One question that comes up frequently about evolutionary biology is whether it really boils down to speculation and assumption. Most of evolution happened in the distant past, after all. We claim that humans and chimpanzees descended from a single ancestral species over millions of years, for example, but none of us was there to observe that process. To a scientist, though, the right question is not, “Were you there?” but rather “What if?” What if we do share a common ancestor–what should we see? How can we test a hypothesis about the ancient past?
One way we can test for shared ancestry with chimpanzees is to look at the genetic differences between the two species. If shared ancestry is true, these differences result from lots of mutations that have accumulated in the two lineages over millions of years. That means they should look like mutations. On the other hand, if humans and chimpanzees appeared by special creation, we would not expect their genetic differences to bear the distinctive signature of descent from a common ancestor.
What do mutations look like, then? DNA consists of a long string of four chemical bases, which we usually call A, C, G and T (for adenine, cytosine, guanine, and thymine). A mutation is any change to that string. In the simplest mutations, one base replaces another when DNA is incorrectly copied or repaired, e.g., a C at a particular site in a chromosome is replaced by a T, which is then passed onto offspring. These substitutions do not all happen at the same rate; some occur more often than others. For example, C and T are chemically similar to one another, as are A and G, and chemically similar bases are more likely to be mistaken for one another when DNA is being copied. Thus, we find an A becoming a G more often than a T.
This means that as they accumulate, mutations create a characteristic pattern of more and less common changes. It is that pattern that we can look for to see if genetic differences were caused by mutations. To determine exactly what the pattern is, we can just look at genetic differences between individual humans, because these represent mutations that occurred since those two people last shared a common ancestor.1 Twelve possible substitutions can occur (A→C, A→G, A→T, C→G, C→T, C→A, etc.), but if we are only looking at differences between two copies of DNA, we cannot distinguish some of the substitutions. For example, if I have an A and you have a C at a specific location, unless we have our ancestors’ DNA to look at, we cannot tell whether it was originally an A that mutated into a C in your DNA, or whether it was originally a C that mutated into an A in my DNA. Thus we have to lump the two possibilities together and just count the number of places one of us has an A and the other a C. An additional complication: our DNA has two complementary strands, and we do not know which strand a mutation occurred on. Perhaps it was not actually our ancestor’s A that turned into your C. On the other strand of his DNA, the A was matched by a T (the complementary base to A); perhaps that was the base that actually mutated. The bottom line is that there turn out to be four distinguishable classes of substitution: (1) What we call “transitions” occur between the chemically similar bases A and G, or C and T; these changes happen more often than the others. (2) A difference between A and T (which I will label A↔T). (3) A difference between G and C (G↔C). (4) A difference between either A and C or G and T (A↔C /G↔T).
Now we are in a position to test whether genetic differences between humans and chimps look like mutations. To determine the pattern for mutations, I calculated the rates for the four classes using human diversity data (which is available online). Then I calculated the pattern seen when comparing human and chimpanzee DNA, also using public data. The first graph is the distribution for humans. As expected, transitions are the most common. That pattern is our signature–the sign that mutation has been at work.
The second graph is the same distribution for differences between human and chimpanzee DNA. The overall rates are different–there are 12 times as many differences between human and chimpanzee DNA as there are between DNA from two humans (note the different scale on the y-axis of the graphs)–but the pattern is almost identical.
Remember my opening question: if humans and chimpanzees shared a common ancestor, what should we see? What we should see is what we do see: genetic differences between the species that look exactly like they were produced by mutations. In scientific terms, I had a hypothesis about the distant past, I tested the hypothesis with data, and it passed the test.
Now, when scientists point to similarities between human and chimpanzee DNA, critics sometimes object that similarities don’t really prove anything, since they could be explained equally well by a common design plan: the creator might well use similar stretches of DNA to carry out similar tasks in separately created species. That objection does not apply here, though, because we are looking at the differences between species. I cannot think of any reason why a designer should choose to make the differences look exactly like they were the result of lots of mutations. The obvious conclusion is that things are what they seem: humans and chimpanzees differ genetically in just this pattern because they have diverged from a single common ancestor.
We can make the same comparison for other pairs of species, all of which are thought to have common ancestry. Here is the breakdown for the differences between humans and some other primates, including apes and Old World monkeys, and some nonhuman comparisons as well. (In order to display the results on a single chart, I have rescaled the other distributions to have the same total rate as the human-human comparison.)
Everywhere we look, the pattern is the same. That’s true even though the overall rate of genetic difference ranges from less than 1% (human vs chimpanzee) to more than 5% (humans vs baboons). The genetic differences between species always look like mutations.
I also took a look at some species that are less similar to humans—mostly out of curiosity, since I was not sure exactly what to expect. Mutation patterns vary subtly even between human populations, probably because of small differences in some of the hundreds of proteins responsible for DNA replication and repair; such variations are likely to become more pronounced as we look at more distantly related species. One thing I did expect, though, was still to see more transitions than other substitutions since that difference is rooted in the basic chemical similarity of some bases. The set shown here includes cats compared to dogs, cows compared to dolphins, a comparison between a couple of species of finch, and even two species of pufferfish.
There is one additional test we can make. When I made the plots above, I excluded a small part (approximately 1%) fraction of DNA because it is known to mutate much faster than the rest. The higher mutation rate occurs when a C is immediately followed by a G in the DNA sequence, a pairing known as a “CpG” (“p” stands for the phosphate group that links adjacent DNA bases). A wide range of animal species chemically modify the C when it occurs in a CpG. This has an interesting effect: modified C can spontaneously turn into a T. As a result, mutation is much more common at CpGs than for other DNA, especially for C mutating into T.
We can therefore define a more comprehensive signature of mutation by measuring the rates for the same categories as before, but now at CpG sites. (This adds three new categories rather than four, since A↔T cannot occur at a CpG site.) This signature is shown in the first plot, which once again comes from human diversity data. The second figure shows the same categories for human compared to chimpanzee DNA. Once again, the two line up almost perfectly. Even at these special sites, differences between species look exactly like they were caused by mutations.
This kind of thing is the reason that most geneticists have no doubt about common descent: it makes sense of everything we see. Even better, it makes predictions. When I started to put together this post, the only data I had seen was for humans and chimpanzees, but I still had a very good idea what I would see when I looked at other primates.
Of course, none of this says anything at all about God’s role in human origins, nor does it rule out miraculous intervention. But it does provide strong evidence that we share ancestry with other species.
So What Is BioLogos?
Well it all began with a scientist and a book. Francis Collins, the physician and geneticist who led the Human Genome Project, wrote the book, The Language of God. In it he describes his own journey from atheism to Christian faith, and the harmony between Christianity and science.
Today, BioLogos continues to carry out the vision of Collins, showing that you don’t have to choose between modern science and biblical faith.