Discovery Institute Jumping the Shark

January 9, 2014 at 4:28 pm (Discovery Institute)

Creationist Casey Luskin, in another attempt to disprove evolution , misinterprets yet another paper.

Now, a new study in BMC Genomics, “Characterization of the heart transcriptome of the white shark (Carcharodon carcharias),” has found more genomic evidence that contradicts the standard vertebrate phylogeny.

No this is not correct; the paper does not show more genomic evidence that contradicts the standard vertebrate phylogeny; instead the paper simply shows that the proportion of proteins in certain biochemical categories, is more similar to human than to zebrafish – indicating that certain biochemical/physiological functions in white shark might be more similar to human than zebrafish – possibly, for example, endothermy.


Permalink Leave a Comment

Casey Luskin’s Pharyngula

July 3, 2013 at 4:27 pm (Discovery Institute)

Creationist Casey Luskin in his post Vertebrate Gene Expression and Other Properties Don’t Support a “Phylotypic” Stage writes:

However, a new article in PLoS Genetics, “The Hourglass and the Early Conservation Models — Co-Existing Patterns of Developmental Constraints in Vertebrates,” shows that even an analysis of the genome based on Darwinian assumptions fails to confirm many predictions of the “phylotypic” stage. The implication is that, as other papers have more explicitly suggested, the phylotypic stage may not clearly exist.

This is not true. The original model makes no predictions on gene age or gene sequence evolutionary rate. So I’d rather say that they are closer to the original model than the reports which they fail to confirm.

Permalink Leave a Comment

Folding On Levinthal’s Paradox

December 30, 2012 at 3:02 pm (Discovery Institute)

This piece by Paul Nelson predictably distorts the actual take home message of this paper.

Basically, the (original) Levinthal paradox related to protein folding was not a paradox, it was a calculation that showed the immense dimensions of folding space. Levinthal simply concluded the conformational space is so vast that a protein cannot fold by random conformational search. That is, there is no time to visit a significant part of the folding space to search out the best energetic solution – what we call the global minimum. Rather – he suggested – there must be mechanisms encoded in the protein sequence, i.e. the protein follows some path(s) from an unfolded to a folded state. The necessity of this conclusion can be stressed by calling the apparent contradiction between the number of conformations and the time it actually takes for a protein to fold, a “paradox”.

This paper makes two points. For one, nobody took notice of the analogy of protein folding and the assembly of the interactome. Their calculations only show that the numbers are even bigger here. It does not really matter how big they are – even 10^7200 is unimaginable, thus it serves the purpose.

The conclusion that there are pathyways of assembly, is trivial, everybody thinks it this way. Their “paradox” only puts emphasis on the magnitude of the problem.

The second, more circumstantial and more serious conclusion they draw is that the analogy is only virtual: whereas protein folding leads to the global energy minimum, the interactome is not in an energy minimum, it is in a “steady state”, and requires continuous energy to maintain. They purport it cannot form from its components, i.e. the interactome cannot “fold” if taken apart. It could only form once during evolution, and its assembled state contains the very information needed for its assembly. It is a rather philosophical conclusion, and has many consequences, far less trivial than folding mechanisms and assembly pathways.

Permalink Leave a Comment

Casey Luskin and the Frogamander, a Cautionary Tale

January 23, 2011 at 9:20 am (Discovery Institute)

With all the hoopla over at the DI about homoplasy I thought this post would be relevant. Gerobatrachus is one of my favorite fossil finds. Here’s an interesting post about it at Discovery Institute’s blog here . Here’s some of my own (hopefully constructive) criticism of this article.

Casey Luskin claims that

” ~75% of the character data conflict with the phylogenetic hierarchy in their tree.”

That is correct, but it can’t be taken as a measure of quality of their data set. The CI ( consistency index ) is strongly negatively correlated with the number of taxa and characters, the more of either the lower it gets. Furthermore, the original matrix this one was based on specifically maximized the number of characters that might conflict because it’s better not to rule things out beforehand, which further lowers the CI. You cannot compare the CI of different analyses based on different matrices as a result. Also, the ankle seems to go into a piece of matrix that remains in place. I’m fairly confident that we have all of the bones that were ossified in this area in this individual.

Permalink 1 Comment