Learning something new: Integrating computer vision into my research.

I recently started my sabbatical year, where I get a chance to re-tool myself, and my research knowledge. The game plan is to learn how to integrate computer vision (machine vision) and machine learning approaches into my research, in particular with respect to the study of animal behaviour and the analysis of images (and videos). We study the evolutionary genetics of complex phenotypes in my lab. While this used to (mostly) mean the complex structure of the shape and size of fruit-fly wings, we are moving more into the study of animal behavior. In particular how flies evade and escape being eaten by predators (more on that at a later date).

 The analysis of such data (both huge sets of wing images as well as video, which is effectively a series of images) can be time consuming and what can be done manually is somewhat limited (such as with JWatcher), in particular if you want to do "high throughput" work with many samples. I have over the past few years interacted, and begun to collaborate with scientists who are using all sorts of techniques from computer vision which have amazed me, both with respect to the speed of the analysis, but also the detailed information gleaned from such approaches. So I am trying to get up to speed and see how to utilize these approaches for my own work.

 To that end I will be now posting about this experience (as well as all of the more usual genetics). This will include useful new tidbits, programming scripts, software I have tried (and tutorials), books, and anything else I can think of. Basically my research progress journal for this new endeavour. I hope that this will help me stay nice and organized, and perhaps will be useful more generally. If you start to follow this thread, and have suggestions for anything, please let me know in the comments or on twitter.

More to follow soon!

Retreating to Communicate

While I know many people expect my job as an evolutionary geneticist is to work alone in a dark tower (but only during a thunder storm), creating new breeds of monstrous fruit-flies (maniacal laugh... maniacal laugh), I actually spend most of my time doing something else entirely, trying to communicate how cool my fruit-flies really are. Well not so much the fruit-flies, but the questions we address using our little critters. Since I am unlikely to be creating a freeze ray to help me take over the city some time soon (Sorry Dr. Horrible, but it is unlikely I will be in your posse anytime soon), I generally use the more traditional means of communication, like presenting my work to my peers. This generally occurs in one of two ways, first presenting a "talk" or seminar at a University or a professional conference, or the far more common (and meat and potatoes of my field), by writing research manuscripts.

Now as many folks are aware of, writing can be (at least for some people, and that includes me) hard work. As a scientist, in addition to helping other people in the lab with their experiments, analyzing their hard earned data, and writing their papers (not to mention teaching, meetings, etc...). It turns out that many others struggle with this as well. Well, a few months ago I received an interesting email for a Writers Retreat held on the campus of the University of Nebraska-Lincoln. So here I am (having just arrived an hour or so ago). In addition to quiet time to write, there will be opportunities for professional coaching, and peer feedback, which I am very excited about.  We have also been asked to read the first thirteen pages of "Explaining Research: How to reach key audiences to advance your work" by Dennis Meredith. This book (or at least the first thirteen pages) reminds us as scientists to do a better job communicating, not only with each other, but with everyone!

In any case I am very excited about this opportunity, and I will let you know how I fare!

Why are scientists so cautious?

An article from the New York Times writer Gina Kolata from June 27th, is getting a fair bit of buzz, both around the blogosphere and among my friends on facebook (many of whom are scientists as well).
http://www.nytimes.com/2009/06/28/health/research/28cancer.html

The gist of it is as follows. The current trend among large scientific research grant agencies is to fund projects that "play it safe". That is, they will not be risky projects, with respect to having significant "productive" output in terms of research articles out the other side (i.e. funds = research articles). Those proposals that may be the most ground breaking (both in terms of basic research, as well as any potential for significant clinical advances) are also often the most risky. This article does a good job getting at the heart of both the political and cultural components of this issue.

However it got me thinking about the culture of science in general, and our mentorship process. In particular about how a major part of the training of scientists with respect to critical reasoning, also leads perhaps to excessive skepticism. Is this possible? Now, I tend to be an overly skeptical person, and like most scientists, and I often look for the flaws in all of the experiments I perform. However, is it possible that we take it to far as a scientific community?

From my own training experience, I know that some of the most valuable time spent was in "Journal club", where a group of students, post-docs and faculty would get together each week over coffee, and argue about a couple of recent papers. However, in most situations, this would turn in to a session to find every possible flaw in the study. While there is certainly value in this (knowing a good experiment from a bad one for instance), I am know wondering if this does not lead to a culture of scientists who are unable to take risks, or appreciate proposals for "risky" science?

This is something I will have to mull over.....


http://www.nytimes.com/2009/06/28/health/28cancerside.html