I wanted to write down what Karen was telling me some weeks ago about her bio research, because when I write things down I understand them better (this tells you something about the kind of learner I am). Karen is a very good teacher because she's so into her work that you can't help but have some of that excitement rub off on you.

She was explaining her work on a protein that triggers cell death - and therefore is important to cancer research - and the amount of work they have to do, the methodologies they have to follow to work with living cells, that just blows my mind. That you can order custom genetic sequences - that they can pipe out short strings of bases cheap enough and fast enough to sell genes to researchers that want to buy them - I remember how much trouble I had just getting the DNA out of my cheek cells back in high school science class, and they can just up and make arbitrary sequences consistently enough to sell this as a service? (I am a terrible experimentalist. Careful, reproducible results are not my forte.)

And how she described mutating a plasmid vector was... damn. Plasmids are circular strands of DNA, so when they replicate they "unzip" into two circles of half-DNA. Floating little blocks of base material get attracted to the now-exposed "zipper-half" of these DNA circles until the entire plasmid is filled.

Imagine ripping two long pieces of velcro apart and throwing them into a pool of little pieces of velcro; when you pick the long strand of velcro up again, the fuzzy side will be covered with the wee pieces of velcro that stuck to it. Now imagine the velcro is striped red and green - there'll be a green section of velcro, then a red one, then a green one, then a red one. And imagine that only green velcro can stick to green velcro and only red velcro can stick to red velcro. And that when the little pieces of velcro hit the long strip of velcro, they magically join together into one long strip of velcro themselves, so you end up with two identical pieces of velcro. That's how DNA replicates.

To be more technical about it, there are really four "colors" of velcro - ATCG - and "A-velcro" and "T-velcro" stick only to each other, and "C-velcro" and "G-velcro" stick only to each other, so you really end up with one piece of velcro, and then the same pattern on the second piece of velcro but in different colors... the point is that the same information is contained.

So the way they mutate plasmids is that they take advantage of the time when the circle of plasmid-velcro is floating around the sea of little base-velcro-snippets, so to speak. And what you do is you throw in a section of velcro that has the sequence that you want it to mutate to. (This is the strand of DNA that's "manufactured" and you buy custom-made, like I mentioned earlier) This short section of velcro exactly matches up with the sequence on the plasmid-velcro except for a short piece in the middle, which is the new sequence you want it to mutate into. And this piece of velcro floats around, it finds a plasmid half it matches up with, and the two ends of the mutant-velcro stick to the plasmid, just like in normal DNA replication. The middle part that's mutated still hangs on, because it's sticking on either side since the bases there match the plasmid. And the rest of the little bits and pieces of base-velcro join on, and it closes up the circle and merges into one continuous loop of velcro and voila, you have your new mutated plasmid.

Note: I don't actually know what I'm talking about. I might be completely off about this.

Anyway, the cool thing about this protein that Karen is looking at is that it performs a similar function to some other protein - I can't remember the names of these things - basically there are these two proteins that can hit the same "DIE NOW!" button on a cell - and they're trying to figure out whether the two proteins help each other, compete with each other, or how do they interact when they're pressing that "DIE DIE DIE" button. (By "pressing the die button" I mean the protein locks into a particular chemical receptor on the cell and triggers apoptosis.) So they've got all these experiments running trying to narrow down how these two proteins work. It's subtle stuff; since you've got these two proteins that sort of do the same thing, if one stops working you might not be able to tell - the other one's working so the person won't get sick. But it's very important to know how these proteins work, because it helps you understand cell death, which helps you understand why cells don't die, which is what we call "cancer."

That's all. I just thought that was fantastically beautiful. Thank you, Karen. (And apologies for completely mangling your dinnertime lesson to me. I'm not very good at understanding or explaining any of this biochemical stuff yet.)