Re: [Chimera-users] scale problem

Dear all, I faced a scale problem when I was trying to fit the density map. A density map was reconstructed from a set of projections which their magnitude are 1.055A/pixel, with a reference model that is 1.055 A/voxel and 5 A in resolution. However the pdb file cannot fit into the map because the map is too much small. So I have to change the voxel size to double and then can be fitted. Here I changed the voxel size, I think it changed the scale for the density. So i just worried about the fitting result. thanks On Sun, Nov 22, 2015 at 10:07 AM, Yaowang Li <liyaowang2911@gmail.com> wrote:
Dear all,
I faced a scale problem when I was trying to fit the density map.
A density map was reconstructed from a set of projections which their magnitude are 1.055A/pixel, with a reference model that is 1.055 A/voxel and 5 A in resolution. However the pdb file cannot fit into the map because the map is too much small. So I have to change the voxel size to double and then can be fitted.
Here I changed the voxel size, I think it changed the scale for the density. So i just worried about the fitting result.
thanks
yaowang

Hi Yaowang, The Chimera Volume Viewer voxel size setting (menu Features / Coordinates) changes the spacing between grid points. It does not change the map intensity values. If it appears to be wrong by a factor of 2 when compared to a PDB file then the voxel size is definitely. I would not just guess that it is a factor of 2 too small — you should find out from whoever did the reconstruction what the correct value is. But if 2.11A is the correct size than your fitting is fine. Tom
On Nov 23, 2015, at 1:31 AM, Yaowang Li wrote:
Dear all,
I faced a scale problem when I was trying to fit the density map.
A density map was reconstructed from a set of projections which their magnitude are 1.055A/pixel, with a reference model that is 1.055 A/voxel and 5 A in resolution. However the pdb file cannot fit into the map because the map is too much small. So I have to change the voxel size to double and then can be fitted.
Here I changed the voxel size, I think it changed the scale for the density. So i just worried about the fitting result.
thanks

Dear Tom, Yeah, it indeed did not change the intensity, I made a mistake, and did not make it clear. I have a little experience on the this, that is my first time. As I stated, the magnitude of projections is 1.055 A/pixel, so I think the volume size of density map should be 1.055 A/voxel. A PDB file should be fitted into the density map theoretically I think, while apparently the density map is small compare to PDB. That is why I changed the voxel size to 2.11 A/voxel. So actually I want to make a confirmation that the grid in the density map whether is 1.055 A/voxel or not. if it is right, I guess something is wrong during the reconstruction. Regards, yaowang On Mon, Nov 23, 2015 at 7:50 PM, Tom Goddard <goddard@sonic.net> wrote:
Hi Yaowang,
The Chimera Volume Viewer voxel size setting (menu Features / Coordinates) changes the spacing between grid points. It does not change the map intensity values. If it appears to be wrong by a factor of 2 when compared to a PDB file then the voxel size is definitely. I would not just guess that it is a factor of 2 too small — you should find out from whoever did the reconstruction what the correct value is. But if 2.11A is the correct size than your fitting is fine.
Tom
On Nov 23, 2015, at 1:31 AM, Yaowang Li wrote:
Dear all,
I faced a scale problem when I was trying to fit the density map.
A density map was reconstructed from a set of projections which their magnitude are 1.055A/pixel, with a reference model that is 1.055 A/voxel and 5 A in resolution. However the pdb file cannot fit into the map because the map is too much small. So I have to change the voxel size to double and then can be fitted.
Here I changed the voxel size, I think it changed the scale for the density. So i just worried about the fitting result.
thanks

Hi Yaowang, I don’t know what the correct grid spacing for your density map is, although it sounds like by comparing to a PDB model you think it is around 2.11 Angstroms. When Sparky opens the file it reads the grid spacing from the file header if the header has the information. But that value might be wrong. Just because your projection images had 1.055 Angstrom pixel size doesn’t mean the 3d reconstruction will use the same pixel size. You will have to get the correct grid spacing from the person who did the reconstruction. Tom
On Nov 24, 2015, at 7:17 AM, Yaowang Li <liyaowang2911@gmail.com> wrote:
Dear Tom,
Yeah, it indeed did not change the intensity, I made a mistake, and did not make it clear.
I have a little experience on the this, that is my first time. As I stated, the magnitude of projections is 1.055 A/pixel, so I think the volume size of density map should be 1.055 A/voxel.
A PDB file should be fitted into the density map theoretically I think, while apparently the density map is small compare to PDB. That is why I changed the voxel size to 2.11 A/voxel.
So actually I want to make a confirmation that the grid in the density map whether is 1.055 A/voxel or not. if it is right, I guess something is wrong during the reconstruction.
Regards,
yaowang
On Mon, Nov 23, 2015 at 7:50 PM, Tom Goddard <goddard@sonic.net <mailto:goddard@sonic.net>> wrote: Hi Yaowang,
The Chimera Volume Viewer voxel size setting (menu Features / Coordinates) changes the spacing between grid points. It does not change the map intensity values. If it appears to be wrong by a factor of 2 when compared to a PDB file then the voxel size is definitely. I would not just guess that it is a factor of 2 too small — you should find out from whoever did the reconstruction what the correct value is. But if 2.11A is the correct size than your fitting is fine.
Tom
On Nov 23, 2015, at 1:31 AM, Yaowang Li wrote:
Dear all,
I faced a scale problem when I was trying to fit the density map.
A density map was reconstructed from a set of projections which their magnitude are 1.055A/pixel, with a reference model that is 1.055 A/voxel and 5 A in resolution. However the pdb file cannot fit into the map because the map is too much small. So I have to change the voxel size to double and then can be fitted.
Here I changed the voxel size, I think it changed the scale for the density. So i just worried about the fitting result.
thanks

Hi Tom, I did the reconstruction. As you said, there is no relationship between pixel size of projection and the voxel size of density map? On Tue, Nov 24, 2015 at 6:08 PM, Tom Goddard <goddard@sonic.net> wrote:
Hi Yaowang,
I don’t know what the correct grid spacing for your density map is, although it sounds like by comparing to a PDB model you think it is around 2.11 Angstroms. When Sparky opens the file it reads the grid spacing from the file header if the header has the information. But that value might be wrong. Just because your projection images had 1.055 Angstrom pixel size doesn’t mean the 3d reconstruction will use the same pixel size. You will have to get the correct grid spacing from the person who did the reconstruction.
Tom
On Nov 24, 2015, at 7:17 AM, Yaowang Li <liyaowang2911@gmail.com> wrote:
Dear Tom,
Yeah, it indeed did not change the intensity, I made a mistake, and did not make it clear.
I have a little experience on the this, that is my first time. As I stated, the magnitude of projections is 1.055 A/pixel, so I think the volume size of density map should be 1.055 A/voxel.
A PDB file should be fitted into the density map theoretically I think, while apparently the density map is small compare to PDB. That is why I changed the voxel size to 2.11 A/voxel.
So actually I want to make a confirmation that the grid in the density map whether is 1.055 A/voxel or not. if it is right, I guess something is wrong during the reconstruction.
Regards,
yaowang
On Mon, Nov 23, 2015 at 7:50 PM, Tom Goddard <goddard@sonic.net> wrote:
Hi Yaowang,
The Chimera Volume Viewer voxel size setting (menu Features / Coordinates) changes the spacing between grid points. It does not change the map intensity values. If it appears to be wrong by a factor of 2 when compared to a PDB file then the voxel size is definitely. I would not just guess that it is a factor of 2 too small — you should find out from whoever did the reconstruction what the correct value is. But if 2.11A is the correct size than your fitting is fine.
Tom
On Nov 23, 2015, at 1:31 AM, Yaowang Li wrote:
Dear all,
I faced a scale problem when I was trying to fit the density map.
A density map was reconstructed from a set of projections which their magnitude are 1.055A/pixel, with a reference model that is 1.055 A/voxel and 5 A in resolution. However the pdb file cannot fit into the map because the map is too much small. So I have to change the voxel size to double and then can be fitted.
Here I changed the voxel size, I think it changed the scale for the density. So i just worried about the fitting result.
thanks
participants (2)
-
Tom Goddard
-
Yaowang Li