Thursday, September 02, 2004

All models, all semester

No, not the Cindy Crawford/Linda Evangelista/Giselle Bundchen type. Rather, computational models of biological systems. They've got curves in them too, but they're less likely to get people excited than the curves on the other type of model.

What I'm referring to is the way my first semester is shaping up. I have to do three one-month long "rotations" in different people's labs, where I basically figure out what the folks in that lab are working on, how I feel about it and hopefully do some of the work myself. The idea is that you do a rotation in the labs of three faculty members you think you might like to do your thesis research under and then pick one of them as your thesis advisor at the end of the first semester.

So far, I've lined up rotations in two labs: one in Drew Endy's lab [working on synthetic biology] and one in Doug Lauffenburger's lab. Doug's lab is working on computational models of various cell processes, like the decision to die vrs reproduce, how stem cells decide to become liver cells vrs heart cells vrs brain cells etc, and they're using various types of models. My rotation in his lab is going to consist of taking two models [partial-least-squares and Bayesian networks, for anybody who cares -- I'm not quite sure what those are myself ;-)] that have been used to analyze a particular data set and see whether I can extend them a bit and re-analyze the data set. So, it'll be a useful introduction to two types of models. Similarly, in Drew's lab, I'm going to be playing with a model they've built and use it to analyze some data that's being generated in the next month or so. So, that's a third type of model I'll learn about.

And, as if that's not enough, I'm taking a class, Biological Engineering 420 ["Biomolecular Kinetics and Cellular Dynamics"], that focuses on yet a 4th type of model, namely coupled differential equations and chemical kinetics to analyze various processes. Hence, all models, all semester.

I'm not sure where my third rotation will be yet, but I think it might be in Tom Knight's lab. Tom Knight used to be an electrical engineer/computer scientist, but about 10 years ago decided that the future was in biology, so he basically turned into a biology graduate student, took a bunch of classes and is now also working on synthetic biology. He's mainly doing experimental work, taking apart one of the smallest organisms discovered to date that is capable of self-reproduction on its own [ie viruses don't count], called Mesoplasma florum, and attempting to put it back together to figure out the way it ticks. It's a simple organism, with "only" about 600 genes, so it's reasonable to think that it's possible to construct a full-fidelity model of what all its genes and proteins do and their interactions with each other. However, at the moment, he's mainly doing experimental [ie lab bench] work, so working in his lab would allow me to satisfy the experimental part that's required in my lab rotations.


Post a Comment

<< Home