Thursday, June 07, 2007

Fitness Landscapes: Behe really, really shouldnt have gone there

As I mentioned before, I have to deal with fitness landscapes every day. HIV fitness, relative to the fitness landscapes of their host and the human population, is specifically what I research.

But Behes ideas of fitness landscapes are so incredibly, bafflingly wrong, I started to second guess my own understanding of fitness landscapes. "He cant be this silly. He cant be this oblivious. Maybe I dont really understand this as well as I thought I did. Oh shit. What if I just made an ass of myself in front of PandasThumb and SciBlogs and all their readers? Oh holy shit."

So I spent a little time on PubMed. Youll be pleased to hear that Im not insane. Not only that, but Im learning lots of cool new things, seeing how other biologists are using fitness landscapes.

**evil grin**

If Behe wasnt so fixated on Darwins Black Box, he might have noticed he was opening a Pandoras Box for ID Creationism. Sure on the surface, Pandoras Fitness Landscape Box might look inviting to IDC... a shiny theoretic mathematic veneer, useful for blinding people. But a demon waits for ID Creationists. Empirical research.

I found this oh-so-handy review in Nature:
Empirical Fitness Landscapes Reveal Accessible Evolutionary Paths

Behe seems pleased with himself over the 'probability' of chloroquine-resistance in malaria (wrong about that too), an alteration of two amino acids.

'Empirical Fitness' gives an example of researchers empirically determining probably fitness pathways of bacterial resistance to cefotaxime, and it will blow Behes mind.


Researchers just used a plug-and-chug method to figure out what kind of pathway bacteria could take to acquire resistance. 32 possible mutants, 120 possible pathways. They tested them all.

Most of the pathways made Behe Evolution Barricades-- dips in fitness keeping the organism from reaching a near-by higher peak.

They found 18 possible pathways that would work just fine. No drops in fitness as the mutations started to pile up. And as the review authors admit, this is extremely conservative. When you take other dimensions of the fitness landscape into account, and the possibility of mutations outside of the 5 building a bridge across a fitness valley then subsequently reverting (epistasis!), there must be many, many more possible trajectories.

Five mutations.

Not two.

This paper also addresses a favorite of ID Creationists in the context of fitness landscapes: ligands and their receptors (biochemistry, what Behe should know). Theyre like locks and keys-- one is no good without the other. And not just that-- How do you make sure the lock is specific for the key, and the key is specific for the lock? You cant have excess keys around, opening locks willy-nilly, or locks that take any key. How could such a thing evolve without God? I mean Designer! How could such a thing evolve without a Designer?

Empirically studying fitness landscapes gives us some awesome examples of how new super-specific ligand-receptors can evolve! How about our hormone receptors? Hormone: key, Hormone receptor, lock. Turns out one of our hormones acts like a master key: 11-deoxycorticosterone. It works on 'ancestral' receptors and modern receptors, which means duplicated receptors (AAAAARG gene duplication! Another IDC bane!) were free to mutate around a bit, and new, more specific keys could evolve, like aldosterone! Then the aldosterone key and aldosterone lock were free to co-evolve with each other and live happily ever after! YAAAAY!

The review calls this process 'molecular exploitation.'

This is only a four page review, but it has some more great examples and describes fundamental aspects of 'empirical fitness landscapes'-- and its a Nature paper! So Im certain you can find this at a local library, or a librarian there can find you a copy. Its worth the hunt if this spike in Behe posts has piqued your interest!


Blake Stacey said...

Thanks for the link!

Anybody who knows a little about the brain can appreciate that the "lock and key" model has its limits. Neurotransmitters provide an excellent example of multiple keys fitting in the same chemical lock, with differing degrees of effectiveness. Morphine bonds to the ยต-opioid receptor, which is normally activated by the enkephalins naturally present in the brain. The morphine molecule binds more tightly to the receptor than the normal, endogenous opioid molecules. It's a key which fits better into the lock than the standard keys do. (Heroin does the same job even better.)

The drug naloxone binds to the opioid receptors even better than heroin — it's a key which fits even better in the lock — but the key doesn't turn! It binds to the receptor but doesn't trigger a response in the postsynaptic neuron. A couple minutes after injection, naloxone can force a patient overdosing on opiates to go into withdrawal.

386sx said...

This is both very silly and utterly brilliant. Very nicely done.

ERV said...

Blake-- As usual, reality is much much cooler than Creationist fairy-tales and stressed analogies. Man I miss brain stuff-- If I didnt go into virology, I was going to go into neuroscience.

386sx-- While the silliness is always intentional, any semblance of brilliance is purely accidental. Id put the odds of a brilliant post at 1 in 10^20.


Rev. BigDumbChimp said...

Love the blog. Good to see someone taking them to task with plenty of snarky goodness to boot.

Anonymous said...

"Not only that, but Im learning lots of cool new things, seeing how other biologists are using fitness landscapes."

In a earlier post, you said "I hate Creationists. They are absolutely worthless to me. I can learn nothing from them ...". As this sentence shows, you were wrong. You can learn from them, even if it is indirectly.

Steven Laskoske said...

To Anonymous:
It wasn't the Creationists she was learning from. It was the scientists. I supposed that after coming across mind-numbing creationism might inspire her to go find some real learning. However, that was not learned from the creationist but in spite of the creationist.