headers
Irina Nesmelova | 4 Jan 2008 19:19
Picon

structure of protein with ligand

Hello,
 
I want to calculate a structure of protein dimer, in which each of the monomers has ligand - non-peptidic small molecule. I have a pdb file for ligand. How should I the information about ligand into python script? If someone has an example of pyhton script for similar system, I would greatly appreciate it.
Thank you,
Irina Nesmelova.
_______________________________________________
Xplor-nih mailing list
Xplor-nih <at> nmr.cit.nih.gov
http://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
Permalink | Reply | 0 comments |
headers
Irina Nesmelova | 11 Jan 2008 20:53
Picon

Re: structure of protein with ligand

Charles,

Thank you very much for your reply.

I have built my small molecule with XPLOR, so now I have both psf and pdb 
files.

In general,  to perform a simulated annealing for the complex of protein and 
small molecule do I have to keep a small molecule rigid to get it in proper 
orientation and without distorsion of its structure?
Thank you,
Irina.

----- Original Message ----- 
From: <Charles <at> Schwieters.org>
To: "Irina Nesmelova" <nesme001 <at> umn.edu>
Cc: <Charles <at> Schwieters.org>
Sent: Friday, January 04, 2008 2:10 PM
Subject: Re: [Xplor-nih] structure of protein with ligand

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
>
> Hello Irina--
>
>> I want to calculate a structure of protein dimer, in which each of the
>> monomers has ligand - non-peptidic small molecule. I have a pdb file
>> for ligand. How should I the information about ligand into python
>> script?
>
> You will need PSF information for the ligand. You might read:
>
> http://nmr.cit.nih.gov/xplor-nih/faq.html#node0150.txt
>
> I intend to create a facility to deal with rigid small molecules (given
> a pdb), but haven't managed to do so yet.
>
> For a dimer you'll also want to use the NCS potential term. For that,
> add this bit to your Python script:
>
> #non-crystallographic symmetry - so that each monomer is
> # (more or less) identical
> xplor.command("""
> ncs restraints
> initialize
> group
>  equi (segid A)
>  equi (segid B)
>  weight = 1
> end
> ?
> end""")
> potList.append( XplorPot('NCS') )
> rampedParams.append( MultRamp(.01,10,"potList['NCS'].setScale(VALUE)") )
>
> hope this helps--
> Charles
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.6 (GNU/Linux)
> Comment: Processed by Mailcrypt 3.5.8+ <http://mailcrypt.sourceforge.net/>
>
> iD8DBQFHfpKiPK2zrJwS/lYRAlq1AKCJYw21d1dH4DfgNVb8U9wD09UhfQCdEFRY
> jkmoRGNYahOeqpz1ysQ40Hk=
> =GX+U
> -----END PGP SIGNATURE-----
>
Permalink | Reply | 1 comment |
headers
Charles | 11 Jan 2008 23:50

Re: structure of protein with ligand


Hello Irina--

> 
> I have built my small molecule with XPLOR, so now I have both psf and pdb 
> files.
> 
> In general,  to perform a simulated annealing for the complex of protein and 
> small molecule do I have to keep a small molecule rigid to get it in proper 
> orientation and without distorsion of its structure?

this is a bit of a judgement call on your part. I'd definitely keep
rigid those portions for which you know the geometry. If the molecule
contains rotatable bonds, but you have a good idea of its configuration,
fix the geometry. In any case, you might perform initial calculations
with a rigid conformation.

I hope this helps--
Charles
Permalink | Reply | 0 comments |
headers
svasudev | 15 Jan 2008 19:41
Picon
Favicon

J coupling restraints.

Hi

I have got J coupling information from my DQF-COSY experiments (for phi
and chi). Can i use them as it is in my refining script without converting
to dihedrals?

When I do it I get the following error messages in the out file.

refine.py(66): jCoup = create_JCoupPot("jcoup","jna.tbl",
refine.py(67):                         A=6.51,B=-1.76,C=1.6,phase=-60.0)
refine.py(68): jCoup = create_JCoupPot("jcoup2","jcacb.tbl",
refine.py(69):                         A=9.5,B=-1.6,C=1.8,phase=0.0)
selection D  must contain one atom.
        selection string:
selection D  must contain one atom.
        selection string:
JCoupPot::addRestraints: error adding restraint.

Thanks for your time!!!

Sheeja.
Permalink | Reply | 1 comment |
headers
Charles | 15 Jan 2008 20:30

Re: J coupling restraints.


Hello Sheeja--

> When I do it I get the following error messages in the out file.
> 
> 
> refine.py(66): jCoup = create_JCoupPot("jcoup","jna.tbl",
> refine.py(67):                         A=6.51,B=-1.76,C=1.6,phase=-60.0)
> refine.py(68): jCoup = create_JCoupPot("jcoup2","jcacb.tbl",
> refine.py(69):                         A=9.5,B=-1.6,C=1.8,phase=0.0)
> selection D  must contain one atom.
>         selection string:
> selection D  must contain one atom.
>         selection string:
> JCoupPot::addRestraints: error adding restraint.
> 

It seems that something is wrong with your restraint table
(jna.tbl). Xplor-NIH sees only three out of four atom selections for one
or more ASSIgn statements. Please check these.

best regards--
Charles
Permalink | Reply | 0 comments |
headers
R.M. van der Werf | 16 Jan 2008 14:39
Picon

Transitioning from Xplor v3.851 to pyXplor

Hi,

I am trying to make the transition from Xplor v3.851 to pyXplor (from
Xplor-NIH-2.18.0).

To understand pyXplor I am trying to make a script which performs a
number of dynamics steps. (No complicated bussiness with restraints
besides holonomics)

I get stuck quite in the beginning. When I read the standard nucleic.par
and nucleic.top, the program complains about duplicate definitions of
parameters.

I am used to having to define
1. topology
2. parameters
3. structure (this I managed to do with protocol.loadPDB('##') )
where do I have to define steps 1 and 2 ? And how do I use multiple
parameter files ?

Sorry for this quite silly question, but I could not find it in online
documentation, FAQs, and the tutorial.

Kind regards,
Ramon van der Werf

Here is my script which I made so far:
--------------------------------------

import protocol, ivm	
protocol.initRandomSeed(1)

#Define topology and subsequent parameters
protocol.initTopology('/home/ramon/COMPROG/xplor-nih-2.18/toppar/nucleic.top')
protocol.initParams('/home/ramon/COMPROG/xplor-nih-2.18/toppar/nucleic.par')

# Read PDB and generate PSF on the fly and initializes coordinates
protocol.loadPDB('par/ED_53_acc.pdb')
Permalink | Reply | 1 comment |
headers
Charles | 16 Jan 2008 18:23

Re: Transitioning from Xplor v3.851 to pyXplor


Hello Ramon--

> 
> I get stuck quite in the beginning. When I read the standard nucleic.par
> and nucleic.top, the program complains about duplicate definitions of
> parameters.
> 
> 
> import protocol, ivm	
> protocol.initRandomSeed(1)
> 
> #Define topology and subsequent parameters
> protocol.initTopology('/home/ramon/COMPROG/xplor-nih-2.18/toppar/nucleic.top')
> protocol.initParams('/home/ramon/COMPROG/xplor-nih-2.18/toppar/nucleic.par')
> 

If you're running /home/ramon/COMPROG/xplor-nih-2.18/bin/xplor
then these files are read in by default when it is determined (within
loadPDB) that your structure is a nucleic acid. i.e. you don't need
these lines. You can specify the string "nucleic" in both cases- it
specifies using default topology/parameter files (the ones you're
loading). You can change the default parameter/topology files using

import protocol
protocol.paramters["nucleic"] = "/path/to/other/params"
protocol.topology["nucleic"] = "/path/to/other/topo"

> # Read PDB and generate PSF on the fly and initializes coordinates
> protocol.loadPDB('par/ED_53_acc.pdb')
> 

When you use loadPDB, you don't need to specify initTopology/Params. The
caveat is that nonstandard residues can't be handled without some other
work.

And you're right- the documentation should be improved.

best regarads--
Charles
Permalink | Reply | 0 comments |
headers
Irina Nesmelova | 17 Jan 2008 23:00
Picon

carbon chemical shift error

Hello,
 
How can I input the error for carbon chemical shift?
Thank you,
Irina.
_______________________________________________
Xplor-nih mailing list
Xplor-nih <at> nmr.cit.nih.gov
http://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
Permalink | Reply | 4 comments |
headers
John Kuszewski | 17 Jan 2008 23:37
Picon

Re: carbon chemical shift error

Hi,

I presume you're asking about the CARBonshift energy term.

Do you mean you want to define an error in your observed Ca or Cb chemical shifts?  If so, you can't.  The term's accuracy in predicting carbon shifts from backbone torsion angles is rather low (~ 1ppm, if I remember correctly) compared to the accuracy of measurement of Ca and Cb chemical shifts, so there's no point in defining a squarewell that way.  

Rather, the carbonshift's squarewell potential works by comparing the difference between observed and expected shift values relative to the known accuracy of the prediction surface at each point.  These accuracies are recorded in the same table as the expected secondary chemical shifts.

So all you need to do is turn on the square well potential shape by calling CARBon SQUArewell, and you're all set.  

If I've misunderstood your question, forgive me.

--JK

On Jan 17, 2008, at 5:00 PM, Irina Nesmelova wrote:

Hello,
 
How can I input the error for carbon chemical shift?
Thank you,
Irina.
_______________________________________________
Xplor-nih mailing list

_______________________________________________
Xplor-nih mailing list
Xplor-nih <at> nmr.cit.nih.gov
http://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
Permalink | Reply | 3 comments |
headers
Irina Nesmelova | 18 Jan 2008 02:54
Picon

Re: carbon chemical shift error

Hi John,
 
Thank you for your reply. That is exactly what I have asked. And the reason for this question is that I don't see the difference in CARB energy term if I change the potential from harmonic to square. The energy is very high (~9000-10000). I am using the script from mef-dna example (sa_mef_dipo.inp) Is there some other parameters I have to change?
Thank you,
Irina.
----- Original Message -----
Sent: Thursday, January 17, 2008 4:37 PM
Subject: Re: [Xplor-nih] carbon chemical shift error

Hi,

I presume you're asking about the CARBonshift energy term.

Do you mean you want to define an error in your observed Ca or Cb chemical shifts?  If so, you can't.  The term's accuracy in predicting carbon shifts from backbone torsion angles is rather low (~ 1ppm, if I remember correctly) compared to the accuracy of measurement of Ca and Cb chemical shifts, so there's no point in defining a squarewell that way.  

Rather, the carbonshift's squarewell potential works by comparing the difference between observed and expected shift values relative to the known accuracy of the prediction surface at each point.  These accuracies are recorded in the same table as the expected secondary chemical shifts.

So all you need to do is turn on the square well potential shape by calling CARBon SQUArewell, and you're all set.  

If I've misunderstood your question, forgive me.

--JK

On Jan 17, 2008, at 5:00 PM, Irina Nesmelova wrote:

Hello,
 
How can I input the error for carbon chemical shift?
Thank you,
Irina.
_______________________________________________
Xplor-nih mailing list

_______________________________________________
Xplor-nih mailing list
Xplor-nih <at> nmr.cit.nih.gov
http://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
Permalink | Reply | 2 comments |

Gmane