New PDB tags and PDBx information in coordinate/restraints

[GoogleCodeExporter]

This issue is to check whether the new NMR-STAR files contain all
information necessary for public release as remediated restraint files at
the PDB. Note that the coordinates are still in here, they will be removed.

For testing purposes are attached a list of PDB ID codes with PDB chain
codes different from the PDBx codes (chainDiff.txt), and a list of PDB ID
codes where insertion codes are used (insertCode.txt), both generated by
Monica.

I generated NMR-STAR files with PDB_ tags for 2otk (chain codes different)
and 1tut (insertion code present), also both attached (gzipped).

To reproduce the PDBx information then, use the 'Asym_ID' chain identifier
in the '_Entity_assembly' loop in the 'assembly' saveframe. For the residue
number, just use the corresponding 'Label_comp_index_ID' tag in the
coordinate/restraint saveframes.

Original issue reported on code.google.com by wfvran...@gmail.com on 16 Nov 2009 at 3:58

Attachments:

• 1tut_linked.str.gz
• 2otk_linked.str.gz
• chainDiff.txt
• insertCode.txt

[GoogleCodeExporter]

The files appear to meet the criteria requested. Of course, I now have a bit of 
a
case of vertigo having tried to visually match values to the extensive list of 
tags
in each category.

Original comment by Eldon.Ul...@gmail.com on 28 Jan 2010 at 9:13

[GoogleCodeExporter]

What's needed to close this issue? 
It seems that the content is ok.
Are all the files there now?
Can PDB take a first look at them now?

Original comment by jurge...@gmail.com on 29 Jan 2010 at 9:04

[GoogleCodeExporter]

I guess PDB should have a look at them. What do you think John and Monica?

Original comment by schulte....@gmail.com on 29 Jan 2010 at 2:07

[GoogleCodeExporter]

Has a new full set of the restraints files been generated? Where are the files 
for
PDB (lacking coordinates) located?

Original comment by webmas...@bmrb.wisc.edu on 29 Jan 2010 at 11:35

[GoogleCodeExporter]

I sent files directly to John Westbrook and Monica but haven't heard anything 
back from them. I've been testing 
the software out on tang and haven't had any problems so far. I'll look for 
possible problem entties in the 
presetDict and test them on tang. Otherwise, I think we can update grunt, once 
we get the go-ahead from John, 
and close this issue.

Original comment by schulte....@gmail.com on 3 Feb 2010 at 4:50

[GoogleCodeExporter]

Hey guys.

I updated Wattos, ccpn, and recoord, then ran makePython.py. I'm not getting 
all the tags filled for the 
_Atom_site loop. I'm attaching files from tang (correct) and grunt. Any ideas?

Original comment by schulte....@gmail.com on 9 Feb 2010 at 5:56

Attachments:

• 1tut_linked_tang.str.gz
• 1tut_linked_grunt.str.gz

[GoogleCodeExporter]

Ok, let's break this down. Please adjust the below wattos script to find the 
input mmCIF file.

wattos < testReadMmCifWriteNmrStar.wcf

And compare your output with my output on my development machine Stella. If 
they are the same; please tell 
me which tags you're missing wrt to those I thought to have put in; see my 
comment 31.

I'm removing people from the cc here so we can work together for a while and 
still have it logged.

Original comment by jurge...@gmail.com on 10 Feb 2010 at 9:55

Attachments:

• testReadMmCifWriteNmrStar.wcf
• 1brv_from_mmcif.str.gz

[GoogleCodeExporter]

This did not create a str file. However, a joinedCoord.str was created 
yesterday. I've attached the output I just 
got.

Original comment by schulte....@gmail.com on 10 Feb 2010 at 2:49

Attachments:

• outputFromTestRead

[GoogleCodeExporter]

I ran this interactively. It needed a 'y' response. The files are the same, 
except for a thiol_state tag.

Original comment by schulte....@gmail.com on 10 Feb 2010 at 3:36

Attachments:

• 1brv_from_mmcif_grunt.str.gz

[GoogleCodeExporter]

I got in my local setup a $dir_star/1brv/1brv_merge.str
with coordinates but Chris and I found no coordinates in the one on grunt he 
was running. Chris will contact 
Wim on this issue by mail.

Original comment by jurge...@gmail.com on 10 Feb 2010 at 4:20

[GoogleCodeExporter]

It looks like everything is working. I will synchronize everything with the 
bmrb mirror, and it should all show up 
on the public sites tomorrow.

We can take a look at it then and go ahead and close this sucker.

Thanks guys.

Original comment by schulte....@gmail.com on 11 Feb 2010 at 6:23

[GoogleCodeExporter]

A couple of issues:

Reviewed 2kt7_linked.str

1. In the gen_dist constraints table the atoms listed for the PDB tags _1 and 
_2 and
those listed for Auth _1 and _2 do not match with those listed for 
Comp_index_ID_1
and _2.

2. The PDB and Auth atom nomenclature hierarchy is not present in the torsion 
angle
restraints table.

Original comment by webmas...@bmrb.wisc.edu on 11 Feb 2010 at 9:40

[GoogleCodeExporter]

Indeed, well spotted. 

1. This is because of complex mapping issues between the constraint item, the
resonance, the atoms, and the original names. It was working fine for the 'old'
distance constraint saveframe, but not for the 'new' one. Now fixed.

2. This is due to only the dihedral angle and residue given in the original data
(CYANA). I've fixed this so that the info is propagated in the final NMR-STAR 
file.

So should all work after an SF update of:

python/msd/nmrStar/IO/NmrStarExport.py

Anyone know an ID code for a file with RDCs? Should check those as well!

Original comment by wfvran...@gmail.com on 12 Feb 2010 at 12:29

[GoogleCodeExporter]

CYANA RDC entry is 2hgh

from oceans12_README.txt:
Why these (originally) 12 test entries were special to JFD, CP, WV

Current count: 14
Current list:  1a4d 1a24 1afp 1ai0 1brv 1bus 1cjg 1d3z 1hue 1ieh 1iv6 1kr8 2hgh 
2k0e

Molecular system:
**********************
1a4d small dsRNA one to start off with
1a24 well known 
1afp beta barrel.
1ai0 complex with H2O and
1brv small VAL H1 is only present in the second model and on.
1cjg nucleic acid/protein complex.
1d3x interest of Vincent B.
1hue well known dimer to me
1ieh well known
1iv6 complex
1otz large number of chains (60)
2k0e Largest in total (35 Mb cif file; 160 models with each 148 residues.
2jzn Entry with wildly different models in the ensemble.

On the restraint side:
**********************
1a24 Hydrogen bonds
1brv Discover
1bus no parsed
1cjg XPLOR
1iv6 EMBOSS
1kr8 Amber with RDCs CYANA with RDCs.

Original comment by jurge...@gmail.com on 12 Feb 2010 at 12:40

[GoogleCodeExporter]

Cut'n pasted one line too little:

2hgh CYANA with RDCs.

Original comment by jurge...@gmail.com on 12 Feb 2010 at 12:41

[GoogleCodeExporter]

The 2hgh joinedCoord file from 

/raid/docr/ccpn_tmp/data/archives/bmrb/nmrRestrGrid/2hgh

as well as the one I downloaded from the web site don't have _Atom_site.PDB_ 
tags
(still auth_ ones).

Anyway tried to fix this, but turns out there's no coupling constant info in the
original file either, as far as I can tell.

Original comment by wfvran...@gmail.com on 12 Feb 2010 at 1:06

[GoogleCodeExporter]

The original 2hgh has NMRPipe   "coupling constant" block but that's not 
parsed, sorry about that Wim.
I checked from the list 1d3z has X-PLOR RDCs. To find parsed RDCs from CYANA we 
need to add an entry like 
1jwe.

Chris, can you redo this set of 15 entries so we can test?
---
Why these (originally) 12 test entries were special to JFD, CP, WV

Current count: 15
Current list:  1a4d 1a24 1afp 1ai0 1brv 1bus 1cjg 1d3z 1hue 1ieh 1iv6 1jwe 1kr8 
2hgh 2k0e

Molecular system:
**********************
1a4d small dsRNA one to start off with
1a24 well known
1afp beta barrel.
1ai0 complex with H2O and
1brv small VAL H1 is only present in the second model and on.
1cjg nucleic acid/protein complex.
1d3z interest of Vincent B.
1hue well known dimer to me
1ieh well known
1iv6 complex
1otz large number of chains (60)
2k0e Largest in total (35 Mb cif file; 160 models with each 148 residues.
2jzn Entry with wildly different models in the ensemble.

On the restraint side:
**********************
1a24 Hydrogen bonds
1brv Discover
1bus no parsed
1cjg XPLOR
1iv6 EMBOSS
1jwe CYANA RDCs (Interest of WV)
1kr8 Amber with RDCs
2hgh CYANA with AMBER coupling constants.
1d3z XPLOR RDCs

Original comment by jurge...@gmail.com on 12 Feb 2010 at 1:26

[GoogleCodeExporter]

yes. I'll let you know when they're done.

Original comment by schulte....@gmail.com on 12 Feb 2010 at 2:55

[GoogleCodeExporter]

These have been processed and are on grunt at:
/grunt/docr/ccpn_tmp/data/recoord/

If you would like, I can tar, gzip, and send them to you.

Original comment by schulte....@gmail.com on 12 Feb 2010 at 3:39

[GoogleCodeExporter]

I don't see a coupling constant saveframe in 2hgh_linked.str. I did a complete 
weekly, so the joinedCoord.str has 
been rebuilt. I'm going to try running it on tang and see what I get.

Original comment by schulte....@gmail.com on 12 Feb 2010 at 3:52

[GoogleCodeExporter]

Same thing on tang.

Original comment by schulte....@gmail.com on 12 Feb 2010 at 4:21

[GoogleCodeExporter]

Look in 1d3z as mentioned in comment 67. It's still early in Madison I know. 
It's waaaaay beyond 5 here.

Original comment by jurge...@gmail.com on 12 Feb 2010 at 4:33

[GoogleCodeExporter]

Sorry, my bad for not catching the NMRPipe bit (before the coffee, and I really 
should see an eye doc). 1d3z 
looks good. 

It even looks like the author tags are in there? Has this been updated already?

I understand that it's way past 5 there - and a friday. I really appreciate 
what you guys have done.

Original comment by schulte....@gmail.com on 12 Feb 2010 at 4:46

[GoogleCodeExporter]

Let's hope we're almost there! I'll try to check 1d3z as well.

Original comment by wfvran...@gmail.com on 12 Feb 2010 at 5:02

[GoogleCodeExporter]

I've tested the new formats with entry 1brv and it looks fine. Great, now I can 
update NRG-CING again.

Can we close this thread?

Original comment by jurge...@gmail.com on 22 Feb 2010 at 9:50

[GoogleCodeExporter]

Everything I've looked at seems good. I finished the conversions over the 
weekend, but found that the structure 
for wwPDB_divided had been changed. Dimitri and I have to work out what's going 
on, then we will, hopefully, 
have all this on the public servers by tomorrow.

Otherwise, this thread can be closed. Would you like the honors, Jurgen?

Original comment by schulte....@gmail.com on 22 Feb 2010 at 2:49

[GoogleCodeExporter]

Original comment by schulte....@gmail.com on 22 Feb 2010 at 4:03

• Changed state: Fixed