- When similar species interbreed, it negatively impacts biodiversity
- Human activity is forcing hybridization in Colorado River Basin fish species
- New technologies and advanced computation give a more complete view of genetic changes
Think about the last time your car broke down. Would you flip open the hood, detach random devices from the engine, and expect everything to work out?
Of course not. A car is a complicated machine, and each of its parts serves an important purpose in getting you where you need to go. Detach one hose, unscrew one spark plug, and you may find yourself stranded.
Sadly, humans are behaving just that recklessly with a machine larger than anything we’ve ever built. The natural world is filled with a multitude of flora and fauna, each of which play a role in their local ecology. This variety of life is called biodiversity, and the Earth can’t function correctly without it.
While biodiversity is influenced by many factors, human activities have a large impact. Tyler Chafin, a recent doctoral graduate at the University of Arkansas, is looking at how human activity has caused similar species of fish to breed together, thereby forcing hybridization.
Taking the science further
Chafin’s study focused on three fish species of conservational concern in the Colorado River Basin: the humpback chub, the roundtail chub, and the bonytail chub. His PhD advisors and coauthors on his study, Dr. Michael Douglas and Dr. Marlis Douglas, have spent time studying these species and their previous work suggested some level of hybridization.
However, these earlier studies were unable to get a fully nuanced view of the implications this hybridization might have on the persistence of populations, given the DNA sequencing technologies available at the time.
“This came down to a limitation in how many different pieces of the genome could be effectively sequenced,” says Chafin. “Quantifying hybridization is a notoriously difficult problem and doing so in a statistically relevant way requires a lot of data from throughout the genome.”
He continues, “Even though previous studies could show good evidence that hybridization had occurred, it wasn’t certain whether or not it was currently occurring, much less if there was any sort of negative impact on populations.“
What Chafin and his colleagues found was that hybridization was having a negative effect on the viability of offspring of the humpback and roundtail chubs. They discovered a lot of evidence for selection against the resulting hybrid individuals. This makes sense considering the fact that the fish in the study live in a sometimes-hostile desert environment and are therefore highly specialized.
The boundaries between species that have been genetically distinct for millions of years are now dissolving, especially in places with a large amount of human activity. This is especially true when human water usage causes reduced water levels.
“The proportion of water being removed at various sites significantly predicted the loss of purity in these populations,” says Chafin. “At around 0-2.5% water removal, four out of five populations were comprised of 75-100% pure individuals; at 22.5-25% usage (the highest we measured), populations were 0-10% pure.”
While Chafin and his colleagues studied other areas of human activity, water removal serves as a good proxy for overall anthropogenic impacts. It’s quite clear that human activity is causing certain fish species to hybridize—and that’s bad news for our planet’s fragile biodiversity.
Big problems, big computers
Previous studies were limited by the capacity of DNA sequencing technologies available at the time. Now that these tools have advanced, scientists are able to answer more questions. That said, these new tools require some heavy lifting that can only be accomplished by high-performance computing (HPC).
“We did a lot of the work on the Trestles cluster here at the University of Arkansas,” says Chafin. “However, many of our applications don’t really take advantage well of the highly distributed traditional HPC model because they often are high memory, long-running, have many dependencies, and are of limited scalability (i.e., hundreds of processor cores can’t be used).”
Chafin continues: “For this reason, we performed a lot of analyses on the XSEDE Jetstream cloud at Indiana University and TACC. This was really convenient for us because I could set up our messy web of dependencies on a virtual machine and boot up across a few high-memory Jetstream instances analyses that could run uninterrupted for 10-20 days at a time.”
Although modern technology is helping us better understand the issue at hand, the problem remains the same – human activities are negatively affecting our world. The kind of car troubles you’ve run into in the past pale in comparison to the interlocking gears that allow life as we know it to thrive on Earth.
But unlike your car, we don’t have an Earth mechanic to turn to. There isn’t a tow truck coming to save us. The only experts we have are pointing to the “check engine” light flashing on our collective dashboard. We can either listen to them or blow a global gasket unlike anything our species has ever seen before.