By Kimberly Mann Bruch and Advay Shindikar, San Diego Supercomputer Center, and Ken Chiacchia, PSC

Red-tailed boa

Red-tailed boa. The species is classified as “vulnerable;” scientists would like to see it protected. A study of a devastating virus that affects it is providing clues that may help, as well as offering possible therapeutic targets for human illness caused by arenaviruses. Adobe Stock Images:

New DNA Sequence Provides Clues to Virus’s Spread, Possible Targets for Future Therapies

Just like humans, snakes can get sick, too. Using PSC’s Bridges-2 and Jetstream2 at Indiana University, a University of Tennessee Veterinary Medical Center team uncovered the complete DNA sequence of an incurable virus found in the vulnerable Colombian red-tailed boa. The work promises to help scientists seeking to protect the species, as well as doctors fighting viral infection in humans.


Every species that goes extinct is a loss in and of itself. But the knowledge represented by that loss is even greater. We can’t know what useful secrets about how living creatures and their internal processes work that we would have learned had the species survived.

A scientific team led by veterinarian Dr. Mohamed Abouelkhair, an immunology and virology assistant professor at the University of Tennessee Veterinary Medical Center, wanted to know more about how reptarenavirus can cause Boid inclusion body disease, a devastating, incurable disease of the vulnerable Colombian red-tailed boa constrictor.

The snake’s official classification as vulnerable is a big part of the story. Scientists have worked hard to create a consistent system that helps them identify threatened species. But it’s much harder to do so accurately when our knowledge about them is incomplete. Among reptiles, 21.1 percent are classified as vulnerable; so are 13.6 percent of birds. But the relative lack of information about reptiles, and particularly about the microbes that infect them, puts an extra question mark over just how endangered they are.

“Our research … [can] aid in developing ways to diagnose and intervene in case of infections, ultimately contributing to the conservation of these reptilian species. By unraveling the mysteries surrounding viral pathogens in snakes, we can better safeguard the health and conservation of reptiles, as well as mitigate potential risks to human health.” — Dr. Mohamed Abouelkhair, University of Tennessee Veterinary Medical Center

The serious illness caused by reptarenavirus in this boa constrictor prompted the University of Tennessee collaborators to carry out a study of the virus’s RNA sequence. Knowing how the pathogen’s genes are put together could offer clues on how to protect the snakes from it — clues that possibly could be applied to related arenaviruses that attack humans. To achieve this gene assembly project, the researchers turned to PSC’s flagship Bridges-2 supercomputer, as well as the Jetstream2 supercomputer at Indiana University (IU). Both of these NSF-funded systems are made available to U.S. researchers through the agency’s ACCESS program, in which PSC is a leading member.


Gene assembly is a brute-force computing task that’s very different from classical supercomputing problems. When you’re trying to calculate the airflow over a new aircraft, it’s a matter of carrying out many complex equations, one for each bit of air streaming over the plane’s surface. With gene assembly, though, it’s less of a math problem and more a game of Concentration.

Scientists sequence RNA by converting the genomic RNA into complementary DNA or cDNA and then cutting it up into many smaller pieces. This is because there’s a limit to how many DNA bases — the links in the nucleic acid chain — a sequencing machine can read in a row. By cutting up a species’ genome into sequencer-digestible bits, they wind up with millions of overlapping chunks of the DNA.

You can figure out where these bits of DNA sit in the whole genome by looking at the overlaps. Where one piece overlaps another, their sequences will be identical. So rather than crunching complex math as in the airflow problem, it’s a matter of storing each sequence in readily accessible memory — the same as RAM in a personal computer — so that when the computer sees a sequence that matches one it’s seen before, it can rapidly match them.

For the assembly task, the team used IU’s Jetstream2 supercomputer as well as the large-memory capabilities of PSC’s Bridges-2 system. They used the computers to assemble and analyze the genome of a reptarenavirus they isolated from a single Colombian red-tailed boa.

Designed from the outset to offer different computing elements that cooperate to accelerate both classical and “new community” science problems, Bridges-2 in particular offers huge memory. “Large memory” nodes in many supercomputers can be as large as 128 gigabytes — individually nearly 10 times as much as the total in a high-end laptop. Bridges-2’s “regular memory” nodes, which the University of Tennessee team has used, have 256 or 512 GB RAM.

The vast size of Bridges-2’s memory enabled the machine to store many more sequences, immediately available for comparison, than most other supercomputers. This helped speed the assembly. But the ability to store more DNA fragments in memory also helped the scientists compare the sequence of their virus to those of other variants, allowing an analysis of specific ways in which its genome helped it attack the snake. Overall, Bridges-2 significantly accelerated the work, reducing the time required from weeks to just several hours.

“[Bridges-2] played a pivotal role in our research by facilitating intricate analyses and computations on the metagenomic data of the reptarenavirus. Leveraging the supercomputer’s immense processing power, we were able to perform complex tasks such as taxonomic classification, identifying the virus variant, assembling genomic sequences, and phylogenetic analysis. This capability was crucial for handling large datasets efficiently and extracting meaningful insights.” — Dr. Mohamed Abouelkhair, University of Tennessee Veterinary Medical Center

A subsequent analysis of the viral genome showed that it carried a particular DNA sequence called the University of Giessen virus (UGV-1) S or S6 (UGV/S6) segment and L genotype 7. This short segment of RNA had been shown by other scientists to give the virus a survival advantage, making it more infectious compared with other S segment genotypes of reptarenavirus. This information offers a clue as to how the virus persists — and how it might be thwarted. The team reported their results in a paper in the Virology Journal in November 2023.

The team plans to next use ACCESS resources to delve deeper into the evolutionary history, molecular epidemiology, and biological properties of reptarenaviruses in various snake species — not just boas. The work also offers clues to combating human illnesses caused by related arenaviruses.