Bio-Origins Project: Hypothesis Disproven

Now it’s my turn to be smug. In last month’s edition of Acts & Facts “Deputy Director for Life Sciences Research” Nathaniel Jeanson announced that he was investigating differential mutation rates as an explanation for the observed differences in sequence in the same gene in different species. His hypothesis was that God had, in effect, a pool of genes to choose from when he created life. All organisms that needed a specific gene would be given the same one, but the particular genes needed by each would vary. These originally identical genes would then diverge through mutations, with Jeanson using lower generation times as proxies for higher mutation rates. His original supporting evidence came in the form of the mitochondrial ATP-6 genes of the elephant, mouse, and fruit fly.

As I pointed out at the time there are a number of flaws in this hypothesis. For one – despite Jeanson’s claims to the contrary – this process would not necessarily create the observed hierarchy in sequence similarity. More importantly, however, the three animals analysed at that point just happened to have their evolutionary relatedness approximately agree with the predictions of Jeanson’s differential mutation rate model. I predicted that the mere insertion of a fourth animal would ruin the correlation, suggesting a turtle as a good test subject.

The ICR has not tested a turtle, but instead has analysed a large number of mammalian ATP-6 genes. Jeanson has written a new article for the November Acts & Facts edition: Bio-Origins Project Update, Evidence Against Differential Mutation Rates.

Last month we showed preliminary evidence suggesting that the molecular patterns we see in animal species may be due to different rates of mutation in each “kind.” Further investigation of the match between mutation rates and genetic differences among species suggests that this initial hypothesis was incorrect.

For all my “I told you so”s, it is important to remember that this admission is a positive thing. The ICR is actually admitting that it got something wrong. The problem, however, is that the flaws were so obvious and they should have seen this coming. The cynic in me is even wondering if the whole thing wasn’t just a set up to be able to say “we admit our mistakes; we do real science too!”

If you didn’t quite get it above, Jeanson lays out his (former) hypothesis in the new article like so:

Our initial hypothesis posited that individual gene sequences were identical in each kind but that the overall genome sequences were different. This hypothesis was consistent with the common design principle of tool re-use—if a tool performs a function well, good engineers re-use it for other applications. Conversely, in the genome, genes may act like tools in the construction (development) of each creature; if so, they might be re-used for the same function in many different creatures. Our hypothesis was also consistent with the common assumption that mitochondrial genes (that we were investigating) were “housekeeping”—they performed the same function in every creature. Hence, there seemed to be no functional reason for designing these gene sequences differently in different kinds.

We also hypothesized that, from this originally created gene sequence identity, modern gene sequence differences arose as a result of different rates of genetic change over time. These differences in rates would eventually produce a hierarchy of differences among modern species.

As I mentioned, even if it were true that all ATP-6 genes diverged from a common sequence in accordance with the mutation rates of the species concerned a hierarchy still would not be produced.

To test our hypothesis, we used a surrogate measure of mutational change, the generation time (the time from conception to sexual maturity) for each species. Since preservation of any mutation in a population depends on successful transfer of the mutation to progeny, the mutation rate for a species is intimately tied to the species’ generation time.

Generation time isn’t a perfect proxy. The mutation rate per generation can also vary, both between species and over time. You may remember that in the Lenski experiment a number of E. coli lineages evolved “mutator” forms by damaging DNA repair systems. When this arose in the citrate-metabolising lineage it allowed the speedy fine-tuning of that process (which originally wasn’t very efficient). The potential for disconnect between generation time and the true mutation rate provides a fair bit of wiggle-room when the evidence goes against the hypothesis, but to his credit Jeanson does not make use of it.

With reference to some figures showing similarity and generation time of a “small subset of [their] comparisons” Jeanson goes on to give examples of counter-evidence to the hypothesis. The TL;DR is that walruses and raccoons ought to be quite dissimilar, but are not. Conversely, baboons and elephants should be similar, but also go against the prediction. Jeanson does not go into the more general patterns of the data, but they too are quite interesting.

He concludes his article:

What might be the real explanation for the genetic patterns among species? Perhaps the differential mutation rate hypothesis applies to only select groups of creatures—perhaps to insects, but not to mammals. Alternatively, the explanation for the genetic patterns might be something entirely different. Perhaps God created different ATP6 sequences in different kinds, after which little diversification happened genetically—this is the subject of our current investigation.

Jeanson was doing quite well up to this point, when it comes to being scientific and honest. But here he blows it – he doesn’t even seem to prepared to consider the possibility that, just maybe, evolution did it. We’ll see what alternative explanation he comes up with when he reports it.