0G and DNA repair

We all know, or could have expected, that DNA gets damaged more often in space. Cosmic rays, solar wind--it is hard to shield against these things. In fact, shielding can be counter-productive, as cosmic ray showers generate more particles than you had to start with.

That will exercise our DNA repair systems. They have some low failure rates--do they work the same up there?

A Queen's University study says that polymerases don't work quite as well in microgravity. They designed a system to carry in "the vomit comet" to make 20-second long tests of copying DNA at 0-G, and compared the error rates with runs taken without the 0-G.

If that result sounds weird to you, it does to me too.

But: convection in 0-G is different; you have to rely on diffusion. Flames look very strange and burn differently, with somewhat different combustion products.

But at this scale, why would convection matter in a mere 20 seconds?

They had some problems: "We were forced to invest much effort into improving the user-friendliness of our mini-laboratory, to make it easier to operate not only in microgravity, but also in the subsequent 2G hypergravity phase of the flight once a zero-gravity parabola has been completed," Since the experiment involved manipulating pipets, yes, some delicacy is needed.

I'd forgotten about the high-G part of the flight. The reaction should have been quenched before that stage began--it sounds like they did it right--but I wonder if vibrations and the initial acceleration (about 1.5 G) could have done some minor damage to the equipment. Their control sample was done during level flight, so take-off would have been the same. They had three 0.2ml reaction mixture tubes, and 80microliters of each of the various ingredients. That's not very big, but depending on the geometry convection might be an issue. It wouldn't be at the nuclear level, though.

Different amino acids had different transcription error rates--significantly--even for a "we only got one measurement at 0G" experiment. And there seemed to be some effect--for some of the comparisons. Their statistics are low and I don't know if their error bars are realistic.

Curious. But one run isn't quite enough.