I've set it to 3 σ, just like for the original density map. Concerning the "zone" function, I think that although the contour level is not directly affected by the zone radius, the density is still clipped below a certain threshold. I wonder which approach (zone or mask) leads to least artifacts when extracting the original density? I haven't tried "Segger" yet.
Best wishes
Philipp
-----Ursprüngliche Nachricht----- Von: Elaine Meng <meng@cgl.ucsf.edu> Gesendet: Samstag, 16. Oktober 2021 17:50 An: Selenschik, Philipp <pselensc@uni-muenster.de> Cc: chimera-users@cgl.ucsf.edu List <chimera-users@cgl.ucsf.edu> Betreff: Re: [Chimera-users] How to set the contour level correctly?
Hi Philipp, The zone radius is independent of the contour level. Think of it in two stages: first the contour level is applied, and then all the stuff beyond the zone radius is hidden. If you want to mask the map by the contour surface, see the "mask" command instead. <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/mask.html>
The sdLevel and rmsLevel options are in the help link below. If either one fits your definition of sigma, then using the option with 3 would give contour at 3 sigma. <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/volume.html#general>
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
_______________________________________________ Chimera-users mailing list: Chimera-users@cgl.ucsf.edu Manage subscription: https://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
Hi Elaine and Tom, thanks for your patience and help! I got it now :) What I meant by "artifacts" in my previous mail were parts of the volume that are shown due to non-ideal parameters (either radius for (surface) zone or contour level for the mask approach). All the best Philipp -----Ursprüngliche Nachricht----- Von: Elaine Meng <meng@cgl.ucsf.edu> Gesendet: Montag, 18. Oktober 2021 19:15 An: pselensc@wwu.de Cc: chimera-users@cgl.ucsf.edu Betreff: Re: [Chimera-users] How to set the contour level correctly? Hi Philipp, The "mask" command just masks the first map by a surface (not caring that the surface is for another map), so yes, it does depend exactly on that contour surface. Using some specific sigma may not be that meaningful, instead you may want to contour the molmap far enough away from the helix to include whatever density from the first map that you think should be included. I don't know what you mean by "artifacts." "mask" command <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/mask.html> As I understand it, zoning is not for editing the data so much as for visualization; it just hides the density rather than deleting it. For deleting/zeroing density see masking, cropping, and/or volume eraser. I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
Hi Philipp, I suggest just using surface zone to show density around a helix. An example of how to do that in ChimeraX is in the "Inspecting side chain density" section of this tutorial https://www.rbvi.ucsf.edu/chimerax/data/stanford-jul2019/tutorial.html Chimera has the same capabilities and is the predecessor of ChimeraX. Commands are similar but not identical. Surface zone just hides the surface beyond the specified distance from the atoms. The surface will look cutoff in some places where it extends beyond that distance. That is a good thing so a person can see the density is cutoff. Masking disguises those cuts by capping the cutoff density and is misleading, so I don't recommend it. Tom
On Oct 18, 2021, at 6:14 AM, Philip via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Hi Elaine,
thanks for suggesting the mask option! For the latter, I use the estimated 0.143 cutoff FSC resolution of the density map (2.5 Å) to create a molmap of the helix: molmap #1 2.5 and then use that map to mask the electron density: mask #0 #1.1 However, the masked volume seems to depend on the contour level of the (intermediate) molmap. I've set it to 3 σ, just like for the original density map. Concerning the "zone" function, I think that although the contour level is not directly affected by the zone radius, the density is still clipped below a certain threshold. I wonder which approach (zone or mask) leads to least artifacts when extracting the original density? I haven't tried "Segger" yet.
Best wishes
Philipp
-----Ursprüngliche Nachricht----- Von: Elaine Meng <meng@cgl.ucsf.edu> Gesendet: Samstag, 16. Oktober 2021 17:50 An: Selenschik, Philipp <pselensc@uni-muenster.de> Cc: chimera-users@cgl.ucsf.edu List <chimera-users@cgl.ucsf.edu> Betreff: Re: [Chimera-users] How to set the contour level correctly?
Hi Philipp, The zone radius is independent of the contour level. Think of it in two stages: first the contour level is applied, and then all the stuff beyond the zone radius is hidden. If you want to mask the map by the contour surface, see the "mask" command instead. <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/mask.html>
The sdLevel and rmsLevel options are in the help link below. If either one fits your definition of sigma, then using the option with 3 would give contour at 3 sigma. <https://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/volume.html#general>
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Department of Pharmaceutical Chemistry University of California, San Francisco
_______________________________________________ Chimera-users mailing list: Chimera-users@cgl.ucsf.edu Manage subscription: https://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
participants (4)
-
Elaine Meng -
pselensc＠wwu.de -
Selenschik, Philipp -
Tom Goddard