Dear Charles,
Thank you Charles for all these prompt updates.
The Talos script handles the dimer very nicely, and it seems to be able to
output the SEP residue in the PDB files much better now. Thank you so much!
The code now doesn’t fully go to completion, especially during the step for:
> xplor makeNEF.py ${name}_new.nef *final.pasd $likelihoodCutoff
In the script.sh, but I am playing around with the script to explore further.
I think this could be because of a choice of syntax that I have been using in
the .nef input.
The current xplor-NIH seems to like this following input the most, so I have
been using it for all my .nef scripts:
The script I use uses SEP in the molecular name for the .psf generation, but
SER in the chemical list. This minimises error in 3.9, 3.10, but might causes
issues in the long run:
in the
save_nef_molecular_system:
…
41 B 45 ARG middle . .
42 B 46 SEP middle . .
43 B 47 PRO middle . .
…
save_nef_chemical_shift_list_1_peptide:
…
B 46 SER CB 64.04604116 . C 13
B 46 SER H 8.873905008 0.00213681375 H 1
B 46 SER HA 4.189790184 0.01496001044 H 1
B 46 SER HB2 3.789559292 0.00655430596 H 1
B 46 SER HB3 3.860571509 0.006689027185 H 1
…
But changing the chemical shift naming to SER works well with the TalosN
script, which will not complain about missing atoms, and allow their NOEs to be
read by the Talos, and this worked well for 3.9, 3.9.10.
However, now, in 3.9.11, it seems that the next part of the code after pass3, a
code breaking error occurs
> xplor makeNEF.py ${name}_new.nef *final.pasd $likelihoodCutoff
Will dislike this ‘trick’ that I have been using up to this point:
reading shift assignments from NOE_2`1`_final.pasd ... X-PLOR>set message off
echo off end
[502 loaded]
reading peaks from NOE_2`1`_final.pasd ... TCLInterp::command: error executing:
readMarvinPeaks -fileName "/dev/shm/xplor-nih-jiang02-31556/tmppggx8job.peaks."
-pot noe_NOE_2`1`
Result or error message is: xplor-nih error:
ShiftAssignment::primarySeqDistToSA: shift assignment
NOE_2`1`219_fromB_46_SER_H selects zero protons with selection: ((((segid B and
resid 46 and resname SER and name H))))
Traceback (most recent call last):
File "<string>", line 2, in <module>
File "/Users/jiang02/xplor-nih-3.9.11/python/trace.py", line 180, in run
exec(cmd, dict, dict)
File "<string>", line 1, in <module>
File "/Users/jiang02/xplor-nih-3.9.11/python/xplorInit.py", line 147, in
execfile
exec(code, globals, locals)
File "makeNEF.py", line 114, in <module>
noePot = create_PASDPot(name,
^^^^^^^^^^^^^^^^^^^^
File "/Users/jiang02/xplor-nih-3.9.11/python/pasdPotTools.py", line 118, in
create_PASDPot
tcl.command('readMarvinPeaks -fileName "%s" -pot noe_%s' %
File "/Users/jiang02/xplor-nih-3.9.11/python/wrappers/tclInterp.py", line 72,
in command
return _tclInterp.TCLInterp_command(self, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Alternatively:
If I changed the SER to SEP in 46 instead, a different problem will emerge
earlier in the run
Input in the nef that I changed:
save_nef_chemical_shift_list_1_peptide:
…
B 46 SEP CB 64.04604116 . C 13
B 46 SEP H 8.873905008 0.00213681375 H 1
B 46 SEP HA 4.189790184 0.01496001044 H 1
B 46 SEP HB2 3.789559292 0.00655430596 H 1
B 46 SEP HB3 3.860571509 0.006689027185 H 1
…
Output warning:
nefShifts: reading table: default
nefShifts: ignoring shift entry with atomname @_2257
401 shifts were read from NEF data.
Warning: selection (segid B and resid 42 and resname GLY and name H) does not
match any atoms
Warning: selection (segid B and resid 46 and resname SEP and name CA) does not
match any atoms
Warning: selection (segid B and resid 46 and resname SEP and name CB) does not
match any atoms
Warning: selection (segid B and resid 46 and resname SEP and name H) does not
match any atoms
Warning: selection (segid B and resid 46 and resname SEP and name HA) does not
match any atoms
Warning: selection (segid B and resid 46 and resname SEP and name HB2) does not
match any atoms
Warning: selection (segid B and resid 46 and resname SEP and name HB3) does not
match any atoms
And more downstream:
In the following line:
> for spectrum in $spectra; do xplor initMatch.py ${name}_new.nef $spectrum;
> done
This error will appear if I used B 46 SEP instead of B 46 SER in the
chemical shift list.
initMatch.py(92): print(unassigned.message)
Of 755 total atoms, 538 atoms were expected to be assigned.
This does not include atoms:
…
B 45 ARG CA
B 45 ARG CD
B 46 SER N
B 46 SER H
B 46 SER P
B 46 SER CA
B 46 SER HA
B 46 SER HB3
B 46 SER HB2
B 46 SER CB
B 47 PRO CG
…
These two warnings go away if I used B 46 SER only in the chemical list, but
still retain B. 46 SEP for the rest of the Nef file.
In short, the script in 3.9, 3.10 seems to give the least initial errors if I
substituted SEP for SER in the chemical shifts list, but retain SEP in the
actual molecular names list for .psf generation and the NOEs.
This is not ideal, and I hope to raise this issue to ask if this causes
problems in the script. I appreciate all your prompt replies and help, I
apologise in advance for the troubles caused.
Thank you.
Best wishes,
Mingxuan Jiang
Postgraduate
Creixell Lab
CRUK Cambridge Institute
University of Cambridge
[A black background with a black square Description automatically generated
with medium confidence]
From: Charles Schwieters <[email protected]>
Date: Wednesday, 5 March 2025 at 19:21
To: Mingxuan Jiang <[email protected]>
Cc: Xplor-NIH mailing list <[email protected]>, Mingxuan Jiang
<[email protected]>
Subject: Re: readNEF doubles protein sizes of hetero-dimers
Hello Mingxuan--
>
> When I entered the directory installed from the Debian format, I compared the
> input scripts to be used for structure solving from nef:
>
Right- I just made the changes required for handling multiple chains
in runTalos.sh yesterday. I prepared a new snapshot for you today:
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fbit.niddk.nih.gov%2Fout%2FtoMingxuan-20250305%2F&data=05%7C02%7Cmj582%40universityofcambridgecloud.onmicrosoft.com%7C38d2e9302b4d4ecebcbd08dd5c1ada56%7C49a50445bdfa4b79ade3547b4f3986e9%7C1%7C0%7C638767992590370530%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=MQ4b1yvKA6Y%2BsIJlgc9aNuezjkJ2fATv85eryrApkRE%3D&reserved=0<https://bit.niddk.nih.gov/out/toMingxuan-20250305/>
Note that, as this point you will need to explicitly specify the
additional segment name when you execute runTalos.sh:
sh runTalos.sh input.nef "A B"
>
> The protein is now of correct size, but the phosphate group is still deleted
> at
> the final steps.
>
>
In today's snapshot I have made some other changes so that the variant
residue SEP is better handled, and I'd appreciate it is you ran
through the process one more time to see if things work better
now. I'd like to iterate on this until it works ok.
>
>
> Will the energetics of the choice of ensemble structures be affected by this
> missing phosphate? Is this just a formatting issue in the .pdb files but not
> the .cifs?
>
It would be best if the correct residue were simply present.
best regards--
Charles
########################################################################
To unsubscribe from the XPLOR-NIH list, click the following link:
http://list.nih.gov/cgi-bin/wa.exe?SUBED1=XPLOR-NIH&A=1
