Oddity in 3DNA Analysis of Non-Natural Nucleic Acid

[rwmolt]

Good afternoon!

I am seeing something rather anomalous in the application of 3DNA toward a given PDB structure. A sample PDB is attached that shows the oddity. If you run 3DNA, it will report an X-displacement of 69Å. It is not unique to this one PDB; I have examples ranging as high as 300Å. Obviously the x-displacement is, empirically, not 69Å for any dinucleotide step. In particular, it occurs for a non-natural nucleic acid

Other parameters look "reasonable," but I lack an independent measurement check, given that this has non-natural nucleic acids. May I ask if it is obvious to you what is amiss in 3DNA's analysis?

[xiangjun]

Hi,

Thanks for posting your questions on the 3DNA Forum.

What you say as 'oddity' in 3DNA output is actually as expected. The unusually large X-displacement you noticed in your attached PDB file (and others) is calculated in exactly the same way as other "reasonable" ones. Thus it is a reflection of abnormal steps in your structure.

To verify the reversibility (rigor) of the 3DNA analyze/rebuild procedure, do the following:

Code: [Select]

find_pair 1623.pdb | analyze
rebuild -atomic bp_step.par model-step.pdb
rebuild -atomic bp_helical.par model-helical.pdb

You will see "model-step.pdb" and "model-helical.pdb" are virtually identical (with RMSD=0.001 Å). Moreover, re-analyze "model-step.pdb" and "model-helical.pdb" will get back essentially the same results as from "1623.pdb" one started with.

If you are interested in the details, please follow the steps as described in "tech-details.pdf". You should get the same results as reported in the 3DNA output.

HTH,

Xiang-Jun

[rwmolt]

Thank you for your expertise. This is illustrative of the care that must be exercised in interpreting what this parameter means; the raw values were disturbing when interpreted in the traditional sense of what this looks like for a normal DNA helix. I appreciate your help.