A lot of rather heating commentary is coming on on the subject of Michael Crichton's gene-patent article, and on gene patents in general. The subject is large enough that it'll need to be broken down to discuss. For today, here's my take on one aspect, what a patent lawyer would call "composition of matter".
The patenting of isolated genes as chemical entities is tricky. Yes, they are chemicals, and when they're isolated and purified like that they really are in a different state than found in nature. But their size is so far removed from many of the other things patented as substances that I can't help but wonder if a principle is being pushed too far. (The obvious other example here is the patenting of isolated proteins, which of course is also well established, for better or worse).
An analogy occurs to me, and working through it will show some of the complications of this area: suppose I isolated and purified a single molecular weight form (one particular isomer) of some long industrial polymer that's usually made and used as a mixture. Can I patent that? Can I then go after people who sell the mixture, because it includes my proprietary substance?
Now, there are some differences here compared to patenting a gene, because the original polymers I'm thinking of are man-made, and there's a lot of prior art around them. And no doubt some of it includes language that covers polymers of a range of molecular weights and the like, and the older ones have long since entered the public domain anyway. The biggest problem with using this as a path to riches is that I don't think I can turn around and go after people whose polymers have my patented isomer in them, because I believe I'd have to show that it's an essential part of their system (and it probably won't be). So I likely won't be able sneak in on some of that big polyethylene money this way.
How about polymers that aren't man-made, like cotton or silk? We have no good way (at present) to produce or isolate individual single isomers of such things, the way we can with stretches of RNA or DNA. If I invent one, I'll most certainly apply for a patent on the method of doing that, just like someone who invents a new way to separate or purify DNA would. I don't think anyone should have a problem with that, because that would be an inventive step by anyone's definition. But can I then turn around and get composition-of-matter patents on some of the things I can isolate with my new technique? Judging from the genomics examples, I'd say that I could, if I could pass a further test.
That's a big one, though: having to show some utility for them. As I mentioned yesterday, that issue that came up with a lot of the early gene patent applications - back in the far-off days of the 1990s, people just immediately shotgunned the PTO with applications for every gene they came across, often with only the haziest uses in mind. Eventually the rules were tightened up - you can't just march in with your gene now and say "could be useful for a diagnostic test for a disease in which this gene is involved" and get a good reception. (There's also the problem that most of the genetic landscape is already the subject of one application or another by now)! I could have some difficulty showing a particular utility for a particular isomer of (say) a silk protein, but it could probably be done. Perhaps the presence of a particular one would prove to be important for imparting some property to the finished silk, for example.
There would be other patenting difficulties, even if I got mine issued. A big one would be the "doctrine of equivalents", which is the patent law way of saying that a difference that makes no difference is no difference. If I claim a newly isolated pure polysaccharide of X hundred or thousand monomer units, is there anything different about it compared to the X+1 isomer? There had better be a difference at some point if I want to have a patent that will do me any good, and to be on the safe side I'd better try to patent everything out to that point.
DNA, RNA, and proteins are perfectly suited to pass that test, though, since very small changes can be demonstrated to lead to totally different properties and functions. The doctrine of equivalents comes in when you start looking at silent mutations - a base change that doesn't change the amino acid that gets coded for, or (in a protein) a conservative amino acid switch in a part of the structure that doesn't affect anything. Court cases have been fought over just these sorts of issues.
As you can see, the thing that makes DNA, RNA, and proteins different is that they're structurally simple enough to make, handle, and isolate, and structurally complex enough so that there are huge numbers of potential variations. There are also highly evolved systems than can be exploited for their production and alteration, which gives everyone a big head start. Biologically, they're leveraged tremendously, so that seemingly trivial changes can sometimes have huge consequences - and, of course, these consequences bear on human health, which makes them of great social and financial importance. A better recipe for intellectual property wrangling I could hardly imagine. Next time, we talk utility, where even more fun is to be found.