3DNA Forum

Questions and answers => General discussions (Q&As) => Topic started by: kailsen on October 08, 2009, 12:29:46 am

Netiquette · Download · News · Gallery · G-quadruplexes · DSSR-Jmol · DSSR-PyMOL · Video Overview · DSSR v2.5.2 (DSSR Manual) · Homepage

Title: 3DNA calculation for Bases that have two conformers in pdb
Post by: kailsen on October 08, 2009, 12:29:46 am

Dear Sir,
 I would like to know which of the two bases does 3DNA consider while calculating the parameters in cases where the pdb has two conformers for a single base
e.g from 437D.pdb
ATOM     47  O4'A  G A2648     -26.371  52.168 -12.884  0.58 13.58           O  
ATOM     48  O4'B  G A2648     -25.879  52.221 -11.854  0.42 16.38           O
ATOM     49  C3'A  G A2648     -24.982  50.605 -13.934  0.58 11.81           C  
ATOM     50  C3'B  G A2648     -24.120  50.873 -12.569  0.42 14.39           C

I some cases one of the bases is completely flipped out, while in some the bases overlap each other except for the few atoms. I note this happening mainly in RNA GA base pairs

Title: Re: 3DNA calculation for Bases that have two conformers in pdb
Post by: xiangjun on October 08, 2009, 06:58:09 pm

This is a subtle point with regard to the ATOM or HETATM records of PDB format (http://http://xiang-jun.blogspot.com/2009/05/pdb-atom-coordinates-record.html), and I am glad that you raised the issue. 3DNA has been designed with this in mind and works as it should.

In 3DNA v2.0, file misc_3dna.par contains the following section:

Code: PHP

1. <span class="syntaxdefault"></span><span class="syntaxcomment"># Section 2&#58; alternative location in ATOM or HETATM records
2. #   default to A or 1 (A1), and ' ' is always added
3. </span><span class="syntaxkeyword"><</span><span class="syntaxdefault">alt_list</span><span class="syntaxkeyword">></span><span class="syntaxdefault">A1</span><span class="syntaxkeyword"></</span><span class="syntaxdefault">alt_list</span><span class="syntaxkeyword">></span><span class="syntaxdefault"> </span>

Thus, by default, only ATOM or HETATM records with alternative locations (column #17) of ' ', 'A' or '1' are processed. Of course, as pointed out in 3DNA FAQ (http://http://3dna.rutgers.edu/x3dna/faqs), you could modify file misc_3dna.par to use other alternative locations: 'B', for example.

HTH,

Xiang-Jun

PS:

• Every 3DNA user should upgrade to v2.0 (http://http://3dna.rutgers.edu:8080/forum/viewtopic.php?f=3&t=122): you have nothing to lose (no backward compatibility issues) but to take advantage of a more refined piece of software.
  
• Are you sure 437D.pdb has the ATOM records you listed?
  
  Code: [Select]
  e.g from 437D.pdb
ATOM 47 O4'A G A2648 -26.371 52.168 -12.884 0.58 13.58 O
ATOM 48 O4'B G A2648 -25.879 52.221 -11.854 0.42 16.38 O
ATOM 49 C3'A G A2648 -24.982 50.605 -13.934 0.58 11.81 C
ATOM 50 C3'B G A2648 -24.120 50.873 -12.569 0.42 14.39 CI cannot find them in 437D.pdb. It is not an important point here, but it would be helpful if you can clarify the discrepancy, just for the record.

Title: Re: 3DNA calculation for Bases that have two conformers in pdb
Post by: kailsen on October 10, 2009, 06:15:57 am

Hi Lu,

 Thanks you for clarifying my doubt. Sorry for the discrepancy in the PDB ID. Pls  see 483d.pdb  instead. I had  removed the ANISOU lines for clarity.

ATOM     27  C4 A  U A2647     -28.663  57.538  -5.049  0.58 17.81           C  
ATOM     28  C4 B  U A2647     -29.492  57.317  -4.727  0.42 18.11           C  
ATOM     29  O4 A  U A2647     -29.141  58.325  -4.234  0.58 16.26           O  
ATOM     30  O4 B  U A2647     -30.281  57.919  -4.000  0.42 18.56           O

Funded by the NIH R24GM153869 grant on X3DNA-DSSR, an NIGMS National Resource for Structural Bioinformatics of Nucleic Acids

Created and maintained by Dr. Xiang-Jun Lu, Department of Biological Sciences, Columbia University