One of the Molecular Visualization Principles I am currently developing is “avoiding snake motion”. This refers to the depiction of long polymer motion, typically mRNA and DNA. To illustrate this principle, we decided to use a short linear strand of DNA.
I wanted to maintain consistency of style with the other principles, so I was aiming for a specific representation that was an interesting challenge to achieve.
I generated a mesh using Molecular Maya, and then decided to retopologize it by hand in order to create clean topology that was based on the atomic structure. Each nucleotide was made from two face loops. Edge loops were continuous along the long axis & backbone as well.
Unfortunately, the short segment of DNA from the pdb (1BNA) is not perfectly linear. It has a twist, which I had to correct in order to duplicate and fuse the segments into a long straight mesh. I used a number of tricks, tweaks, and deformers, which resulted in a repeatable model, albeit one that still had a slight twist at a longer scale.
I duplicated the segments and used a three-point snap to place them correctly. Again, because of the larger scale deformation, I couldn’t just translate each segment a certain distance. The edge loops that run along the backbone were very helpful in connecting these without any guess-work.
The model smooths fairly well, isn’t too dense, is nicely editable, and still maintains some surface variation rather than mathematical perfection.
I chose a shader similar to the ones I’ve used previously in the series of molecular visualization principles. In order to easily differentiate the ends, I wanted to create some color variation along the length of the linear DNA segment. However, the initial color scheme didn’t have meaning from a biological stand-point, potentially leading to confusion.