Below are some ongoing projects I'm working on.
Statistical modeling of HIV adaptation. Here's the big idea: HIV mutates so fast, that there's a tremendous
amount of genetic variation both within and between patients. This means that HIV "samples the search" space
very efficiently. The bad news is, this means HIV adapts to dugs and our immune system very quickly. The good news is,
we can use this as a research tool: identify what is adapting, and what is driving adaptation, and you've learned something
about virology and immunology. See my home page and paper descriptions for some cool examples
of how this plays out. A webservice running these tools, as well as source code, is available here.
False discovery rates for contingency tables
If you could design the best tool imaginable for monitor viral infections, what would you do? How about randomly sample blood
from people and vertabrates (since most nasty viruses hang out in natural hosts, many of which are unknown)? Well, good luck
getting a robot to do that (or an IRB to approve it), but mosquitos are doing that for us already. So let's reinvent how
mosquitos are captured--smart traps, delivered by automonous drones--then sequence the blood meal of the mosquitos. Sounds crazy right?
Well, we're trying it anyway!