Monday, January 29, 2007

Shh! Fins in the making!

Last week's Nature contains an interesting article on fin development in sharks, rays and other cartilaginous fish. The article both illustrates how science doesn't always get things right the first time but rather sometimes approaches truth through successive attempts. It is also a case that illuminates the difference between the science of evolution and the pseudoscience which purports to compete with it.

The hedgehog (hh) gene was originally identified in Drosophila; the phenotype of the hh mutant larvae suggested a hedgehog, and with the whimsical naming tradition in fly that is the name it was given. When later work searched for hh homologs in vertebrates, a whole family was found, and these were named after various hedgehogs (african hedgehog, indian hedgehog, etc). With surely a mischievous smile, one was named after the video game character Sonic. As fate would have it, Sonic hedgehog (Shh) is the most studied of the bunch.

An important role for Shh is in the patterning of the vertebrate limb. While mapping out this pattern across the limbed vertebrates, a curious thing happened: a previous Nature paper reported that the cartilaginous fishes, who left the vertebrate line first among still living limbed vertebrates, seemed to lack a dependence on Shh.

The new paper reverses the previous finding. The previous paper had looked only for an Shh message, but the new paper uses a full court press on the problem.

First, they sequenced elements resembling Shh appendage-specific regulatory elements (ShAREs) from multiple cartilaginous species -- these are the DNA elements which drive Shh expression in the limbs. Both the sequence and position relative to Shh were found to be conserved. Second, appropriate expression of Shh was shown in fin buds from multiple species. Third, the developmental program in these buds was shown to respond to Shh or retinoic acid (a powerful developmental trigger) in a manner similar to the response observed in bony fish and other vertebrates.

Nature recently raised a firestorm by publishing a pro-creationism letter from a prominent European creationist, which provoked a flurry of letters in response. Some of these letters supported the publication, on the grounds that it denied creationists the claim that they are censored, while others bemoaned Nature besmirching itself with pseudoscience.

The shark/ray fin paper illustrates neatly what makes evolution a science and creationism (or its sheep's clothing sibling, intelligent design) a pseudoscience. There is a lot of evidence from the fossil record and embryology that shark fins are developmental homologs of bony fish fins, bird wings, and our arms & legs. If this work had shown that the molecular program was entirely different, then the world would be turned upside down. A major inconsistency would exist between the classical evolutionary view and the molecular developmental view. While such inconsistencies have generally been resolved with the molecular side shifting more (as recently pointed out by Carl Zimmer), there always exists the possibility that the two will be irreconcilable. In that case, a serious crisis would exist for the modern evolutionary synthesis.

We can contrast this with intelligent design and creationism (hereafter ID/C). ID/C posits an unknown (ID) and/or unknowable (C) designer, who had total freedom in designing species. Therefore, not finding ShAREs or Shh-dependent development of shark fins would be no big deal; it would just be one more design choice. Indeed, there would be no particular reason to expect ShARE's of similar sequence and similar location. In contrast, the evolutionary view predicted the existence of conserved ShAREs and conserved mechanism -- and without it would be in serious trouble. Because ID/C is equally compatible with all possible evidential outcomes, it has zero predictive power -- and therefore zero true explanatory power. That which explains anything equally well actually explains nothing.