By Michaela Webb, NASA Space Grant Intern for CyVerse
Ashley Maynard recently used CyVerse tools to complete her master’s thesis at California State University, East Bay, where her work focused on understanding why certain populations of Olympia Oyster in the San Francisco Bay are better at surviving baywater salinity changes than others. She now uses those same CyVerse tools to understand cancer biology in humans.
The Olympia Oyster, Ostrea lurida, which once lived abundantly in California’s estuaries, is now all but extinct. The important ecosystem services it provided, including habitat for other species, shoreline protection, and improved water quality through filtration, have disappeared with the oyster, leading to a decrease in the health of estuary ecosystems.
Efforts at restoring the important species are being made, but need to take into account an increasingly fluctuating climate. “Here in the Bay Area we see increased rainfall during the winter months and increased freshwater in our estuaries, meaning that salinity changes in a very short period of time,” Maynard said. “I started this project trying to identify why one population seemed to be more robust to salinity changes. If I could identify the molecular markers, specifically genetics, that could help me to understand salinity fluctuation tolerance in these oysters.”
Maynard used RNA-seq, a next generation sequencing technique for mapping the presence and quantity of RNA in a biological sample, to compare the genes of three different oyster populations in the San Francisco Bay. She then built a transcriptome, which is all of the active RNA in the oyster.
But she had a dilemma: Unlike human genetics, Olympia Oyster genetics haven’t had much scientific interest. This means that understanding the functions of different genes in the transcriptome (also known as annotation) wouldn’t be an easy task.
“When you’re working with a non-model organism you need to try to annotate the transcripts and later identify the function. We had about 50,000 different transcripts to annotate,” Maynard said. This presented a computational challenge that Maynard initially didn’t know how to solve.
Then she met Blake Joyce, a CyVerse science informatician, at the 2016 conference of the Society for Integrative and Comparative Biology in Portland, OR. Joyce connected her with an annotation program hosted by CyVerse called Trinotate. Using Trinotate, Maynard annotated most of the transcriptomes over the course of a few weeks. “It was really critical to be able to annotate that much data,” Maynard said.
To understand the function of different oyster genes, Maynard and her team compared their transcriptome to the genomes of organisms that have already been studied extensively. When they found a match between an oyster gene and a known gene in another species, they could infer the name and function of the oyster gene.
Often when researchers publish transcriptomes for non-model organisms, they are only able to match between 15 and 40 percent of the genes in their transcripts to known genes. Using Trinotate, Maynard and her team were able to match about 50 percent of their transcriptome to known genes. “We did pretty well compared to the other people that are publishing in the de novo transcriptome world,” Maynard said.
Another challenge that Maynard confronted in her research was finding a reliable place to store her data. “I was having problems with my school campus internet connection. It kept breaking down,” Maynard said. “Having access to the CyVerse Atmosphere cloud was really nice because it gave me consistent access to my data.” Atmosphere is CyVerse’s computational cloud service.
Maynard and her team hope their work on Ostrea lurida will help resource managers determine which oyster populations are best suited for reintroduction to the San Francisco Bay and elsewhere.
Maynard now works as a junior research specialist at the University of California, San Francisco in Trever Bivona’s precision cancer medicine lab, where she still uses many of the same tools as she did for her oyster research. “I love working with human genes because I have a puzzle I can map to now,” Maynard said. “The coding that I did with Blake Joyce was really helpful. A lot of the skills were transferrable.”