Spark is an interactive clustering and visualization interface built with epigenomics data in mind. Instead of visualizing data along the linear genomic axis, as in a genome browser, Spark lets you cluster your data along points of interest, like transcription start sites, and rapidly uncover core patterns in your data. 

​

I designed and developed Spark at the Michael Smith Genome Sciences Centre and later supervised students to extend its functionality. My contributions included:

​

• Concept: Envisioned the project based on observed research needs.

• User research: Collaborated extensively with members of the US National Institutes of Health Roadmap Epigenomics Program to understand analytical needs and iterate on the software.

• Interface + visualization design: Designed all aspects of the visualization and UI

• Implementation: Wrote the application in Java including backend data management and multi-threaded clustering for computational efficiency.

• Supervision: Later supervised junior developers to extend functionality.

• Project management: Planned and prioritized features.

• Scientific writing: I wrote the Genome Research paper describing the methods and applications.

​

More information and videos available at sparkinsight.org.