Gould, earthworms and you:

February 19, 2001 9:38 AM

Gould, earthworms and you: Stephen Jay Gould discusses the recent discovery that the human body has only about 1/4th of the DNA originally estimated. NYTimes op-ed piece. One of the best results of this discovery is that it sounds death knell of reductionist biology; as usual, the human body turns out to be more complicated than anyone could have imagined. ("Gee, we haven't explained life, the universe and everything? Gosh darnit!") I have always thought it was silly to ascribe artistic talent, criminal behaviour, musical aptitude or computer savvy to the foibles of some single gene. Now here's independent confirmation of that opinion...

So once again we find that we ourselves, and not our parents or our grandparents, are responsible for who we are and what we become...
posted by hanseugene (14 comments total)

As Kass said to me: given that programmers favour a smaller, more elegant codebase over a crufty one, shouldn't we be praising the apparent compactness of the genomic algorithm? (Even with the apparent cruft of "junk" DNA...)
posted by holgate at 10:00 AM on February 19, 2001

Gould's comment on our dependence on the bacteria accidentally traveling through us indicates, I think, that saying that we ourselves are responsibe for who we are is not the right way to put it. In fact, I guess that it's the opposite: The genes don't explain us, so we have to be a product of dozens of contingent factors: sensory experience, diet, chance reactions during development, socialization.

At any rate, referring to ourselves when discussing how we come to be is a case of begging the question.
posted by argybarg at 10:07 AM on February 19, 2001

Define "ourselves" as the sum of the circumstances and environments over which we can excercise control. Certainly we don't decide how, when and where we will be born and raised. We do, however, eventually decide what we are going to make of these circumstances. We choose what we put in our mouths, we choose our friends, we choose to a great extent how we will react to the events taking place around us, etc. At some point we have to make these choices; they are very much what make us human. Chemical determinism doesn't explain who we are any more than finding the region of the brain that controls religious experience explains religious experience itself.
posted by hanseugene at 10:18 AM on February 19, 2001

With any luck, the we ... are responsible for who we are argument isn't going to be used by some groups to change other groups to conform to some arbitrary "normalcy". For instance, the christian right trying to "cure" homosexuals.
posted by dithered at 11:04 AM on February 19, 2001

"I am responsible for what I am" is, on the level of therapeutic generality, a truism about adulthood.

But as a deeper and more precise explanation for how we become humans and how the organism acheives its complexity -- it's hopeless. Where does the "I" come from?

My point is that Gould's argument sucks the hope out of the argument that we are self-creating. If our genes merely set the process in motion rather than program each step, then it is largely our lives, as we experience them in the world, that create us.
posted by argybarg at 11:14 AM on February 19, 2001

I don't think this decides anything. A factor of 4 is neither here nor there in this kind of thing. There is no way we know enough to say OK, if there had been this number of genes then this, but with a quarter then this. A factor of a hundred or a thousand, perhaps, but four?!

My guess is that bits of both will be right - sometimes when Gould gets upset over some quite obviously daft reductionist claims he procceeds to dismiss all of nature. Nature and nurture, as always, are both important...
posted by andrew cooke at 11:27 AM on February 19, 2001

Holgate, while it means that the code which is used is perhaps more elegant than we thought, it also means that an even higher proportion (something near 95%) of the total code we carry is dead code, which isn't exactly elegant.

More to the point, it's known that even the live code contains dead segments. When a gene is transcribed to RNA for use, some enzymes go in and clip out sections. Those are known as "introns" and it's suspected that they were introduced by retroviruses.

In other words, the code has to be patched as it's read from storage (DNA) into working memory (RNA) before it can be run (making proteins). That is extremely inelegant, to say the least.
posted by Steven Den Beste at 12:05 PM on February 19, 2001

Cheers, Andrew, I unintentionally mimicked your last sentence in another thread.
We're coming closer to understanding our biology, but I don't think we're very close to grasping the whole picture once and for all . People are too predisposed to classify things as either black or white- "pick your side and start swinging"- because we hate to admit that we don't really know it all.
posted by xtrmntr at 12:12 PM on February 19, 2001

Steven: everything I've read about the genetic system sounds like it's the work of a thoroughly mad old-style "Real Programmer", the sort of person who believes that unexpected side effects were designed to be used, and predictability be damned. This is the sort of machine language that inspires awe, respect, and fear - but little admiration and even less imitation.

I eagerly await the day when the folding problem has been solved and genetics becomes the domain of programmers. Ahhh, then we will have fun.

-Mars
posted by Mars Saxman at 12:27 PM on February 19, 2001

any copy around for those of us who feel we're highly evolved enough not to have to register for sites to be able to read their free content?
posted by DiplomaticImmunity at 12:33 PM on February 19, 2001

the fact that the creation of various protiens requires bits and pieces of DNA from all over the chromosome to create RNA doesn't necessarily imply ineffieciency; it has yet to be shown that a given intron for a protein can't be reused to make another protein. Or that the exact shape of an intron isn't immutable; what my be cruft for generating a particular RNA is used in the creation of other RNA. Rather like building a command with xargs using a case statement.

SJG's comments apply, I believe, very neatly to this article that appeared today in the Washington Post, about the sociobiologist's quandries explaining homosexual behavior. The sociobiologists, while not explicitly trying to pin a behavior on any single codon yet, are struggling to apply their reductionist model onto sexual expression, and are finding they have to expand their suppositions to include the idea that environmental factors play a profound role in the selection of possible behaviors.
posted by katchomko at 3:01 PM on February 19, 2001

Mars, you're right. No-one would study this crap to learn how to program -- except for one thing. Anyone trying to write new code for this machine has to make it work within the mechanisms already in place. So stuff like introns and exons doesn't appear optional in eukaryotes. Synthetic genes will have to work even when edited by the transcription mechanism. That should be fun.

I've tried enhancing spaghetti code, but at least I had the benefit of cryptic inline comments and incomplete and obsolete external documentation. The poor bastards trying to figure out the genetic code have to work it all out on their own. I wish them luck.
posted by Steven Den Beste at 8:31 PM on February 19, 2001

First of all, I'm getting a little sick of the media harping on how the human genome is only twice as big as the fruit fly's, when we're clearly so much more complex. Well, what sort of objective standard are they using to define complexity? Can you fly? I can't.

It's a mistake to believe that our species is at the pinnacle of creation and is somehow more evolved than other species on the planet. Drosophila isn't a lower rung on the evolutionary ladder, it's another endpoint on the evolutionary tree, no more nor no less perfect for its particular biological niche than we are for ours.

The report that there are only 30,000 or so human genes is surprising, but it doesn't require any retooling of biology's central dogma. We already knew that extragenomic, environmental factors play an important role in development--everything from socialization to to chemical environment to mammalian nurse cells. We've also known for a while that single genes can code for multiple proteins (antibodies being the paradigm here). I fail to see how chemical reductionist biology has been disproven.

Besides--Stephen Jay Gould hasn't ever been humbled by anything.
posted by shylock at 12:05 AM on February 20, 2001