search

AlphaBetaTest / cing

Automatically exported from code.google.com/p/cing
0 stars 0 forks source link

Duplication of resonances when importing to CING leads to lowerbound violations #293

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago 
What steps will reproduce the problem?
1. For entry 1a24

Original data:
Both 26 and 28 are Phe residues
assign (resid  28 and name He*)   (resid  26 and name He*)      4.0 0.7 4.8     

from DOCR:
http://nmr.cmbi.ru.nl/NRG-CING/recoordSync/1a24/1a24.tgz
to:
http://nmr.cmbi.ru.nl/NRG-CING/input/a2/1a24.tgz

In DOCR it is nicely one restraint but after FC's stereoswapping it becomes an 
ambiguous restraint with the four atom pairs listed separately (see attached 
figure). 

This leads to a larger violation of 1.8 Ang in CING at:

http://nmr.cmbi.ru.nl/NRG-CING/data/a2/1a24/1a24.cing/1a24/HTML/Restraints/dista
nce_constraint_list.html

than the 1.4 Ang. in NRG that is more correct.
        779 1  26 PHE QE  1  28 PHE QE  4.000 3.300 8.800 2.471 1.884 3.391 1.416  2 15 "[*+*****- **** **   *]" 1 

I need to perfect the importer and/or add a collapser in order to fix this. A 
lot of work. It might be required for other cases.

Original issue reported on code.google.com by jurge...@gmail.com on 28 Mar 2011 at 2:47

Attachments:

GoogleCodeExporter commented 9 years ago 
Wim, is it easy for you to collapse these ambis to non-ambis?

Original comment by jurge...@gmail.com on 28 Mar 2011 at 2:49

GoogleCodeExporter commented 9 years ago 
Not really - it relates to the way these are treated as resonances in CCPN. I 
don't fully understand why the violation is different though... isn't is the 
same thing even with the four items and the atoms individually listed?

Original comment by wfvran...@gmail.com on 29 Mar 2011 at 1:16

GoogleCodeExporter commented 9 years ago 
The violation is higher because the sum averaged distance is lower. Because 
there are duplicate pairs of the atoms.

Anyway, it might be correct CCPN I don't know  I 'll need to look into this one.

Original comment by jurge...@gmail.com on 30 Mar 2011 at 7:18

GoogleCodeExporter commented 9 years ago 
Right apologies Jurgen I'll look into why this is happening - it is clearly 
wrong (I didn't check the atom names thoroughly enough).

Original comment by wfvran...@gmail.com on 30 Mar 2011 at 7:37

GoogleCodeExporter commented 9 years ago

Original comment by jurge...@gmail.com on 30 Mar 2011 at 9:09

GoogleCodeExporter commented 9 years ago 
OK so I get the same 1.416 angstrom violation out of the current StereoSwap, 
which makes sense, but the resonances are indeed messed up afterwards - I was 
making a new connection to an HE2 or HE1 atom (but it already existed). The 
reason for this was that aromatics are treated slightly differently from 
prochirals in CCPN.

I think it's fixed now - please try again after updating 
ccpnmr/format/process/stereoAssignmentSwap.py

Original comment by wfvran...@gmail.com on 30 Mar 2011 at 9:34

GoogleCodeExporter commented 9 years ago 
Nice fix see attached, thanks Wim!

Original comment by jurge...@gmail.com on 30 Mar 2011 at 10:05

Attachments:

GoogleCodeExporter commented 9 years ago 
I need to reopen this issue as to remember to update/check some of my code 
tomorrow.
I have add some more smarts to the DR simplify routines.

Original comment by jurge...@gmail.com on 30 Mar 2011 at 6:14

GoogleCodeExporter commented 9 years ago 
Committed simplify code to r959.

Original comment by jurge...@gmail.com on 31 Mar 2011 at 12:48