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

Author Topic: using x3dna to analyze a coarse grained model mapped into an atomistic model  (Read 28382 times)

Offline amirhossein

  • with-posts
  • *
  • Posts: 4
    • View Profile
Hello,

I have a coarse grained DNA model(two bead model one for bases and the other for phosphates and sugar or backbone) and using NAB (from Amber) I have mapped it into an atomistic CG model. the primary structure is about 400 base pairs. when I feed it to X3DNA it just recognize 75 base pairs and not the rest.
I also receive the following   warnings after executing the code:
***Warning: structure with overlapped base-pairs***
This structure contains intra-chain direction reverse
(I have attached the pdb file of the model if you are interested to have a look).
 
Thanks in Forward for your help!



Offline xiangjun

  • Administrator
  • with-posts
  • *****
  • Posts: 1650
    • View Profile
    • 3DNA homepage
Hi,

Thanks for using 3DNA and posting your questions on the Forum. Attaching a PDB file is especially helpful in identifying where the problem is.

Quote
the primary structure is about 400 base pairs. when I feed it to X3DNA it just recognize 75 base pairs and not the rest.

This is normally due to fact that your presumed base-pairs are out of norm as would be expected by 3DNA's already generous criteria by default. For example if the shortest 'H-bond' distance between two bases is > 4.0 Å, 3DNA would not take them as forming a pair (in the default settings -- see file config/misc_3dna.par). An example case for G40-C761 is shown below. Note the long distances for base-base H-bonds.


Whenever 3DNA 'fails' to identify a pair, it is worth checking carefully for abnormality in the analyzed structure. In your case, you started with a CG model, and then built an atomic model using NAB. So I won't be surprised if there are irregularities in the resultant structure. Normally, by visualizing your structure in Jmol or PyMOL, you'll be able to notice clearly some issues.

Quote
***Warning: structure with overlapped base-pairs***

The message means that two base-pairs identified by 3DNA are overlapped in 3D. I am aware of one such erroneous entry in the PDB. In your case, however, it may well be due to the 'coarse' nature of your model.

Quote
This structure contains intra-chain direction reverse

Again, it could be due to abnormality in your structure.

Could you run AMBER or another tool to 'regulate' your structure?

Xiang-Jun
« Last Edit: February 24, 2016, 10:53:19 am by xiangjun »

Offline amirhossein

  • with-posts
  • *
  • Posts: 4
    • View Profile
Dear Dr.Xiang Jun,

Thanks a lot for your reply.actually your point helped me to solve the problem.as the equilibrated base distance in the CG model (regarding the PMF analysis) is 7 A, I changed the default value from 4 to 7 and now all the bases are recognized.
I have some more questions:
1) is the periodic box information important for X3DNA in analysis of the structure.
2) regarding the data for groove analysis, it seems that the measured values are switched off for the minor and major groove (the values for the minor groove are bigger than the major groove).so I was wondering if you have any suggestions about this issue.

Minor Groove        Major Groove
                 P-P     Refined     P-P     Refined
   1 GC/GC       ---       ---       ---       ---
   2 CC/GG       ---       ---       ---       ---
   3 CC/GG      20.6       ---      11.8       ---
   4 CC/GG      16.1      15.9      13.1      11.5
   5 CC/GG      16.3      15.8      11.8       9.7
   6 CC/GG      15.3      14.9      12.2      10.4
   7 CC/GG      15.0      14.8      12.3      11.3
   8 CC/GG      16.9      16.6      11.6      10.7
   9 CC/GG      18.4      17.8      12.0      10.9
  10 CC/GG      18.1      17.6      14.0      12.9
  11 CC/GG      17.8      17.6      13.2      12.4
  12 CC/GG      17.4      17.3      12.8      11.1
  13 CC/GG      16.8      16.6      12.9      10.7


again, thanks a lot for your help I really appreciate it.

Offline xiangjun

  • Administrator
  • with-posts
  • *****
  • Posts: 1650
    • View Profile
    • 3DNA homepage
Quote
actually your point helped me to solve the problem.

Glad to hear that it helped. One step forward!

Quote
1) is the periodic box information important for X3DNA in analysis of the structure.

No. 3DNA does not care about any info other than ATOM/HETATOM records in the PDB or PDBx/mmCIF input file.

Quote
2) regarding the data for groove analysis, it seems that the measured values are switched off for the minor and major groove (the values for the minor groove are bigger than the major groove).so I was wondering if you have any suggestions about this issue.

This is possible. For A-DNA, for example, the minor-groove width is larger the major-groove width. It may also be related to the algorithm implemented in 3DNA, or peculiarity in your structure. See the thread NUPARM vs 3DNA for further details.

Xiang-Jun
« Last Edit: February 24, 2016, 11:05:26 am by xiangjun »

Offline amirhossein

  • with-posts
  • *
  • Posts: 4
    • View Profile
Hello Dr. Xiang-Jun,

In order to get the helical axis I have used the rebuild command and every thing seems ok for the axis but there is something unusual in the image that I can not understand.It seems that some of the bases are elongated to a far distance!
I have attached the image (I use jmol to view it).
I was wondering if you could have a look at it and help me out!

Thanks in Forward

Offline xiangjun

  • Administrator
  • with-posts
  • *****
  • Posts: 1650
    • View Profile
    • 3DNA homepage
Quote
I have attached the image (I use jmol to view it).

What format is the image in? I cannot open it (including with Jmol). Please attach a regular .png image and the corresponding .pdb file.

Xiang-Jun

Offline amirhossein

  • with-posts
  • *
  • Posts: 4
    • View Profile
Hi,

Sorry for the trouble.I have attached the .png, .alc and the .pdb file I do the analysis on.

Thanks!

Offline xiangjun

  • Administrator
  • with-posts
  • *****
  • Posts: 1650
    • View Profile
    • 3DNA homepage
Quote
I have attached the .png, .alc and the .pdb file I do the analysis on

It seems no .png file is attached this time. I saw one attached .pdb file which helped a bit to understand your initial question:

Quote
In order to get the helical axis I have used the rebuild command and every thing seems ok for the axis but there is something unusual in the image that I can not understand.It seems that some of the bases are elongated to a far distance!

I looked at your attached 'atomistic_join.pdb' using Jmol, and noticed that one end is as shown below:



As you can see, the structure is highly irregular, not even complete. Moreover, I am not sure of how you could get the helical axis from the rebuild command -- certainly not an option in 3DNA. Did you mean the 3DNA analyze command instead of rebuild?

Please be specific and provide a (minimal) reproducible example to make your point unambiguous.

Xiang-Jun

 

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