Powerful supercomputers around the country continue to push advances in critical areas, such as cancer research. One of the latest such success stories comes out of The University of Texas at Austin, where Texas Advanced Computing Center machines revealed a connection between cross-shaped (cruciform) DNA segments and human cancers. By shedding light on the pathways involved in cancer formation, the research may lead to improved cancer prevention and treatment efforts.

TACC science writer Jorge Salazar relates how UT Austin scientists relied on its Stampede, Lonestar and Corral machines to uncover a link between short inverted repeat sequences of DNA nucleotides (aka cruciform DNA) and and human cancer. It’s the kind of study and results that would not be possible without cutting-edge HPC resources.

One of the most important aspects of cancer research is figuring out what makes a normal cell turn into a cancer cell, and that’s why this new insight is so meaningful.

Cruciforms are a common DNA structure. Comprised of short inverted repeats (a DNA sequence followed by its reverse compliment), they play an important role in the regulation of essential processes, such as DNA replication and gene expression.

Using Stampede and the Lonestar clusters, the researchers combed the COSMIC database of non-inheritable human cancer mutations to locate short inverted repeats of 30 base pairs and under. What they found was surprising. The short inverted repeats were enriched at the site of translocation breakpoints in human cancer genomes. Translocations are spots where DNA has been broken and repaired. Although a common occurrence, the breaks are implicated in cancer development.

These cruciform-forming sequences are essentially markers for the chromosome breaks, but did they initiate the cancer development? The research team discovered that there are at least two ways the cruciforms could stimulate the formation of DNA double-strand breaks, one involving the impeding of DNA replication and the other an aborted repair process (see figure below).

The work is highly computationally-intensive due to the large number of combinations being searched, as researcher Albino Bacolla explained.

Albino Bacolla works with lead investigator Karen Vazquez, the James T. Delucio Regents Professor in the Division of Pharmacology and Toxicology at The University of Texas at Austin.

“For every position along the DNA, the program has to perform several hundred iterations,” said Bacolla. “Then the number of these iterations needs to be multiplied by the length of the DNA, then by the number of the translocations in our cancer patients.”

That comes out to about two billion iterations, putting this project outside the realm of the fastest desktop machines and into the realm of supercomputing.

“It would not have been possible to do this job without the TACC resources,” Bacolla said. “The center is an incredible resource in terms of its capacity and support. We have been using the resources and staff support for some time now. It’s a wonderful opportunity for researchers at UT Austin.”

Professor Vasquez sees this as a plausible explanation in human cancer etiology. “That gives us hope, inspiration, and enthusiasm to move forward,” she shares.

The study was funded by The National Cancer Institute, part of the National Institutes of Health. The project is covered in more depth at TACC’s website and published in the journal Cell Reports, coauthored by Steve Lu, Guliang Wang, with Albino Bacolla, Junhua Zhao, Scott Spitser, and Karen M. Vasquez.