Clint Howarth

Software

Worked on a production pipeline that facilitated both manual and high-throughput, automated genome annotation. This involved intricate applications with rapidly changing heuristic analyses, reading and writing arcane bioinformatics file formats, web design, automating the software whose execution case takes five pages to explain, database tuning — all using a big server farm. When I started, our department annotated and published four genomes per year. Now, we have the ability to do the same for over five thousand a year, with better quality and consistency. My work directly helped make that transition possible. I've designed, implemented, and maintained critical components, and more.

The Broad Institute isn't a software company, so working here is like working for a very small software company with an unusually steady client. I work closely with geneticists to try to help them sort, understand, annotate, and publish their technical data. In a small company, each team member must be flexible and able to do many things at the same time, handle turnover, maintain a low cost, and an ability to deliver regularly; this is the kind of environment that I thrive in.

Part of working as a small team means working closely with our customers, in this case geneticists. It means friendly, productive relationships with people from vastly diverse backgrounds, each of whom is very, very smart and know where they want to be — but not how to get there. My job is not just listening to what they say they want, but learning the nature of the goal and understanding what they need. This is one of my favorite parts of this job.

Wrote GenePidgin, an open-source suite of Python tools related to gene naming. I've presented this work at a Cold Spring Harbor Labs bioinformatics conference.

Biology

The Broad Institute is a hotshot research lab, which means in part it's hard to nail down things like departments. Teams, initiatives, and organizations are fluid. It's easier to describe the work I've been involved in:

• Genomic Sequencing Center for Infectious Diseases.
• Fungal Genome Initiative
• Human Microbiome Project
• Sequencing and annotation effort in response to the recent European E. coli outbreak

My software has assisted in annotating many horrible infectious diseases, including tuberculosis, malaria, influenza, the dengue virus, tuberculosis, Staphylococcus Aureus (of staph infection fame), HIV, and many more. Of course, I'm just working on their source code, but the compiled executables are troublesome.

From my work in annotation (and naming), I've been published as a contributing author in a few journals, most notably in Science magazine ("A Catalog of Reference Genomes from the Human Microbiome"), related to the Human Microbiome Project. A few other papers as well. For better or worse, I'm also listed as part-author on more than one million genes.

The automated classification and tagging of coding DNA is cool.