Biology, Software Development, and Bioinformatics at the edge of discovery
About me
I am a software engineer with a meandering work history, and highly interdisciplinary training background. I received my undergraduate degress in microbiology and molecular genetics from The Ohio State University, and my Ph.D in organic chemistry from the University of Missouri. I trained as a postdoc, first as a T15 training fellow, and then as a regular postdoctoral researcher in the Department of Biomedical Informatics at the University of Pittsburgh, and eventually transitioned to a Research Software Engineer role at Pitt. I am currently the Developer Engagement Lead for Bioconductor and am based out of Limerick, Ireland.
Current Role
My current role in developer engagement is funded through the Chan Zuckerburg Institute, and has a broad mandate for improving developer interactions with the Bioconductor project. This includes, but is not limited; improving developer training resources, improving coding practices and standards within the Bioconductor ecosystem, addressing technical debt and planning for forward compatibility within the Bioconductor ecosystem, and decreasing the technical and conceptual gulfs between Bioconductor users who are not developers, and bioconductor users who are very much developers.
My current role is a change of pace from solely software development and an opportunity to have a bigger impact on how R as a language and Bioconductor as a project can continue to play a pivotal role in scientific discovery writ large, and both academic and non-academic bioinformatics.
Current Projects
In my current role, I have some flexibility to spend a small amount of time on technical projects of my own. The following projects are under current, albeit somewhat slow paced, development.
BiocMetal
Robust and reliable programmatic access to GPU compute is still largely absent from the R programming language. The reasons for this absence are mostly historical, and I have started to explore ways to address that gap in R and Bioconductor’s technical capabilities with a package currently under the working name BiocMetal. Though this project currently lends itself to more generalizable programmatic GPU compute, ideally it will eventually encompass access to, and leveraging of, large sequence based language models in R.
SynExtend
SynExtend was originally envisioned as a relatively simple tool for performing orthology inference from synteny maps alone. It has grown over time to include contributions that are outside of that scope and is still a work in progress. Development, validation, and testing of tools within the package is ongoing.
Others
I am generally excited to lend my expertise to projects that can benefit from my unconventional mix of skills and experience. Feel free to reach out!