How does 3DNA calculate rise

[auffinger]

Hi Xiang-Jun,

I just have a silly question but I am unable to
find a simple description of how you calculate the rise values
and, incidentally, all other values.

Thanks,

Pascal

[xiangjun]

Hi Pascal,

Have a look of $X3DNA/doc/tech-details.pdf, and/or read the source code.

HTH,

Xiang-Jun

[auffinger]

Hi Xiang-Jun,

I found this definition regarding rise in the Calladine 1995 JMB paper:

Rise, Dz , is simply the component along the global
helix axis of the vector joining the mid points of
successive C6-C8 lines.

Is it calculated the same way in 3DNA ?
Sorry, I am not good at looking at the code.

Best,

Pascal

[xiangjun]

I found this definition regarding rise in the Calladine 1995 JMB paper:

Rise, Dz , is simply the component along the global
helix axis of the vector joining the mid points of
successive C6-C8 lines.

I was puzzled by your above finding of 'Rise definition' in the Calladine 1995 JMB paper, titled The Assessment of the Geometry of Dinucleotide Steps in Double-Helical DNA; a New Local Calculation Scheme". Checking carefully, I noticed that the above citation comes from the "Appendix: NEWHELIX Definitions" (p.662) which uses a global helical axis, as further described below:

The expressions for Rise and Twist are clearly defined in a global sense, and they therefore depend on the overall conformation of an oligomer. Thus while these expressions are probably valid for, say, B-DNA oligomers where all base-pairs are nearly perpendicular to the overall fitted helix axis, they may not be appropriate for the analysis of other oligomers such as those that adopt the ‘‘A’’ form. It should be noted that since the magnitude of Helical Twist is in general substantially greater than the inclination of the base-pairs to the average best-fit helix-axis, the NH-values for Helical Twist are unlikely to be very different from those given by CEHS.

The CEHS definition of translational parameters (Shift-Dx, Slide-Dy, and Rize-Dz) comes on p.654:

The displacements: Dx, Dy and Dz can now be obtained by resolving the vector joining the origins of the two original base-pair triads into components along the axes of the mid-step triad.

Yes, that's the definition used in 3DNA as well.

HTH,

Xiang-Jun