Nathaniel Jeanson’s Null Hypothesis

I’ve been investigating ICR researcher Nathaniel Jeanson’s recent paper in Answers Research Journal, “Recent, Functionally Diverse Origin for Mitochondrial Genes from ~2700 Metazoan Species.” As it’s a topic I’ve seen before I’m going to write about it, but I see this morning that Hemant Mehta at his Friendly Atheist blog has actually beaten me to it, writing “A Creationist’s Desperate Attempt to Sound Like a Credible Scientist.” However Mehta’s post mostly mocks Jeanson, going so far as to dig up a promotional video the ICR made about him, and doesn’t really address his arguments in any detail. All the more for me then.

In October of 2012 there was an article in the ICR’s Acts & Facts magazine in which an exited Jeanson explained a new hypothesis he had come up with to explain why different animal “kinds” had different DNA sequences. Animals tend to share the same collection of genes in their mitochondria, and there is no reason on the surface why these core genes should have different sequences – yet they do. Could it be, Jeanson argued, that all species started off with identical genes that since the creation have mutated randomly away from this common direction? He hypothesised that species with a faster generation time would accumulate more mutations and therefore be more diverged from species that lived longer. He had already compared three species – an elephant, a mouse, and a fruit fly – and the short lived fly was indeed much more different from the relatively long lived mouse and elephant.

Of course, as I pointed out at the time, this hypothesis was doomed. If he simply added a turtle into the mix he would find that even though it had a long lifespan it would be less closely related to the Elephant than the mouse was, just as evolution would predict. Indeed, as soon as November Jeanson had figured out his mistake: his neutral hypothesis was disproved.

Since then he fell off the radar a bit, at least so far as I noticed, but it seems that he was quite busy working on this paper. He is still concerned that:

The young-earth creation model currently lacks a robust explanation for molecular diversity. No comprehensive method exists by which absolute or relative sequence differences among species can be predicted, and no method has been formulated to rigorously predict the function of molecular residues, especially those in so-called “house-keeping” proteins.

But:

The evolutionary model is so robust that it leads to predictions of molecular function. Under the assumptions of this model, species will grow more and more distant molecularly over time, unless some natural force constrains random variation.

While:

Unlike the evolutionary model, the creation model lacks a clear, predictive explanation for molecular diversity. Because Scripture is silent on the identity of the sequences that God created during the Creation week, a number of competing explanations still exist.

There are a few other problems as well: it seems that genetic diversity within fruitflies, which creationists like Jeanson would presumably class as a single kind, is greater than that between humans and chimps, which they insist are unrelated. I don’t think he manages to tackle this issue at all, but concentrates on the other.

With his favourite hypothesis ruined by an uncooperative reality Jeanson casts around for an alternative. Unfortunately, he doesn’t have the resources to do the tests that he wants, and instead tries a third option:

A third method for testing these hypotheses is nuanced and somewhat counterintuitive, but powerful. Rather than test each hypothesis directly, a strict null hypothesis could be constructed and then refuted. This would necessarily imply that one of the alternatives to the null must be true. For example, elimination of the null might point towards created diversity as the likely candidate. Hence, by process of elimination, the real explanation might be discovered.

In short, Jeanson’s previously explained neutral hypothesis is turned into a null hypothesis, which once refuted means that something else must be true. In short, nothing that we didn’t already know.

Jeanson clearly spent a lot of time comparing the sequence of multiple mitochondrial genes of around 2700 animals, apparently preforming tens of millions of comparisons over the last year. The result is a series of figures that look like this, one for each gene:

Jeanson's figure 2

Fig. 2. Kingdom-wide violations of the null hypothesis with ATP6 alignments.”

In this figure the colour at any given point represents the degree of sequence similarity between the species at that point on the y and x axis – red means similar, blue dissimilar, and white in the middle. The whole graph is mirrored along the top-left to bottom-right axis as this is where each species is compared to itself. Red/white squares (you see them as rectangles as the figure seems to be a bit distorted – it was apparently made in Excel) along this axis come from groups of species which are consistently more closely related to each other than to outsiders. As you can see from Jeanson’s helpful annotation, vertebrates are more closely related to each other than to non-vertebrates – the red/white square in the middle, which is probably so large only because more vertebrates have had their mitochondria sequenced than anything else – while within that group mammals make up a subgroup to which there even more subdivisions. Needless to say, patterns like this completely refute Jeanson’s null, as there should be no reason why any group larger than a kind should clump together.

But it’s also the exactly the hierarchical pattern we would expect from evolution, and indeed I’m torn between being happy that even the creationists are producing results in support of the opposing view and just rolling my eyes and pointing out that this is what I said would happen a year ago and that the whole exercise was pointless.

With the null disproved the solution must lie elsewhere. The correct solution is of course evolution, but this is a creationist publication so that’s not going to be seriously considered. Instead, Jeanson posits that his god, for reasons unknown but presumed to be related to function, created groups of kinds which show up when you do comparisons. In addition, the null hypothesis assumed that natural selection was not in operation, so it’s likely that, you know, it had some kind of effect in the real world.

(1) the null; (2) evolution; (3) what Jeanson wants you to believe

To summarise a bit more graphically then, the neutral, or null hypothesis (1) posits that everything radiated randomly and independently from a single common point. This fails because it would not produce the pattern observed in the figure above. Evolution (2) also begins at a point but it’s lineages split at various points, producing the hierarchy we see. In the model that Jeanson now is forced to go with (3) those groups were built in by the YEC god, but don’t mean that the kinds are actually related to each other. All the evidence that suggests that they are must merely be the result of some unknown divine plan – it couldn’t possibly be, say, a personal message from an omniscient god to Jeanson pointing out to him that he’s on completely the wrong track.

Needless to say, this isn’t a comfortable position for Jeanson. Indeed, it’s probably worse than the one he started in. Jeanson has half-buried the tree of life, and expects us to believe that only the part left above ground actually exists. But his paper is a bit of a two-for-one deal, and so unable to achieve victory in the part we’ve already seen Jeanson tries to recover in the other.

While before he was trying to compare kinds to each other, in this portion of the paper he instead investigates variation within kinds. Because of a lack of data he only looks at the nematode C. elegans, the fruit fly, the water flea of the genus Daphnia, and humans. We’ll concentrate on the latter for now.

Calculating time of origin from diversityBy scouring the literature for information about the observed mutation rate in human mitochondria, along with the observed amount of living variation, Jeanson calculates the time since the last common ancestor. Surprisingly, he comes out with a result that is, he says, consistent with the YEC timescale but not with evolution or an old Earth. I should point out here that Jeanson is using the round figure of 10,000 years and not 6000 for the age of the young Earth. He presents his results like so:

Figure 16

Fig. 16. Recent origin of mitochondrial genome sequences.

Looking at part D we see that he has calculated that the variation observed within humans is similar to what would be created in 10,000 years but not in the 180,000 year history of Homo sapiens. Even if true this is not, however, a problem.

Mitochondria is generally inherited from the mother, and so you can exclude men entirely from your analysis. Two women can have the same mother, but one women cannot get her mitochondria from two different mothers. In other words, if you trace mitochondrial lineages backwards in time they can only merge, never split – there is no hybridisation.

Mitochondrial descentGoing back far enough you will by necessity eventually arrive at a single woman, who is the maternal great great grandmother (times however much) of everyone alive. But that doesn’t mean that she is the only woman living at the time, nor the only woman who has living descendants, and certainly not the first woman. Indeed other now extinct lineages (e.g. in blue above) will have existed at that time, with earlier origins.

As you may have already guessed, all Jeanson has actually done is calculate that the person who is commonly referred to as “Mitochondrial Eve” is much younger than generally believed. This is not in and of itself a problem for evolution, as she does not represent the first human woman and so need not be as old as the species, though I suspect that a number of methodological flaws in Jeanson’s reasoning is the cause of his low date.

But not only has he failed to disprove evolution his calculations are not actually consistent with a 10,000 year old young Earth either. As you may recall, a paper in Nature recently investigated the mitochondrial DNA of a 400,000 year old Sima de los Huesos skeleton, which like the Neanderthals and the Denisovans would most likely be classed by YECs as being within the human kind. Here’s figure 4 of that paper:

Figure 4: Bayesian phylogenetic tree of hominin mitochondrial relationships based on the Sima de los Huesos mtDNA sequence determined using the inclusive filtering criteria.

Living humans make up the lower right area of that diagram, and Jeanson believes that all the change in that region can be fit into 10,000 years. That’s all very well but if so then all of the other diversity, involving the extinct groups that we now have DNA for, cannot fit into the YEC timescale. That’s a real problem.

As for the other “kinds” he examined it would similarly appear that he hasn’t sampled all living and historical variation – indeed, he could not have done so. This argument just doesn’t work.

Jeanson’s paper is extraordinarily long and excruciatingly detailed in explaining every last tiny step. But don’t let that fool you into thinking that his conclusions are anything other than a shoddy mess. Mehta may have been a little harsh, but Jeanson really does need to go back to the drawing board. He may also wish to remember the old maxim: quality not quantity. There’s far, far too much of the latter here.

Advertisements

9 thoughts on “Nathaniel Jeanson’s Null Hypothesis

    • He didn’t do much with it save for calculating the number of “nucleotide differences” that 180,000 years would produce relative to 10,000, and figuring that the actual figure was much closer to the latter. There’s also a table that shows “Years elapsed until a single mutation occurs” – apparently to create the observed diversity in 180,000 years you’d need to wait 36,364 years for a mutation to come along, but under the creationists 10,000 year model you’d still need 2,020. Both of those numbers seem absurdly high.

      The paper is very long and hard to navigate (there’s not table of contents, and the figures are not distributed with the text that discusses them) – what were you looking for specifically? I’ll see if I can find it.

  1. Excellent overview and analysis here. Jeanson sure spends a lot of time to show nothing. I haven’t looked at that paper yet but I’m guessing he doesn’t really deal with homoplasy in mtDNA genes which make comparisons between divergent groups sorta pointless. Once a whole mtDNA genome accumulates 2500 to 3000 total differences (out of 16,000) it bascially isn’t going to become anymore divergent because there will be so many backmutations. To compare single gene sequences across all metazoans just does’t makes sense. Of course he doesn’t think there has so much time so as to reach this equilibrium point.

    • He does note somewhere that you can’t get a sequence similarity below 25%, but I don’t think it figures much into his calculations.

  2. Nice job, Peter. It’s now clear that ICR has added another 25W bulb to its pack of luminaries.

  3. Nathaniel Jeanson said …

    “The young-earth creation model currently lacks a robust explanation for molecular diversity.”

    As far as I can tell the YEC “model” currently lacks a robust explanation for anything at all. Picking at individual data points trying to find a “something-anything” that differs from mainstream science is not a “model”.

  4. Got to say nice job here again. That you could actually read this article and make any sense of it at all is quite an accomplishment. I really could not figure out how he compile some of his charts. I am still struck by figure 16 which you reproduced above. If I am reading this right he is saying the evolutionary theory predicts that Daphnis should exhibit 120,000 changes if it has been around as long as evolutionary theory expects. Uh, the mtDNA genome is only 16,000 bp in size does he really think that a 16,000 bp genome can show 120,000 differences? Obviously there are other factors involved but that is a ridiculous graphic to show. Just frustrated with this nonsense that Jeanson and Tomkins are dishing out lately and then Ham refers to the great work in hi journal as providing details of how animals have evolved since the Flood.

  5. I feel like the whole paper was done to be as confusing as possible – in this case, what he should have done is taken the observed diversity and mutation rate and used that to calculate the amount of time since the ancestor.

    I’m considering writing a second post on this sometime, concentrating on the second part. Jeanson has put a write-up in the latest Acts & Facts and is promising more to come.

Thoughts?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s