incorygible
Posts: 374
Joined: Feb. 2006
|
Dave, while you're parsing the sequence similarities between chimps, humans and gorillas, I have this nagging fear that once again you're going to miss the point. Those similarities are interesting, but they aren't as relevant as this is going to be. (I nevertheless eagerly await your response.)
In the meantime, please pay attention to this post. It’s going to be long, but I’m really going to try to meet you halfway. It is often said by some overzealous "evolutionists" that Creationism makes no testable predictions. While this is often true (“goddidit” predicts nothing), it is by no means universal: there are many places where Creationists say “goddiditthisway”. We’re going to talk about one of those. The age of the earth is another great example yet to come, but we’re going to talk about the relationship between humans and (other) apes. We’re going to assume that your theory (I’ll bite the bullet and avoid the scare-quotes) is, as you have claimed many times, “just as good” as ours. We’re going to use our respective theories to make predictions. You game?
A few notes before we begin: When I make predictions on your behalf regarding Creation theory, I will disregard age of the earth, resulting rates of mutation, etc., and assume only the following (correct me if I’m wrong on either): (1) God originally created a human kind and an ape kind, the latter of which includes gorillas and chimpanzees; and (2) DNA is a valuable tool for examining and comparing exactly how God designed his creations. Are you okay with those? I will use parentheses to denote phylogenies, with H=humans, C=chimpanzees, G=gorillas. For example, (H(CG)) represents a phylogeny where chimps and gorillas are most similar and humans are an outgroup, whereas (G(HC)) represents a phylogeny where humans and chimps are most similar and gorillas are an outgroup. Finally, note that when we talk about frequencies of predicted phylogenies below, these are the percentages of sequences for which two species are predicted to be more closely related than the third. These percentages are not the same as actual sequence similarity. In other words, don’t get confused with the percentages below and the percentage sequence similarities in my earlier post – they’re related, in that the percentages we’re talking about here reflect how often chimps are more similar to gorillas, etc., but they are not the same thing.
All good? Away we go.
Let’s assume it is 1985, and you and I are in a coffee shop having a congenial scientific discussion about the new-fangled genetic technology that is just being developed (and won’t really come into its own for another 10-20 years or so). We’ve been over the same old ground many times about your Creation theory and my theory of evolution, including why you distrust dating methods, why you distrust the fossil record, etc. These are accepted areas of disagreement. Today (1985), we’re going to use our theories to predict what genetics will reveal about the relationships between humans, chimpanzees and gorillas.
Specifically, we’re interested in novel mutations. We both believe these are random changes in the genome. I think they are responsible (along with natural selection and a host of other mechanisms) for the diversity of life on earth, whereas you think they reflect degeneration of God’s Creation since the Fall. This disagreement in views won’t matter. Since we only have the back of the envelope, we’re going to simplify mutation as completely random changes in any sequence of DNA that occur at the same rate in each of our three species. We’re going to assume that the rate at which these random novel mutations accumulate is dependent only upon time, but we’re going to keep time relative (so as to avoid that whole millions vs. thousands of years problem).
We start with a few null hypotheses that neither of us believes. We believe genetics will reveal some sort of phylogenetic relationship (as opposed to none, or a purely random relationship). For example, from the evolutionary perspective, if humans, chimps and gorillas were unrelated, or if they diverged from a common ancestor at the exact same time, I might predict that when we look at their genomes, 1/3 of my predicted phylogenies would be (H(CG)), 1/3 would be (C(HG)), and 1/3 would be (G(CH)). However, the fossil record gives me good reason not to believe the null hypothesis (which doesn’t mean we don’t check it! . Similarly, from a Creationist perspective, if humans, chimps and gorillas were created as separate kinds, you might predict the same 1/3 for each phylogeny. However, you believe chimps and gorillas were created as part of a single “ape” kind, and even if they weren’t, you might predict “common design” to create the appearance of relationships that would refute the null hypothesis.
So I start with my Theory of Evolution prediction, based on what we know of the fossil record in 1985 (the timelines have changed a bit since then).
Predicted initial conditions: Humans, chimps and gorillas shared a common ancestor as recent as approximately 8 million years ago. From that LCA (8 mya), the gorillas diverged from the line that would eventually become both humans and chimps. Humans and chimps themselves diverged about 5 million years ago.
Predicted genetic relationships: If we assume random, time-based mutations occurring independently in each line, then we can expect that each of the three phylogenies may be produced, depending on the sequence we are looking at. For example, if a novel mutation in a given sequence occurs independently in the human line, than phylogenies based on that sequence will group chimps and gorillas: (H(CG)). If the mutation occurs in the gorilla line, the sequence will group humans and chimps (G(CH)). However, we should be able to roughly estimate the frequencies at which these predicted phylogenies will occur, based on the ancestry pattern found in the fossil record and the relative timeframes for each lineage to mutate.
As in our null hypotheses, if they all diverged from the LCA at the same time, we would predict a 33% occurrence of each "tree". However, I believe they diverged in the manner and times above. Chimps and humans shared a lineage for 3 million of the 8 million total years, and this would tend to increase the frequency of (G(HC)) phylogenies by an amount we can estimate. I therefore predict the following frequency of phylogenies:
(G(HC)) = 39% (from independent mutations in the gorilla line: 0.5*(3/8)+0.33*(5/8)) + 19% (from accumulation of mutations in the shared human-chimp line: 0.5*(3/8) = 58%
(C(HG)) = 21% (from independent mutations in the chimp line: 0.33*(5/8))
(H(CG)) = 21% (from independent mutations in the human line: 0.33*(5/8))
So I predict 58% of the sequences we look at will group humans and chimps as closer to each other than to gorillas, 21% will group humans and gorillas as closer to each other than to chimps, and 21% will group chimps and gorillas as closer to each other than to humans.
You then counter with Creationist Theory.
Initial conditions: the human kind and the ape kind were separately created, and never shared a common ancestor. Already we’re in trouble, because we have no information on the genome of those two ancestral kinds. We have reason to suspect they were similar (common design, like Escorts and Tauri in 1985), but we don’t know how similar. We can’t do the same kind of relative calculations that I did by assuming one common ancestor (which do not require knowledge of its actual genome, just that it was shared). However, we do know that any differences between these two ancestral kinds should inflate the frequency of (H(CG)) phylogenies predicted. So right from the initial conditions, you predict that, when we look at a lot of genes to get overall frequencies, the predicted frequency of the relationship (H(CG)) will be greater than 33%.
Creationist Prediction: We don’t have any information on when (relative to initial Creation – actual years don’t matter for this) chimpanzees and gorillas diverged via “microevolution” (changes within a Created kind). However, we know it was some time since the Fall. Without relative time-spans like I had, we can’t do similar estimates like I did, but we can predict that the shared ancestry of chimps and gorillas prior to divergence will increase the frequencies of (H(CG)) even further (as it did for the (G(HC)) phylogenies in my example).
So you end up predicting that more than (far more than?) 33% of sequences we look at will group chimps and gorillas as closer to each other than to humans, less than 33% of sequences will group humans and gorillas as closer to each other than to chimps, and less than 33% of sequences will group humans and chimps as closer to each other than to gorillas.
So, armed with our predictions, we meet back up in a bar 20 years later to discuss the results. I bring along some papers from the prolific new genetics literature. Specifically, I show you the following:
Satta, Y., J. Klein, and N. Takahata. 2000. DNA archives and our nearest relative: the trichotomy problem revisited. Mol. Phyl. Evol. 14:259–275.
Chen, F.-C., and W.-H. Li. 2001. Genomic divergence between humans and other hominoids and the effective population size of the common ancestor of humans and chimpanzees. Am. J. Hum. Genet. 68:444–456.
O’hUigin, C., Y. Satta, N. Takahata, and J. Klein. 2002. Contribution of homoplasy and of ancestral polymorphism to the evolution of genes in anthropoid primates. Mol. Biol.
Evol. 19:1501–1513.
Kitano et al. 2004. Human-Specific Amino Acid Changes Found in 103 Protein-Coding Genes. Mol. Biol. Evol.:936-944.
Combined, these studies examined hundreds of sequences for their predicted phylogenies. Each one found that, on average, approximately 60% of these sequences predicted the (G(HC)) tree (i.e., humans and chimps closer to each other than to gorillas), and the remaining 40% predicted the remaining two trees in roughly equal frequencies (i.e., humans and gorillas closer to each other than to chimps, and chimps and gorillas closer to each other than to humans). (You can look this up if you don’t believe me Dave – I’m more than halfway here.)
I order you a double scotch (you’re gonna need it! as we pull out the faded napkin and look at our predictions.
If you’re still with me, here’s the pop quiz:
What did Creation theory predict?
What did the ToE predict?
What did we actually see?
|
