From DNA sequence to DNA structure

[cif077]

Dear All,
I'm newbie in this field, however, I did not find any topic that can answer my question.
I want to study whether a pair of genes that separated by a distance of DNA sequence would close together in 3D conformation.

So, how to use 3DNA to generate a 3D DNA double-helix structure like this: http://http://hydra.icgeb.trieste.it/~kristian/dna/images/modelit_long2.jpg from a DNA sequence?
 I've surveyed some tools like ADN-Viewer (not publicly available), model.it (can not accept long DNA).
And I've alos tried fiber, but all I got just a linear string of DNA structure like attached "Picture 1.png".

I know the "reconstruction function" in w3DNA provide a function: "Customized base-pair-step/nucleotide parameters".
But user seems have to provide all the parameters between every adjacent base pairs (dinucleotides): (example: http://http://w3dna.rutgers.edu/public/example/example_dna.txt)
So I think I need a set of general parameters to describe Shear, Stretch, Stagger, Buckle, Prop-Twist, Opening, Shift, Slide, Rise, Tilt, Roll, Twist for 16 dinucleotides.
There is a database called: DiProDB http://http://diprodb.fli-leibniz.de/ShowTable.php that already provides some parameters for dinucleotides.

Is 3DNA provide "general" parameters to reconstruct a naked double-strand DNA just like model.it did?

Thanks first.

[cif077]

I figured out an approach to deal with my question. However, I'm not sure it is good (proper) or not. Could anyone give me some comment or suggestion please?
1. I used only dinucleotide shift, slide, rise, tilt, roll and twist parameters that download from DiProDB, but ignored Shear, Stretch, Stagger, Buckle, Prop-Tw and Opening.
2. I wrote a small program to generate a base step parameter file from DNA sequence just like the bp_step.par in X3DNA example.
3. I use rebuild program to build a 3D model by using the base step parameter file in step 2.
   (rebuild -atomic M1_sense.3dna M1_sense.3dna.pdb)
 
Everything works fine. But when I use SPDB Viewer to visualize my model, I found some segment did not connected well (please see Picture 2.png in attached file).
Is this problem can be fixed?
Or this is a viewer problem? I've tried JMol, but Jmol cannot show it properly.

I upload my base step parameter for reference.

Thanks

[xiangjun]

The basic issue is that there is no one-to-one correspondence between DNA sequence and its 3D structure. For example, a 146-bp fragment could be in left-handed super-helical form as in nucleosomal DNA,  or other other conformations (straight or not). In the literature, there have been several bending models, each of which assigns a set of step parameters (roll, twist etc) to specific sequence (at di-, tri- or tetra-nucleotide level).

As noted in the 3DNA 2008 Nature Protcols paper (3DNA-NP):

"Thirdly, as demonstrated in SCHNArP, various DNA bending ‘rules’, with different sets of tilt/roll/twist values at the constituent dinucleotide or trinucleotide steps, can be easily incorporated within a script that transforms a sequence with an assigned ‘bending model’ into an input file that feeds into rebuild."

So overall, you are in the right direction, as far as building a DNA model is concerned. Now to your specific questions:

1. I used only dinucleotide shift, slide, rise, tilt, roll and twist parameters that download from DiProDB, but ignored Shear, Stretch, Stagger, Buckle, Prop-Tw and Opening.

The "rebuild" program in 3DNA can take two types of input, as documented in 3DNA-NP. You can ignore Shear, Stretch ... etc base-pair parameters, which will be taken as zeros.

2. I wrote a small program to generate a base step parameter file from DNA sequence just like the bp_step.par in X3DNA example.

It can be simpler, as noted above. Please see also 3DNA-NP: working through recipe #2 would clarify the issues.

3. I use rebuild program to build a 3D model by using the base step parameter file in step 2. (rebuild -atomic M1_sense.3dna M1_sense.3dna.pdb)

Yes, it is right.

Everything works fine. But when I use SPDB Viewer to visualize my model, I found some segment did not connected well (please see Picture 2.png in attached file). Is this problem can be fixed?
Or this is a viewer problem? I've tried JMol, but Jmol cannot show it properly.

That's normal. Again as noted in 3DNA-NP:

Secondly, rebuild allows for the construction of atomic-level nucleic-acid structures (‘-atomic’ option) with sugar–phosphate backbones in pre-assigned, fixed conformations (specified by standard residue files; see the FAQ section at the 3DNA website). Such models, which have precise base-pair geometry but approximate (sometimes distorted) backbone connections, provide a useful starting point and basis for analysis of all-atom simulations.

One more note, 'rebuild' generated PDB files have full CONECT records, and should display with all proper connections with RasMol. Of course, some of the O3'(i) to P(i+1) bonds would be quite long.

So, I would suggest you to download 3DNA v2.0, and play with the worked examples.

HTH,

Xiang-Jun

[cif077]

Thanks for your reply, for me a newbie, I do learn a lot in this new field from this forum and your kindly reply.
I've wrote a mail to Dr. Olson. I hope that I could get 3DNA 2.0 to test my data again.