Hi there,
To my knowledge the PDB does not include unmerged data in entries - you are 
required to deposit the data that the coordinates were refined against. 

I believe you can deposit other data which will be archived and can be 
requested by other users - problem is that none of the distribution web sites 
fesses up to the existence of these extra datasets as part of the entries - so 
users of your structure will be ignorant of them.

You could ask for an authors' remark to be included in the appropriate part of 
your coordinate header to alert users to the archived material. 

Currently you can upload data as mtz files and this format would of course be 
most that would be the most convenient. Of course the mmcif format could 
include the deposition of unmerged data.

Currently some distributed mmcif SF files in fact have multiple datasets - so 
for example relating to other crystals or other wavelengths. So you could 
simply provide your unmerged data as an mmcif block to include at the end of 
the dataset used for refinement. I'm guessing eds and pdbredo just pick up the 
first block as representative of the entry. But you can include other stuff 
after it. Again it would be useful to ask for explanatory remarks to be 
included for users of the data.

For example, the authors of the excellent structure 3ne5 submitted a 'phasing 
dataset' with their SFs. This appears (after 'Format standardization' 
2013-08-28) at the end of the native refinement dataset and I imagine refers to 
a SeMet crystal. (Although the values are so small I think they are perhaps the 
estimated intensities of just the anomalous-substructure? If you are 
interested, below is a bit from that file starting with the end of the native 
data - I imagine the negative sigmas indicate absent reflections).

You can check the mmcif dictionary at 
http://mmcif.rcsb.org/dictionaries/mmcif_pdbx.dic/Categories/refln.html

to see whether you can find an appropriate category for your data. I hope the 
PDB will be happy to spend the time required to negotiate with you an 
appropriate representation (although I was told that time constraints often 
mitigate against this).

all the best
   Martyn 


<<snip>>
1 1 1   47    1   25 o    306.1  133.1 1034.6586 1520.8234 
1 1 1   47    1   28 o    348.2  137.8 1795.0702 1421.0972 
#END
data_r3ne5Asf
# 
_cell.entry_id      3ne5 
_cell.length_a      160.076 
_cell.length_b      160.076 
_cell.length_c      680.069 
_cell.angle_alpha   90.000 
_cell.angle_beta    90.000 
_cell.angle_gamma   120.000 
# 
_diffrn.id                  1 
_diffrn.crystal_id          1 
_diffrn.ambient_temp        ? 
_diffrn.crystal_treatment   ? 
_diffrn.details             '  SeMet phasing set' 
# 
_diffrn_radiation_wavelength.id           1 
_diffrn_radiation_wavelength.wavelength   0.97920 
# 
_entry.id   3ne5 
# 
_exptl_crystal.id   1 
# 
_reflns_scale.group_code   1 
# 
_symmetry.entry_id               3ne5 
_symmetry.space_group_name_H-M   'H 3 2' 
# 
#
loop_
_refln.crystal_id
_refln.wavelength_id
_refln.scale_group_code
_refln.index_h
_refln.index_k
_refln.index_l
_refln.status
_refln.pdbx_I_plus
_refln.pdbx_I_plus_sigma
_refln.pdbx_I_minus
_refln.pdbx_I_minus_sigma
1 1 1    1    0  -29  o       13.6      1.1        0.0     -1.0 
1 1 1    1    0  -26  o        4.3      0.4        0.0     -1.0 
1 1 1    1    0  -23  o       13.1      0.8        0.0     -1.0 
1 1 1    1    0  -20  o      141.7      5.0        0.0     -1.0 
1 1 1    1    0  -17  o      305.0     10.6        0.0     -1.0 
1 1 1    2   -1   15  o      189.0      6.2      178.5      6.8
<<snip>>


________________________________
 From: Raji Edayathumangalam <[email protected]>
To: [email protected] 
Sent: Friday, 6 September 2013, 3:42
Subject: [ccp4bb] What kind of reflection data to deposit to PDB
 


Hi Folks,

Sorry for the non-ccp4 post.

I am trying to determine what is the best form of unmerged reflection data to 
deposit to the PDB. I have single wavelength anomalous data for my structure 
and I have two flavors of scaled files from the same exact set of diffraction 
images: (1) data indexed and scaled in p1, and (2) data indexed in p222, scaled 
in Scalepack using the "no merge original index" option and converted to .mtz 
since the unit cell in the header of the output .sca file was missing. 

The space group for the dataset is p212121. 

Please could you let me know what might be the best approach.

Many thanks and cheers,
Raji

-- 
Raji Edayathumangalam
Instructor in Neurology, Harvard Medical School
Research Associate, Brigham and Women's Hospital
Visiting Research Scholar, Brandeis University

Reply via email to