Netiquette · Download · News · Gallery · Homepage · DSSR Manual · G-quadruplexes · DSSR-Jmol · DSSR-PyMOL · DSSR Licensing · Video Overview· RNA Covers

Author Topic: NMR analysis  (Read 22285 times)

Offline ghzheng

  • with-posts
  • *
  • Posts: 23
    • View Profile
NMR analysis
« on: December 23, 2008, 07:43:19 am »
Hi Xiangjun,

Your suggestion of using find_pair -s / -p is very very useful, especially for RNA structures.

After running:
1) find_pair -s 1KX5.pdb 1KX5.inps     --- I will get chain, residue information of each base in order
2) analyze 1KX5.inps     ---I will be able to obtain information of base step parameters

Now, I assume that bases are in the same order in both files of 1KX5.inps and bp_step.par. What I mean here is, the first base in the bp_step.par is the first base listed in the 1KX5.inps, second to second, and so on. By presenting the base step parameters I would need to tell users the associating chain and residue information of each base.

I also have questions for analyzing nmr structures.

1) find_pair/analyze and nmr_strs  both give an output of inp, auxiliary.par, bp_step.par .... , which contain information of only one structure. Now, is this structure the first model of the NMR pdb, or an average structure?

2) Based on your experience, normally how many (range, maximum) models are contained in one deposited nmr pdb file? I found that not many nmr structures provide the whole ensemble on the NDB. Your answer will help me determine the capability of providing outputs regarding all nmr models.

Thanks.

Guohui

Offline xiangjun

  • Administrator
  • with-posts
  • *****
  • Posts: 1650
    • View Profile
    • 3DNA homepage
Re: NMR analysis
« Reply #1 on: December 23, 2008, 10:41:56 pm »
Code: Text
  1. Now, I assume that bases are in the same order in both files of 1KX5.inps and bp_step.par. What I mean here is, the first base in the bp_step.par is the first base listed in the 1KX5.inps, second to second, and so on. By presenting the base step parameters I would need to tell users the associating chain and residue information of each base.
With "find_pair -s" option, and "analyze", you will get an output file with ".outs" extension It contains all information necessary to local each residue unambiguously -- see the section "RMSD of the bases". The newly added Perl script 'expand_ids' in v2.0 does this.

Regarding analysis of NMR structures, I do not think there is a fixed number of models in the ensemble]representative[/i] NMR structure. Use "ex_str -h" to extract it.

Xiang-Jun

 

Funded by X3DNA-DSSR, an NIGMS National Resource for Structural Bioinformatics of Nucleic Acids (R24GM153869)

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