SAXS

Small Angle X-ray Scattering Initiative for Europe :: Forum
It is currently 2018.02.24 18:00

All times are UTC + 2 hours




Post new topic Reply to topic  [ 6 posts ] 
Author Message
PostPosted: 2016.06.21 19:15 
Offline
Active member

Joined: 2011.09.07 15:53
Posts: 70
Location: Stanford
Hi!

I am using EM2DAM to convert an electron density map to a coarse-grained dummy-atom model. However the meaning of the output is not entirely clear to me. For example, I get
Code:
> em2dam emd_6584.map -t 0.095 -o emd_6584_dam.pdb
[...]
> head -30 emd_6584_dam.pdb
REMARK                                                       
REMARK     Created by E M 2 D A M     Version 1.5  28/12/2011
REMARK                                                       
REMARK     Electron Microscopy density map 2 Dummy Atom Model
REMARK                                                       
REMARK     AUTHOR:  Alexander V. Shkumatov, EMBL HAMBURG     
REMARK                                                       
REMARK    Density map parameters:
REMARK     EM density map file name .................... : emd_6584.map
REMARK     Minimum threshold ........................... : -0.3867
REMARK     Maximum threshold ........................... : 0.6159
REMARK     Absolute  minimum ........................... : 0.0
REMARK     First number read ........................... : 0.0
REMARK     Last  number read ........................... : 0.0
REMARK     Average   density ........................... : 3.011e-3
REMARK
REMARK    Parameters for processing the map:
REMARK     Number of skipped bytes ..................... : 1024
REMARK     Map dimensions --- Ni, Nj, Nk:  324 324 324                                                                 
REMARK     Pixel size .................................. : 1.280                                                       
REMARK     Pixel volume ................................ : 2.097                                                       
REMARK     Threshold ................................... : 9.500e-2                                                     
REMARK     Model Volume ................................ : 1.077e+6                                                     
REMARK     Output PDB file name ........................ : emd_6584_dam.pdb
REMARK
ATOM      1  CA  ASP    1      -58.880 -48.640 -94.720  1.00 20.00 0 2 201
ATOM      2  CA  ASP    1      -57.600 -44.800 -93.440  1.00 20.00 0 2 201
ATOM      3  CA  ASP    1      -62.720 -43.520 -93.440  1.00 20.00 0 2 201
ATOM      4  CA  ASP    1      -57.600 -43.520 -93.440  1.00 20.00 0 2 201
ATOM      5  CA  ASP    1      -62.720 -42.240 -93.440  1.00 20.00 0 2 201

Using a threshold of 0.095, em2dam creates a model with volume 1.077e+6 ų. Since each voxel has a volume of 2.097 ų the number of dummy atoms I would expect in the model should be 1M/2 = 500k. But it creates only about 10k.
Code:
> tail -10 emd_6584_dam.pdb
ATOM  10637  CA  ASP 1064       16.640 -42.240 163.840  1.00 20.00 0 2 201
ATOM  10638  CA  ASP 1064       15.360 -40.960 163.840  1.00 20.00 0 2 201
ATOM  10639  CA  ASP 1064       16.640 -40.960 163.840  1.00 20.00 0 2 201
ATOM  10640  CA  ASP 1065       26.880 -40.960 166.400  1.00 20.00 0 2 201
ATOM  10641  CA  ASP 1065       14.080 -32.000 166.400  1.00 20.00 0 2 201
ATOM  10642  CA  ASP 1065       15.360 -32.000 166.400  1.00 20.00 0 2 201
ATOM  10643  CA  ASP 1065       16.640 -32.000 166.400  1.00 20.00 0 2 201
ATOM  10644  CA  ASP 1065       15.360 -30.720 167.680  1.00 20.00 0 2 201
TER
END

1) What am I missing? How to interpret the numbers in the output? (I could not find this in the documentation)
2) Ultimately, I would like to know the electron density per dummy atom. Is there a way to find out?

Thanks a bunch for any comment!


Top
 Profile  
 
PostPosted: 2016.06.21 22:08 
Offline
Active member

Joined: 2007.08.09 21:10
Posts: 692
Location: Planet Earth
Quote:
I am using EM2DAM

nice to hear someone is using it:)

Quote:
Using a threshold of 0.095, em2dam creates a model with volume 1.077e+6 ų. Since each voxel has a volume of 2.097 ų the number of dummy atoms I would expect in the model should be 1M/2 = 500k. But it creates only about 10k.

No, you are absolutely right. However, PDB format does not allow more than certain number of digits (more than 9999) so the counter re-sets

If you do simple counting, you get expected number:
em2dam alexander.shkumatov$ cat emd_6584_dam.pdb | grep ASP | WC
510589 6622161 38294175

Quote:
1) What am I missing? How to interpret the numbers in the output? (I could not find this in the documentation)

Typically one "residue" consists of 10 atoms. I think the one at beginning/end might be less, e.g.:
ATOM 99980 CA ASP 9999 -65.280 -11.520 -10.240 1.00 20.00 0 2 201
ATOM 99981 CA ASP 9999 -5.120 -11.520 -10.240 1.00 20.00 0 2 201
ATOM 99982 CA ASP 9999 -3.840 -11.520 -10.240 1.00 20.00 0 2 201
ATOM 99983 CA ASP 9999 0.000 -11.520 -10.240 1.00 20.00 0 2 201
ATOM 99984 CA ASP 9999 1.280 -11.520 -10.240 1.00 20.00 0 2 201
ATOM 99985 CA ASP 9999 2.560 -11.520 -10.240 1.00 20.00 0 2 201
ATOM 99986 CA ASP 9999 6.400 -11.520 -10.240 1.00 20.00 0 2 201
ATOM 99987 CA ASP 9999 10.240 -11.520 -10.240 1.00 20.00 0 2 201
ATOM 99988 CA ASP 9999 11.520 -11.520 -10.240 1.00 20.00 0 2 201
ATOM 99989 CA ASP 9999 38.400 -11.520 -10.240 1.00 20.00 0 2 201

I guess you can recognize X, Y, Z
Then for changeable phase, there is '1.00 20.00 0 2 201'
and for fixed phase '1.00 20.00 1 2 201'.
This is used during DAMMIN refinement.

Quote:
2) Ultimately, I would like to know the electron density per dummy atom. Is there a way to find out?

each dummy atom is roughly assumed to be a C-alpha atom of ASP residue.


HTH, Alex


Top
 Profile  
 
PostPosted: 2016.06.21 22:30 
Offline
Active member

Joined: 2011.09.07 15:53
Posts: 70
Location: Stanford
Thanks for your quick response!

Alex wrote:
If you do simple counting, you get expected number:
em2dam alexander.shkumatov$ cat emd_6584_dam.pdb | grep ASP | WC
510589 6622161 38294175

Oh yes. forgot about that. Thanks!
Alex wrote:
Quote:
1) What am I missing? How to interpret the numbers in the output? (I could not find this in the documentation)

Typically one "residue" consists of 10 atoms. I think the one at beginning/end might be less, e.g.:
ATOM 99980 CA ASP 9999 -65.280 -11.520 -10.240 1.00 20.00 0 2 201
[...]
I guess you can recognize X, Y, Z
Then for changeable phase, there is '1.00 20.00 0 2 201'
and for fixed phase '1.00 20.00 1 2 201'.
This is used during DAMMIN refinement.

Thanks for the short reminder on pdb format. However, unclear to me is the header of the output file. What do these numbers mean? E.g. model volume, this seems to be only number changing when I use different values for "threshold".
Alex wrote:
Quote:
2) Ultimately, I would like to know the electron density per dummy atom. Is there a way to find out?

each dummy atom is roughly assumed to be a C-alpha atom of ASP residue.

Hmm. I thought em2dam places the dummy atoms at the positions of the voxels with a density value higher or equal to the set contour level threshold. But if that's true then the density of each dummy atom should change depending on the set threshold, right?
Maybe I am missing a good understanding of the EM contour level here...


Top
 Profile  
 
PostPosted: 2016.06.22 11:18 
Offline
Active member

Joined: 2007.08.09 21:10
Posts: 692
Location: Planet Earth
Quote:
E.g. model volume, this seems to be only number changing when I use different values for "threshold".

basically, just like you calculated earlier: pixel volume * number of beads (atoms) = model volume
Map dimension is either specified by user or extracted from file header (see MRC format for details).

The rest parameters are extracted from density map (quite standard stuff, which you can also get using programs in EMAN package).
Minimum threshold ........................... : -0.3867
REMARK Maximum threshold ........................... : 0.6159
REMARK Absolute minimum ........................... : 0.0
REMARK First number read ........................... : 0.0
REMARK Last number read ........................... : 0.0
REMARK Average density ........................... : 3.011e-3

Quote:
Hmm. I thought em2dam places the dummy atoms at the positions of the voxels with a density value higher or equal to the set contour level threshold..

that is correct. .. but the positions of the voxels with a density value higher to the set contour level threshold.

Quote:
But if that's true then the density of each dummy atom should change depending on the set threshold, right?

not really, it will place a dummy atoms in the place where density is more than threshold, if less - no dummy atom is placed. The density of dummy atoms is uniform. it is
an approximation, where each voxel (should it have density higher than certain level) will be replaced by dummy atom.


Top
 Profile  
 
PostPosted: 2016.06.22 20:20 
Offline
Active member

Joined: 2011.09.07 15:53
Posts: 70
Location: Stanford
Alex wrote:
Quote:
But if that's true then the density of each dummy atom should change depending on the set threshold, right?

not really, it will place a dummy atoms in the place where density is more than threshold, if less - no dummy atom is placed. The density of dummy atoms is uniform. it is
an approximation, where each voxel (should it have density higher than certain level) will be replaced by dummy atom.

So that means ideally, I should use a threshold that will lead to an average density equal to the density of the dummy atoms, right?
You say the dummy atoms have a density of a C-alpha in ASP. So I wonder how one could find the corresponding threshold in EM maps. Do you have some guidance or rule of thumb? Or could you point to any literature or so where to find out more?


Top
 Profile  
 
PostPosted: 2016.06.23 09:58 
Offline
Active member

Joined: 2007.08.09 21:10
Posts: 692
Location: Planet Earth
Quote:
So that means ideally, I should use a threshold that will lead to an average density equal to the density of the dummy atoms, right?
You say the dummy atoms have a density of a C-alpha in ASP. So I wonder how one could find the corresponding threshold in EM maps. Do you have some guidance or rule of thumb? Or could you point to any literature or so where to find out more?

Threshold is determined by person who is doing the analysis. There are some criteria for that but that is out of the scope of this forum.
See em2dam as a mean to 'quick-and-dirty' convert density map into pseudo PDB file.
At the end, it all depends what you will use it for, e.g., I saw several papers where em2dam was used to convert density map to bead model
and fit it to SAXS data. Of course if you want to be more accurate, i strongly recommend to fit all available partial structures/models in density
map and validate resulting PDB.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 6 posts ] 

All times are UTC + 2 hours


Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group