BPS Day 3: Monday, must be Paris

Finally found 1) a seat, 2) an outlet, and 3) an IP address (the Moscone center wifi seems to be pretty overloaded. You'd think a famous important science blogger would be able to get an IP address, but I guess the hardware doesn't distinguish.)

Some thoughts on last night's single molecule workshop:

• The workshop was chaired by Steve Quake, and the first talk went to one of his colleagues who basically used the time to demonstrate the commercially available sequencing machine that Steve's company, Helicos, has on the market. He talked about the science behind it, but it wasn't really a "workshop" insofar as he didn't demonstrate single molecule techniques that you could do in your lab. It was a promotional session for a self-contained commercial product for sequencing. I think this was in pretty poor taste. Aside from that, he spoke very highly of their read lengths, topping all of 35 bases. Steve Block very pointedly asked what he thought the upper limit on their read lenghts were, and the speaker started waxing about the biochemical limitations. When Steve pressed him for a number, he said, I think, 100 base pairs (though I might have misheard, and he may have said 500. But if he said 500, I don't believe him anyway.) The reason that Steve pressed him, of course, is the fact that Pacific Biosciences is now boasting an instrument with several hundred base pair read length, which suggests that the Helicos instrument has a very limited future. Of course, Pac Bio doesn't have an instrument on the market yet, and Helicos does, and there are a lot of advantages to being first to market. So who knows.
• Zev Bryant's talk was interesting. He described double headed mutants of myosin that could be made to switch directions at will, and described processive motors made out of myosin heads that had basically protein logs between them. I'm surprised that you get processivitiy out of such things, but I guess if the ATP is low enough, the heads spend most of their time bound to the actin, so they're naturally processive. I'd like to see optically switchable myosins, and I'm sure he's working on it. Genetically encodable optical switches seem to be a growing theme, I'm seeing a lot of it all over the place here.
• Adam Cohen's talk was a bit silly, I think. He used glass lenses on top of coverslips to create highly confined spaces, 1 nm up, between the glass and the lens, and looked at proteins diffusing. Things undergo confined quasi-2D diffusion, so they stay in the field of view longer without being tethered. But most people are trying to develop assays that get molecules further from the surface, to minimize surface effects. Here, the surface effects are huge, and this is sold as a benefit. The technique didn't solve any particular problem, it didn't present any new science, and it presents a host of new problems. On the other hand, at least it was fun, and fun is good.