user queries for the Xplor-NIH biomolecular structure determination program

headers Debora Marks | 7 Jan 2011 14:09 · /Xv1PdCc9L7z6/

_______________________________________________
Xplor-nih mailing list
Xplor-nih <at> nmr.cit.nih.gov
http://dcb.cit.nih.gov/mailman/listinfo/xplor-nih

Hello Debora--

> >
> > >  I am using distance restraints with both upper and lower bounds,
> > > starting from extended conformations of proteins ( 100-300 aas)
> > > and want to know 
> > >
> > > 1. Can we also WEIGHT the distance restraints INDIVIDUALLY  as
> > > well as set the bounds ? ( or can we 'only' do this in the anneal
> > > program) 
> > >
> >
> > This is done on the level of the script. You may have different NOE
> > objects (Python interface) or classes (old XPLOR interface) which have
> > different force constants. A force constant is not specified in the
> > restraint file.
> >
> 
>  I wonder if you could explain to me what you mean by "classes" ?
> in your reply to my question.  I * think* you mean that I cannot enter force
> constants for individual restraints in the 'anneal.inp' files but can
> somehow use different NOE objects to achieve this
> 

An example of using different force constants for different subsets of
NOEs using the old XPLOR interface can be found in this script:
eginput/rna_refi/rna_database_refine.inp
  (search for noe - there are two classes in this case)

For an example in the Python interface, you can start looking at line
326 in eginput/pre/newRefine.py. There are two sets of NOES there, and
they both are using a force constant (scale) of 1, but, those values can
be easily changed. The introduction of multiple NOE force constants
can be added to eginput/gb1_rdc/anneal.py in the same way by making
modifications starting at line 141.

best regards--
Charles

Hello,

I am trying to make an ethanolamide C-terminal modification: -N-CH2-CH2-OH. Because of the atom naming, I
am making it as a residue, rather than a patch. I have added an appropriate segment to protein.top and one of
the missing impropers to protein.par.

The problem I ran into is that "psfGen.py" automatically patches the C-terminus with a "CTER". I can work
around this by setting cterm='' in psfGen and also modifying the xplor.command a few lines later, but I
would like some permanent solution. Something similar to the case of an acetyl group "if seq[0]=='ACE':
nterm='' ", but for the last residue would be ideal.

Thank you,
Vitaly

Hello Vitaly--

> I am trying to make an ethanolamide C-terminal modification:
> -N-CH2-CH2-OH. Because of the atom naming, I am making it as a
> residue, rather than a patch. I have added an appropriate segment to
> protein.top and one of the missing impropers to protein.par. 
> 
> The problem I ran into is that "psfGen.py" automatically patches the
> C-terminus with a "CTER". I can work around this by setting cterm=''
> in psfGen and also modifying the xplor.command a few lines later, but
> I would like some permanent solution. Something similar to the case of
> an acetyl group "if seq[0]=='ACE': nterm='' ", but for the last
> residue would be ideal. 
> 

I'm glad you found a workaround. Here
http://nmr.cit.nih.gov/xplor-nih/updates/2.27/
I've posted a modified psfGen.py with seqToPSF updated to take a new
argument, ctermPatch. If it is set to a blank string, no c-terminal
patching will occur. Does this suit your needs?

best regards--
Charles

Hello Charles,

Thank you for the fast response, the solution you propose suits me very well. There is still an issue with a
psfGen though: when seqToPSF is supplied with ctermPatch='', the error message results:

Traceback (most recent call last):
  File "stdin", line 1, in <module>
  File "/opt/xplor-nih-2.24/python/psfGen.py", line 553, in seqToPSF
    ''' % (nterm,cterm,splitSeq))
UnboundLocalError: local variable 'cterm' referenced before assignment

Thank you,
Vitaly

----- Original Message -----
From: Charles <at> Schwieters.org
Date: Tuesday, January 11, 2011 10:03 am
Subject: Re: [Xplor-nih] Custom C-terminal residue
To: Vitaly Vostrikov <vvostri <at> uark.edu>
Cc: xplor-nih <at> nmr.cit.nih.gov

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> 
> Hello Vitaly--
> 
> > I am trying to make an ethanolamide C-terminal modification:
> > -N-CH2-CH2-OH. Because of the atom naming, I am making it as a
> > residue, rather than a patch. I have added an appropriate segment to
> > protein.top and one of the missing impropers to protein.par. 
> > 
> > The problem I ran into is that "psfGen.py" automatically patches the
> > C-terminus with a "CTER". I can work around this by setting cterm=''
> > in psfGen and also modifying the xplor.command a few lines later, 
> but> I would like some permanent solution. Something similar to the 
> case of
> > an acetyl group "if seq[0]=='ACE': nterm='' ", but for the last
> > residue would be ideal. 
> > 
> 
> I'm glad you found a workaround. Here
> http://nmr.cit.nih.gov/xplor-nih/updates/2.27/
> I've posted a modified psfGen.py with seqToPSF updated to take a new
> argument, ctermPatch. If it is set to a blank string, no c-terminal
> patching will occur. Does this suit your needs?
> 
> best regards--
> Charles
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.10 (GNU/Linux)
> Comment: Processed by Mailcrypt 3.5.9 
> <http://mailcrypt.sourceforge.net/>
> iEYEARECAAYFAk0sf0QACgkQPK2zrJwS/lY6rwCeNzg/i31qJ7qRhzXC++85bbb6
> NbEAn2ganzV4yy4OMj/Xv1PdCc9L7z6/
> =uqUZ
> -----END PGP SIGNATURE-----
> 

Hello Vitaly--

> Thank you for the fast response, the solution you propose suits me
> very well. There is still an issue with a psfGen though: when seqToPSF
> is supplied with ctermPatch='', the error message results: 
> 
> Traceback (most recent call last):
>   File "stdin", line 1, in <module>
>   File "/opt/xplor-nih-2.24/python/psfGen.py", line 553, in seqToPSF
>     ''' % (nterm,cterm,splitSeq))
> UnboundLocalError: local variable 'cterm' referenced before assignment
> 

Sorry to post such an untested solution. The patch below should fix the
problem (I also updated the web site-posted version).

best regards--
Charles

*** python/psfGen.py	2011-01-11 15:54:24 +0000
--- python/psfGen.py	2011-01-11 17:56:37 +0000
***************
*** 510,515 ****
--- 510,517 ----
              pass
          if ctermPatch:
              cterm = "LAST  %s   HEAD - *                    END" % ctermPatch
+         else:
+             cterm=''
              pass

          if seq[0]=='ACE':

Charles,
The script works fine now, thank you very much!

Vitaly

----- Original Message -----
From: Charles <at> Schwieters.org
Date: Tuesday, January 11, 2011 12:00 pm
Subject: Re: [Xplor-nih] Custom C-terminal residue
To: Vitaly Vostrikov <vvostri <at> uark.edu>
Cc: Charles <at> Schwieters.org, xplor-nih <at> nmr.cit.nih.gov

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> 
> Hello Vitaly--
> 
> > Thank you for the fast response, the solution you propose suits me
> > very well. There is still an issue with a psfGen though: when 
> seqToPSF> is supplied with ctermPatch='', the error message 
> results: 
> > 
> > Traceback (most recent call last):
> >   File "stdin", line 1, in <module>
> >   File "/opt/xplor-nih-2.24/python/psfGen.py", line 553, in seqToPSF
> >     ''' % (nterm,cterm,splitSeq))
> > UnboundLocalError: local variable 'cterm' referenced before 
> assignment> 
> 
> Sorry to post such an untested solution. The patch below should fix 
> theproblem (I also updated the web site-posted version).
> 
> best regards--
> Charles
> 
> *** python/psfGen.py	2011-01-11 15:54:24 +0000
> - --- python/psfGen.py	2011-01-11 17:56:37 +0000
> ***************
> *** 510,515 ****
> - --- 510,517 ----
>              pass
>          if ctermPatch:
>              cterm = "LAST  %s   HEAD - *                    END" 
> % ctermPatch
> +         else:
> +             cterm=''
>              pass
>              
>          if seq[0]=='ACE':
> 
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.10 (GNU/Linux)
> Comment: Processed by Mailcrypt 3.5.9 
> <http://mailcrypt.sourceforge.net/>
> iEYEARECAAYFAk0smqoACgkQPK2zrJwS/lYWCgCdHRdOF8yxiTjUGWtn4l7HcBA5
> 8WgAni0sqE25jr9Ybxf7tbkMXbxKdzlj
> =4HXH
> -----END PGP SIGNATURE-----
> 

_______________________________________________
Xplor-nih mailing list
Xplor-nih <at> nmr.cit.nih.gov
http://dcb.cit.nih.gov/mailman/listinfo/xplor-nih

Hello David--

> How do I set the ceiling for NOE energies?
> i.e., what's the python equivalent to
> 
> noe  ceiling 100 end
> 
> (also, what is its default value?)
> 

No analogous value exists in the Python interface. If you think you have
some bad NOEs, you might wish to use the soft potential type. And if
there are known bad NOEs I suggest removing them from the table. I have
never found a need for introducing a maximum NOE energy value, but if
you can convince me otherwise, I might add it.

best regards--
Charles

Hi every one
I have launched calculations with RDCs (two media, peg and phage) on my
protein. Since my protein has 4 modules, with some of them being mobiles,
I have used up to 8 tensors. (assuming every module was behaving
independently), or only 2 tensors (assuming that the protein is completely
rigid). All calculations converge, with more or less violations, but the
the calculations made under the rigid assumption. (not surprising). But my
problem is the following:

Numbering of my protein begins at 355 and finish at 626.

I have hence introduced the tensors in the pdbfile as ANI 700, ANI 701,
..., 707. (if necessary)
and in the RDC restraints filesas well. But I have the feeling that I may
have skipped something somewhere, and that the software hence still makes
calculations with tensors numbered 500, 501...

In the out file, I will have the following messages (I only write the one
corresponding to the tensor 700, but I have also the 7 others,(in the case
of the calculation with 8 tensors) with different numberings of course..)
:

 calcTensorOrientation(media[medium])
AT_Build::buildNode: cycle link found between atoms 2141 VAL 500 C and
2143 ALA 501 N
        removing bond.
*-- POWELL ------ step=      0 --- stepsize=   0.01000 --- energy evals=  
 2 -*
|   E(poten)= 26362.5488668       grad=  3373.7631740   peg1_NH=
26362.5488668 |
*------------------------------------------------------------------------------*
*-- POWELL ------ step=     10 --- stepsize=   0.01000 --- energy evals=  
24 -*
|   E(poten)= 19189.7677656       grad=  2425.1536755   peg1_NH=
19189.7677656 |
*------------------------------------------------------------------------------*
Powell::step: normal exit.
*-- POWELL ------ step=     15 --- stepsize=   0.01000 --- energy evals=  
10 -*
|   E(poten)= 19189.7277415       grad=  2425.4538268   peg1_NH=
19189.7277415 |
*------------------------------------------------------------------------------*
refinement1234.py(165):     rampedParams.append(
StaticRamp("calcTensor(media['%s'])" % medium) )

And more bothering for me, if I have a RDC restraint concerning residues
500, 501, 502 ... , the energies becomes incredibely high at the beginning
of the calculation.
Finally, I observed that for the calculations under the rigid assumption
(only two tensors, ANI 700 and ANI 701), the high level of energies was
due to residue 500 and 501 which were completely exploded.

Hence, I really have the impression that somewhere, I did something wrong
in the calculation.

All the protocol has been derived from the protocole for refinement of
protein G (tutorial).

Any help appreciated!

-- 
Hélène Déméné, PhD
Centre de Biochimie Structurale
UMR 5548 CNRS-UM1-INSERM UMR 554
29, rue de Navacelles
34090 Montpellier Cedex
France
Tel: +33 (0) 4 67 41 77 01
Fax: +33 (0) 4 67 41 79 13
e-mail: Helene.Demene <at> cbs.cnrs.fr

--

-- 
Passerelle antivirus CBS

12	May 2013
21	April 2013
7	March 2013
4	February 2013
8	January 2013
1	November 2012
4	October 2012
2	September 2012
2	August 2012
5	July 2012
11	June 2012
4	April 2012
25	March 2012
6	February 2012
3	January 2012
3	December 2011
14	November 2011
7	October 2011
9	September 2011
4	August 2011
4	July 2011
7	June 2011
6	May 2011
5	April 2011
5	March 2011
6	February 2011
21	January 2011
16	December 2010
17	November 2010
27	October 2010
15	September 2010
5	August 2010
13	July 2010
36	June 2010
9	May 2010
9	April 2010
14	March 2010
27	February 2010
12	January 2010
17	November 2009
8	October 2009
17	September 2009
16	August 2009
33	July 2009
27	June 2009
17	May 2009
29	April 2009
11	March 2009
23	February 2009
17	January 2009
2	December 2008
25	November 2008
15	October 2008
26	September 2008
17	August 2008
21	July 2008
15	June 2008
26	May 2008
61	April 2008
19	March 2008
23	February 2008
16	January 2008
45	December 2007
20	November 2007
20	October 2007
11	September 2007
46	August 2007
29	July 2007
38	June 2007
29	May 2007
34	April 2007
33	March 2007
21	February 2007
7	January 2007
16	December 2006
22	November 2006
27	October 2006
10	September 2006
29	August 2006
16	July 2006
15	June 2006
20	May 2006
20	April 2006
13	March 2006
18	February 2006
7	January 2006
25	December 2005
11	November 2005
6	October 2005
2	September 2005
4	August 2005
10	July 2005
5	June 2005
6	May 2005
3	April 2005
4	March 2005
1	January 2005
4	November 2004
7	October 2004
8	September 2004
10	August 2004
6	July 2004
13	June 2004
5	May 2004
9	April 2004
10	March 2004
14	February 2004
10	January 2004
7	December 2003
15	November 2003
10	October 2003
6	September 2003
2	August 2003
12	July 2003
4	June 2003
8	May 2003
2	April 2003
6	March 2003
6	February 2003
8	January 2003
4	December 2002
5	October 2002
1	September 2002
7	August 2002
8	July 2002
6	June 2002

Re: Individual Distance restraint weightings in dist geometry and anneal ?

Re: Individual Distance restraint weightings in dist geometry and anneal ?

Custom C-terminal residue

Re: Custom C-terminal residue

Re: Custom C-terminal residue

Re: Custom C-terminal residue

Re: Custom C-terminal residue

noe ceiling

Re: noe ceiling

problems of numbering residue/tensors with RDC

Mon	Tue	Wed	Thu	Fri	Sat	Sun

					1	2
3	4	5	6	7	8	9
10	11	12	13	14	15	16
17	18	19	20	21	22	23
24	25	26	27	28	29	30
31