UnifyBio: Power Tools for Translational Data Science - Benjamin Kamphaus

Benjamin Kamphaus presents UnifyBio, a Datomic-based platform for data harmonization, ETL, and scalable access in translational research.

Datasets in translational research are intrinsically relational and deeply interconnected. Despite this fact, the difficulty of handling data from raw clinical and molecular sources produces a multitude of siloes around institutional and subdisciplinary seams. Computational biologists are forced into narrow specializations around particular data types, with the scope of their efforts bounded by a folkloric understanding of the relevant pipelines and analysis packages. UnifyBio is a set of power tools aimed at dismantling these barriers to big picture thinking. Built with Datomic at its foundation, Unify simplifies data harmonization, ETL, and scalable data access. At RCRF, we use it to co-locate high-quality clinical and molecular datasets by extracting data out of disparate raw files and ad hoc tables and into unified representations in our Pattern Data Commons. This effort has opened up new lines of inquiry, allowing us to see across interconnections in data that often remain invisible. This talk details how using this toolkit has enabled us to take novel approaches to unraveling the puzzles underlying rare cancers. Biography Benjamin Kamphaus is a computational scientist, composer/producer, keytarist, and occasional sci-fi author based in Colorado. He serves as Technical Fellow at the Rare Cancer Research Foundation, where he handles the design and implementation of the Pattern Data Commons—a unified repository for patient-contributed molecular and clinical data. Recorded Nov 14, 2025 at Clojure/conj 2025 in Charlotte, NC. clojure-conj.org