difference between rise/twist and helical-rise/twist

[Rajendra Kumar]

Hello,

I would like to know difference between rise/twist and helical-rise/-twist. For example, if I would like to calculate contour length of the long DNA, whether sum of rise or sum of helical-rise would give the contour length of the entire DNA.

Thank you very much in advance.

With best regards,
Rajendra

[xiangjun]

Hi Rajendra,

The differences between rise/twist and helical-rise/-twist were described in the 2003 "3DNA NAR paper".

Two sets of such parameters are commonly used in the literature (Fig. 1, upper and lower right): the set of local base pair step parameters, Shift (Dx), Slide (Dy), Rise (Dz), Tilt (τ), Roll (ρ) and Twist (ω), which describe the stacking geometry of a dinucleotide step from a local perspective; the set of helical parameters, x‐displacement (dx), y‐displacement (dy), helical rise (h), inclination (η), tip (θ) and helical twist (Ωh), which describe the regularity of the helix, e.g. helical twist is the angle of rotation about the helical axis that brings successive base pairs into coincidence. In 3DNA, the helical axis is defined at a local level, following Bansal et al. (31), as (x1 – x2) × (y1 – y2). This axis corresponds to the single rotational axis that brings the reference frames of successive base pairs, (x1, y1, z1) and (x2, y2, z2), into coincidence, and its location follows Chasles’ theorem as detailed in Babcock et al. (29). The calculations of x‐displacement, y‐displacement, tip and inclination in this reference frame are based on the definitions of Lu et al. (34).

Exact details can be found in the "tech_details.pdf" document distributed with 3DNA (in the $X3DNA/doc directory).

3DNA does not calculate contour length, and I cannot give an advice as to whether sum of rise or helical-rise (OR some other distance) should be used. I welcome your comment in this specific area.

HTH,

Xiang-Jun

[Rajendra Kumar]

Hi Xiang-Jun,

Thank you very much for your reply. In some of the publications, sum of rise was considered to be the contour length of the DNA. However, when I used rise/twist to calculate some quantities, MD simulations did not reproduce these quantities with experiments (optical tweezers and AFM). These quantities were calculated using contour length, which was directly measured in these experiments. In contrast, when I used helical-rise/-twist, these quantities are exactly reproducible with experiments. How would I explain this discrepancy? Any comment would be helpful.

I have also observed that the coordinates of local helical axis that are calculated from 3DNA, are very noisy (fluctuating) during MD simulations. If the helical-rise is calculated from local helical axis, sum of helical-rise is larger than the contour-length. For example, as depicted in the attached figure, length of red-curve (position of local helical axis from 3DNA) is larger than the blue-curve (smoothed helical axis).

Thank you very much in advance.

with best regards,
Rajendra

[xiangjun]

Hi Rajendra,

Thanks for your feedback, and your figure looks nice!

In contrast, when I used helical-rise/-twist, these quantities are exactly reproducible with experiments. How would I explain this discrepancy? Any comment would be helpful.

I do not quite get it. Are you saying that you can reproduce experiment exactly using helical-rise/-twist? Do you have a concrete example? Or am I missing your point?

I have also observed that the coordinates of local helical axis that are calculated from 3DNA, are very noisy (fluctuating) during MD simulations.

The local helical axis is sensitive to distortions of a dinucleotide step, and it only aligns perfectly for regular structures (as shown in Figures 4 and 9 of the 2003 3DNA NAR paper). The smoothed helical axis you produced not just looks pretty but it also makes sense (to me). What's wrong going along with it?

Best regards,

Xiang-Jun