One of the things I like most about science is that you really don't know what's going to happen next. That's especially true in the areas where things have just barely settled down. Before that, when a field is new, no one knows what to expect, so in a way there aren't really any surprising results: everything's a surprise. A much more settled area, by contrast, is far less likely to produce surprises, although when one shows up it really stands out. But a field where people are just starting to exhale and think that maybe they've finally figured out what's going on - that has the best combination of high contrast and a real likelihood for craziness.

Here's a perfect example, since I was just expressing some doubts about the immediate commercial potentials of RNA interference the other day. In a paper coming out in PNAS, a group at UCSF was investigating the use of some small double-stranded RNAs, just the sort of thing that can be used for RNAi experiments. But they found (to their great surprise) that their experiments were stimulating the transcription of their targeted genes, rather than shutting them down. Needless to say, this was not what anyone expected, and I'll bet the folks involved repeated these things many, many times before they could trust their own eyes. There are plenty of other people who won't believe it until they've seen it with theirs.

On a molecular biology level, it's hard to say just what's going on. The authors, according to this news item from Science (probably subscriber-only) say that they've found some rules about which genes will be susceptible to the technique and which won't, which will be released soon. (Translation: as soon as they can be reasonably sure that they won't make fools of themselves - this paper took enough nerve as it is).

The Science article includes a good deal of if-this-holds-up language, which is appropriate for such a weird discovery. (Are the editors there wondering why they didn't get a chance to publish the article themselves, or did they have the chance and turn it down?) At any rate, if-it-holds-up this effect will simultaneously complicate the RNAi field a great deal (it was gnarly enough already, thanks) and also open a door to some really unusual experiments. Upregulating genes isn't very easy, and there are no doubt many ideas that have been waiting on a way to do it. There are therapeutic possibilities, too, naturally - but they'll have to wait on the same difficulties as the other RNA therapies.

Anyway, I'm happy to see this. It opens up some completely new biology, and it opens a door to a potential Nobel for the discoverers should everything work out. And it always cheers me up when something totally unexpected flies down like this and lands on the lawn.