Getting the most out of Eddie and DataStore
High computational power is required for this work, which involved an annotated high quality genome assembly along with resequencing data for all the individuals in the study. For this, Emily turned to Eddie, the University of Edinburgh’s supercomputer.
She explains, “Eddie is what everyone in our group uses for genomic analysis. It is useful, because it has various preinstalled tools, for instance tools for calling genetic variants, mapping sequences to the genome and annotating variants”.
Emily has a background in bioinformatics, and used similar tools during her PhD, so she is familiar with how Eddie functions. For newbies to her research group, she always recommends the Eddie course on the university website. “The course goes through all the capabilities, from start to end”. Moreover, as Emily is a bit of a pro herself, she has started an informal bioinformatics drop-in clinic where she helps people with analysis problems or issues with Eddie.
One of the biggest challenges Emily faced using Eddie was maximising the potential of the supercomputer. As her datasets are so large, some jobs take several days to run. She has just requested priority access to some nodes, to minimise the time and optimise output. Eddie is free for every researcher at the University of Edinburgh, however, with a bit of funding, it is possible to “skip the queue” of jobs and get results faster.
All raw data, and all completed analysis are stored on DataStore. Several members of the Conservation Genetics group work with large scale datasets, and they require sophisticated data storage and organization tools. They take advantage of the large DataStore capacity by requesting additional space beyond the initial 500 GB free allocation.
More on Eddie
More on DataStore
Tools for funding and publishing
The cost of Eddie and DataStore give a glimpse into the funding required for the digital aspects of her project. A lot of her work is supported by grants from philanthropic organisations. “I use Research Professional to keep on top of all these grants and application deadlines. It is a nice platform that consolidates all the different grants in one place”.
Her project on the antelopes has just been published. The team found that the managed population of oryx has reduced levels of close inbreeding, a reduced burden of genetic mutations and enhanced genetic diversity, favouring their ability to make adaptions to changing environmental conditions in the wild.
Emily explains that she will use Pure to document her research outputs. She will also make her code public on GitHub. “I am passionate about making data analysis pipelines open access and support reproducibility across research. On several occasions, I have been contacted by people that have made use of my code”.
A promising outlook for the scimitar-horned oryx
scimitar-horned oryx have already been released back into the wild. The wild population has grown to 400 individuals over the past five years. The research findings from Emily and her colleagues will help guide further reintroduction efforts, optimising the mixing of the gene pool and driving population growth. Above all, they have also demonstrated the benefit of captive breeding programs.
This case study was written by Dr Sarah Janac, Research Facilitator for the College of Medicine and Veterinary Medicine.
Dr Emily Humble is a researcher fellow in Conservation Genetics at the Royal (Dick) School of Veterinary Studies and the Roslin Institute.
Twitter: @emilyhumble
"Meet the Scientist" profile
