Clarification on how DNA radius is computed from 3DNA/DSSR

[Mamta]

Dear Xiang-Jun,
I am using 3DNA to analyze DNA structures from molecular-dynamics simulations.
I get the standard 3DNA output files with base-pair and helical axis information.

I want to calculate the DNA radius from these 3DNA results.
Could you please tell me:

Which output values or parameters from 3DNA should be used to compute the DNA radius, and how exactly should it be calculated from the 3DNA analysis?

Thank you.

Best regards,
Mamta Yadav

[xiangjun]

Hi Mamta,

Thanks for using 3DNA and for posting your question on the 3DNA Forum.

3DNA performs a least-squares fit of linear global helical axes using equivalent C1' and RN9/YN1 atom pairs along each strand of a DNA duplex. It then calculates mean and deviation of helix radius based on P, O4', and C1' atoms. Of course, the linear global helical axes is only meaningful when the DNA duplex is not strongly distorted. You may also want to check Curves+ for deriving curvilinear helical axes.

Using 355d as an example, running 3DNA as below:

find_pair 355d.pdb | analyze

You will get an output file 355d.out, which contains the following content:

Global linear helical axis defined by equivalent C1' and RN9/YN1 atom pairs
Deviation from regular linear helix: 3.30(0.52)
Helix:    -0.1269   -0.2753   -0.9530
HETATM 9998  XS    X X 999      17.536  25.713  25.665
HETATM 9999  XE    X X 999      12.911  15.677  -9.080
Average and standard deviation of helix radius:
      P: 9.42(0.82), O4': 6.37(0.85),  C1': 5.85(0.86)

You can extract the two HETATM lines into the original PDB file, and draw a line between them to visualize the global helical axis (in PyMOL, for example).

The underling algorithm is based on the NewHelix/FreeHelix program, and you can check the implementation details in the 3DNA source code.

Hope this helps.

Best regards,

Xiang-Jun