New at Reason

Julian Sanchez | November 11, 2005

Ron Bailey explains why we may never have to live in fear of sneezing chickens again.

Comments

you know who|11.11.05 @ 1:08PM|#

So science doesn't want to predict how long the mutations will take to develop some functionality (in this case human-to-human trans capability). This despite the fact that these are organisms we can observe right here on Earth now. So agnostic. So honest. So vulnerable. I love it!

Yet in the Intelligent Design threads, we say that this kind of uncertainty regarding time required for a given set of mutations casts no doubt upon evolutionary theory, even though: (1) the mutations were bigger (at least in the aggregate); and (2) they were not nearly as observable (bcs they happened a long time ago).

How to reconcile? How to reconcile?

|11.11.05 @ 1:14PM|#

It isn't that hard to reconcile. Inherent in an article concerning public policy is that we aren't talking about geological time frames. Evolutionary theory is to virus mutation as climate is to weather prediction.

|11.11.05 @ 1:16PM|#

I still can't trust a man whose book has a left-handed alpha helix on the cover when every single high school biology teacher in the nation makes a point out of DNA helices being right handed.

Larry A|11.11.05 @ 1:18PM|#

But if we cure AIDS, where does that leave the God-hates-gays-that's-their-punishment theory?

If God is omnipotent, can He change His mind?

|11.11.05 @ 1:26PM|#

Reconcile? I suspect you're misunderstanding something about how this stuff works.

Let's say you and I went to Las Vegas and we got our hands on a bucketful of dice. Each morning, we dump out the dice and record how many of them come up sixes (apparently we're incapable of finding anything more interesting to do in Vegas). Given a long enough series, it is virtually certain that eventually, one of those bucket-dumpings will result in *all* of the dice coming up sixes. However, it could be the first time, or it could be the twelfth, or it could be the twenty millionth, or it could come far, far later than that.

If you demand that I tell you what month - or even what year - all of those dice will come up sixes, I won't be able to do it - not in advance, anyway.

Science is in the same boat when it comes to predicting what is essentially a random process. The bird flu is even more problematic, because without knowing which changes need to happen to become human-transmissable (is that a word?), we can't even start calculating the probability of it happening. It's also possible that it'll *never* happen - perhaps another strain, without any ability to infect humans, will become dominant in the next decade, and the one that's scary to us will die out naturally. We just don't know. There's no way to predict beforehand exactly how an organism will mutate.

We can tell when the bucket *has* rolled all sixes (as with evolution), but not when the bucket *will* roll all sixes. Does that help clear things up a bit?

Note: I'm not a scientist, so take what I say with a large grain of salt. I'm just fascinated by evolutionary biology.

you know who|11.11.05 @ 1:38PM|#

We can tell when the bucket *has* rolled all sixes

The question with ID isn't whether the mutations happened within the time frame. Presumably they did. The question is whether these were random and whether the timeframe is consistent with the level of random walking that would be expected to be required. If you can't make some good predictions about small mutations (eg, virus characteristics), then I don't see where you get additional certainty because the mutations are more drastic and compounded (eg, mutating virus DNA into chimp DNA).

I mean literally I have seen people present some numbers on how long it would take an eyebal to develop random-mutationally. Sounds like we need those ppl on the bird flu project -- stat!

|11.11.05 @ 1:46PM|#

you know who:

The famous eyeball evolved on a computer model with a starting point of a photosensitive layer of skin and constant evolutionary pressure (i.e. sight was always beneficial) over geologic time. Time was given in a number of generations.

|11.11.05 @ 2:00PM|#

All the time, money and effort we spend on West Nile, AIDS, the Bird Flu and other potential plagues are distracting us and taking valuable resources away from combating the real threat to American citizens. Sharks.

|11.11.05 @ 2:06PM|#

If you can't make some good predictions about small mutations (eg, virus characteristics), then I don't see where you get additional certainty because the mutations are more drastic and compounded (eg, mutating virus DNA into chimp DNA).

The scientists are perfectly capable of predicting how often something will mutate. What they cannot predict is how it will mutate. That's why they can't predict when the virus will, or if it will, become more virulent.

|11.11.05 @ 2:08PM|#

Dave W.-

People can make good estimates of how many mutations will occur over a given number of generations. What they can't predict is when a particular mutation will occur.

Viruses will mutate, and the mutations that are beneficial to viral propagation will be preserved. That we know. But we don't know exactly which mutations are beneficial to viral propagation. Biology is not yet to the point where somebody can look at a gene sequence and predict exactly what sort of organism will result.

And not every mutation that's beneficial to viral propagation is also conducive to an epidemic. Right now, avian flu might be acquiring a mutation that will make it more easily transmissable from one chicken to another, but coincidentally make it even less transmissable to humans. Or it might be acquiring a mutation that makes it more easily transmissable to humans but also, coincidentally, less deadly in humans. (Not every contagious virus is deadly to humans, otherwise everybody with a cold would die.) Then again, maybe a mutation will occur that will make it both easily transmissable to and between humans, and also deadly. We simply don't know.

So, to recap:
1) We know the average rate at which mutations occur.
2) We know that mutations which confer a survival/proliferation advantage will be preserved by natural selection.
3) We don't know when a particular mutation will occur, just that over a given time frame a number of mutations will occur and some will be beneficial while many others are harmful, and still more are actually benign.

|11.11.05 @ 2:18PM|#

Anyone know whether a cell can be infected by one virus when it's already been infected by another virus?

Actually, let me be more specific: anyone know the short answer to that question--one suitable for a management consultant who thinks in bullet points?

|11.11.05 @ 3:12PM|#

No, Scape, not sharks. Snakes. On a plane.

Dave W.|11.11.05 @ 3:49PM|#

"What they can't predict is when a particular mutation will occur."

Which is exactly why all the eyeball models mean nothing as far as ID goes. Don't get me wrong. I don't think the eyeball proves God or anything. I just have trouble getting into the eyeball because we really have no idea how long it should be expected that a random mutation will occur: and that applies to both eyebals and whatever kind of balls virii have.

t. rev|11.11.05 @ 4:11PM|#

JMoore: Yes.

One small objection to the article: it isn't clear that the relative lethality of the virus in birds correlates strongly with the relative lethality of the same virus in humans. In the long run, the deadliest viruses are evolutionary non-starters (e.g. ebola, which seems to keep species-hopping into humans and then burning out before it can spread), but major epidemics are short-term phenomena.

t. rev|11.11.05 @ 4:16PM|#

JMoore: In fact, according to http://gsbs.utmb.edu/microbook/ch062.htm, the human genome contains copies of somewhere between a hundred and a thousand retroviruses. That reference may not fit in a bullet point, though.

|11.11.05 @ 4:25PM|#

t. rev

Thanks much!

Beautifully answered, by the way. If I hadn't retired, I would hire you immediately.

|11.11.05 @ 6:58PM|#

On this issue, I'll take Darwin over god in a New York minute.

|11.12.05 @ 2:28PM|#

Dave W. says:

Which is exactly why all the eyeball models mean nothing as far as ID goes. Don't get me wrong. I don't think the eyeball proves God or anything. I just have trouble getting into the eyeball because we really have no idea how long it should be expected that a random mutation will occur: and that applies to both eyebals and whatever kind of balls virii have.

I think you're failing to appreciate the time scales involved. Consider this analogy I've just stolen from someone else:

If years were miles, your whole life wouldn't even get you halfway from New York to Boston. Evolution, on the other hand, has traveled from the Earth to the Sun and back... about a dozen times.

|11.12.05 @ 5:32PM|#

Ron:

I think many people are crying wolf when it comes to the avian virus," Peter Palese, chairman of microbiology at Mount Sinai School of Medicine in New York, told the Chicago Tribune.

A good part of the hyperbole comes from folks who see an opportunity to get their hands on government money. Kinda like construction companies raising alarm about our "crumbling infrastructure".

Since 1967 the number of American vaccine manufacturers has dropped from 26 to just 4 today. The problem is that vaccines are low?profit margin products sold for a few bucks per dose, yet potentially they expose manufacturers to hundreds of millions of dollars in liability.

There are lotsa products that are cheap per unit and expose the manufacture to liability risk. What harms the drug/bio tech industry and engenders less profit are the years of expensive regulatory hurdles that have to be cleared prior to the product being brought to market and benefiting folks. These regulations also discourage new entrants into the market and benefit the better-healed concerns at the expense of the others and the consumers.


But the development of novel technologies like DNA vaccines point to the really good news.

And should a bird flu pandemic break out, the usual regulatory requirements for safety testing would have to be set aside in an emergency.

Exactly my point. When exciting products like DNA vaccines becomes available, we have to hope that the onset of an emergency will pressure the government to finally suspend these rules that are supposed to exist for our safety so that we may use the product and actually be safer.

|11.12.05 @ 5:45PM|#

I think that the idea of Intelligent Design is even more assailable when we consider results as opposed to process. I've raised the following objection to the ID idea before and have never had a strong answer from an ID proponent:

If it was actually design, it certainly couldn't be called intelligent. There are eye conditions, retinopathies, where blood vessels in the retina leak. Sometimes, in various parts of our anatomy, the body limits the damage caused from leaking blood vessels and saves organs and functions by growing new blood vessels. But with some retinopathies, the new blood vessels grow on the surface of the retina as well, causing blindness.

It's easy to see the mechanisms of adaptation and selection (evolution) at work here. It's rather harder to see design, unless the designer was thought to be malicious or not too intelligent.

|11.12.05 @ 6:35PM|#

Let's say you and I went to Las Vegas and we got our hands on a bucketful of dice. Each morning, we dump out the dice and record how many of them come up sixes (apparently we're incapable of finding anything more interesting to do in Vegas). Given a long enough series, it is virtually certain that eventually, one of those bucket-dumpings will result in *all* of the dice coming up sixes.


I guess that depends on how big of a bucket you are talking about.

If your bucket holds more than 22 dice, you could toss the bucket every second since the big bang and would expect to never see them come up all heads.