MIT DNA Wizard Predicts How to Make Pyramids, Cubes, and More Out of DNA

Technology doesn't exist yet to make his predicted 3D DNA wireframes, but he's validated his model on exotic 2D structures

When you're a super-intellect at the Massachusetts Institute of Technology (MIT) sometimes you don't really worry about what practical uses of your science experiment. Sometimes you just want to make something that looks cool.

I. Spicing Up the Double Helix

That seems to be the story with MIT's new DNA (deoxyribonucleic acid) "coasters". They're beautifully patterned, outlandish polymeric constructs. Oh there's the usual justification about potentially using the technology to create custom DNA scaffolds for novel drug targeting. But mostly they just look really incredibly awesome.

In nature, of course, DNA is most commonly found in the familiar double helix. Chromosomes are formed of incredibly long strands of DNA coiled around polarized proteins, which then further assemble into a tightly compacted, information dense structure. The genome is selectively uncoiled for processing.

In the past two decades scientists have toyed with the idea of more exotic DNA structures. Some of the early efforts have gone towards creating so-called "DNAzymes" (aka deoxyribozymes). While not thought to occur naturally, these synthetic enzymes can mimic the action of ribozymes and have potential in targeted killing of cancer cells. There's also been investigation into using different interesting DNA structures such as hairpin loops to create DNA computers.

To move beyond such simple tinkering, MIT Biological Engineering Professor Mark Bathe devised algorithms that take into account the associative behaviors of certain DNA microstructures and molecular weights of the chain. The algorithms start by predicting small ordered patterns that typically looked like squares, rhombuses, or other simple shapes. These shapes are then used to form the basis of larger molecules.

II. From 2D to 3D in No Time Flat

His model can predict a variety of coaster-like exotic cylindrical relatively flat DNA patterns, some of which can be validated with experiment.

The principal innovation was in recognizing that we can virtually cut these junctions apart only to reassemble them in silico to predict their 3-D structure. Predicting their 3-D structure in silico is central to diverse functional applications were pursuing, since ultimately it is 3-D structure that gives rise to function, not DNA sequence alone.

Excerpt from:
MIT DNA Wizard Predicts How to Make Pyramids, Cubes, and More Out of DNA