3 Seminars, 1 afternoon
Nov 17th 2004Wallabynerd alert
Three seminars in one day, after an aimless morning. And then we had decent cider and average donuts at the department mixer. Caramel apples, too, although I’ve never been partial to those.
And then my mens’ group did a little party for our “sister” women’s group. With miniature quiche. From a box of frozen ones. And similar “impressive” treats. (Don’t tell them they were frozen, though.)
Noon: A mechanical engineer talked about how to describe engineered tendons in mathematical terms. Most of the model came from fluid dynamics and other physics-type stuff, instead of chemistry and biology; It predicts some stuff that makes sense, but it seems like a very naive model. Cross-disciplinary approaches and stuff are great, but hand-waving that ignores decades of work in biomechanics and biochemistry doesn’t sound like the best way to go about it. Might be a good starting point for a “real” model though. He suggested that, within 5-10 years, they’d be able to model tendons well enough to predict how they’d grow back after a particular surgery, for instance.
4:00: Ion mobility spectrometry and differential mobility spectrometry, for detection of chemical warfare agents. (Student seminar, so he just summarized a couple papers with some background on the techniques.) Ion mobility spectrometry is a very simple idea – if you attract ions with a strong electric field in one direction, and at the same time you blow air on them to push them in the opposite direction, then you can separate them by size and shape. The hardware can be made quite small and light weight, and it doesn’t take much power — perfect for a portable instrument. It’s also quite fast – less than a second from start to finish. The military uses them a lot; so do airports (when they swab down your purse or backpack with a cotton patch and stick the patch into a big box – that’s an ion mobility spectrometer, looking for drugs or solvents from explosives).
I hadn’t heard of DMS much before; it’s very similar in concept to IMS, but instead of the drift gas flowing counter to the electric field, it flows in the same direction (towards the detector). The electric field is constant, but the drift tube has a pair of electrodes on either side of it, with the ions traveling between them (much like a quadropole mass spec, but with only two poles instead of four). The charge on these electrodes is swept (square wave); if the charge is “wrong” (too high) for a particular ion, then it’ll get pulled to its doom and hit the plate, never reaching the detector. (Same thing if it’s too low.) So, it winds up acting as a filter that only lets ions of a certain mobility (size/charge) through to reach the detector. Changing the charge on the filter electrodes provides a “spectrum” of ions in the sample. Advantage: Can see both positive and negative ions (IMS only sees one charge at a time); potentially more sensitive. Disadvantage: The examples the seminar-giver showed had HORRIBLE resolution; peaks were very broad. (I wonder if this could be improved by either slowing down the drift gas flow, or reducing the size of the detector, or what.)
Second student seminar was on new fluorescence labeling techniques for cell biology. Very much biochemistry, and she really didn’t give quite enough background on that side of things, or perhaps it was just scattered around too much. I barely followed it and I was a biology major in undergrad. (Then again, that WAS 6 years ago now.) Pretty pictures, at least; that’s my usual reaction to fluorescence labeling stuff in general. She talked about yet another variant of FRET, and the addition of a solvatochromic dye label to a protein to serve as its own “sensor” (for some loose definition of “sensor”). The second technique was a lot more interesting; one attaches an appropriate dye to the protein, and its fluorescence output changes depending on its local environment. So, if the dye folds into the oily middle part of the protein (the example system was calmodulin binding to Ca++), then the change in fluorescence makes a very clear “tag” to show Ca++ binding, or whatever. Again, lots of pretty pictures. (I suspect that’s really the motivation for most microscopic techniques these days — pretty pictures are popular covers for Nature.)
