Deoxyribonucleic acid, or DNA, is the hereditary material in almost all known organisms. It is a molecule composed of two chains of nucleotides that coil around each other to form a double helix. The nucleotides are the building blocks of DNA. These blocks, unlike Legos, don’t link up with just any other block; there are four nitrogen-containing nucleobases (cytosine, guanine, adenine, or thymine), and, as you might remember from biology class, A goes with T and G goes with C.
Dr. Robert Molt, a scientist in the National Security Solutions division of ENSCO in Florida; graduate student Isha Singh; and Dr. Millie Georgiadis, professor of biochemistry and molecular biology at the IU School of Medicine, wondered: “How would DNA be different if we had more than just A’s, G’s, T’s, and C’s?”
Molt, Singh, and Georgiadis began to think through the possibility of synthesizing nucleotides that would be similar enough to bond with the others, and began working with several experimental laboratories to synthesize artificial nucleic acids.
Through their experiments and computational work, they found that the following “rules,” which have been taken as law for DNA molecules, can be reevaluated:
- that one of the nucleotides is a hydrogen bond donor, and the other is an acceptor
- that one had to have one ring, and the other, two.
Regarding the first rule, the team discovered that the synthesized nucleotides exhibit mutualistic relationships where both parts are donating and accepting on the same molecule, which might form a stronger hydrogen bond so the DNA won’t break apart easily, or be prone to mismatches. Regarding the second, though all natural DNA on earth pairs one ring with two rings, the pairs involving synthesized nucleotides could also be either skinny (a one ring-one ring pair) or fat (a two ring-two ring pair).
Similarly, though biological processes for DNA are generally all intermolecular, Molt, Singh, and Georgiadis observed a case in which an atom left one molecule and went to the other because it was more stable that way. There would be no way to see this in an organism, however, because this would be observed through crystal structures, and it’s not physically possible to see hydrogens in crystal structures.
For this project, Molt, whose background is in physics, needed to calculate the energetics of these nucleotides and how they interact in the DNA. To do so, he had to find solutions to equations using quantum mechanics. These equations require the use of thousands of processors at once in order to solve them.
Enter IU supercomputer Big Red II. Molt wrote parallel code to run on 2,240 processors so that he could solve the equations of quantum mechanics to predict the energetics, showing how nucleic acids bond to each other. As he explains, this required the use of hundreds of 32-processor nodes all together working in communication with one another to solve the equations fast enough.
Big Red II accommodates this, while also allowing access to the memory Molt needed for calculating integrals on the fly. In light of the computing power that proved essential to his research, Molt advises learning as much as possible about computer sciences, but adds that “The staff at Big Red II is very supportive and very helpful to developers.”