We continue our review of the book, Adam and the Genome: Reading Scripture after Genetic Science, by Dennis Venema and Scot McKnight. Today, Chapter 3- Part 1
And now we get to main event. What does genetics say about the human race descending from one and only one pair? Dennis recounts the controversy following the June 3, 2011 Christianity Today article, “The Search for the Historical Adam”, and the resulting NPR interview in which he was quoted saying:
“But now some conservative scholars are saying publicly that they can no longer believe the Genesis account. Asked how likely it is that we all descended from Adam and Eve, Dennis Venema, a biologist at Trinity Western University, replies: ‘That would be against all the genomic evidence that we’ve assembled over the last 20 years, so not likely at all.’”
Did I say controversy? — raging furor would be more like it. Trinity Western administrators fielding angry calls. It is a wonder Dennis didn’t lose his job for no longer believing the Genesis account (and I imagine that a few of you reading this blog post think he should have). We will delve into what it means to “believe the Genesis account” when we get to Scot’s section of the book. For right now, let’s try to stay focused on the science. Dennis was naïve to assume that people understood that evolution was a population-level phenomenon. If humans evolved, then they evolved as a population; he thought everyone knew that. But he was to learn that saying the population genetics data indicates we descend from a group of about 10,000 individuals was more controversial than the data supporting common ancestry that we covered in the last chapter.
The brilliant analogy of Anglo-Saxon changing over time to Modern English shows how speciation occurs. It is the incremental change of average population characteristics after two populations separate. No one expects a new language to start with two speakers suddenly speaking in ways radically different than the population it arose from. Yet many people think this how speciation occurs. They assume all species are founded by an ancestral breeding pair that is suddenly markedly different from the population it arose from. He says:
“Thus, I’ve encountered many folks who, upon understanding the evidence that humans and other apes share common ancestors, assume that humans—like all other species in their thinking—got their start when a founding couple “mutated away” in tandem from their ape-like ancestors. These folks then wonder if the Genesis narratives may be portraying this radical shift, with perhaps God intervening to create the necessary, large-scale mutations that made us a biologically distinct species.”
But that is not how speciation works. The populations are genetically separated in some way, usually physical isolation (although not always) and no longer breed together. When mutations occur they are no longer shared across the divide. Now the two populations begin to drift apart in their average characteristics. Trying to draw a sharp dividing line on this biological gradient is as useless as deciding what day Anglo-Saxon became Modern English. Dennis uses an example from his teaching. He will ask the class to spell “lose”. Usually, a number of students will spell it as “loose” which is becoming more common, especially on social media. So the ability of a language to hold variant spellings or grammatical variants is dependent on the number of speakers in the language. Dying languages have no variation at all since they have so few speakers (like some Native American languages). Modern English, as it has become a world-wide language can support a large number of variants. The same occurs with species; a large population size allows for maintaining a large number of variants, since each member of the species is able to hold up to two distinct variants (alleles) of any given DNA sequence in its genome. Therefore, there is a connection between the number of variants in a population and the size of that population. Scientists can use that connection to estimate the size of the population from the number of variants. And since the rate of change over time is slow, it is straightforward to extrapolate backward from the present to the past.
It is technically possible that a species could be founded by a single breeding pair, just as it is technically possible a new language could be founded by two speakers. However, that would be highly unusual and, with respect to the genetics anyway, there would be a tell-tale mark on the genome that would persist for hundreds of thousands of years—a severe reduction in genetic variability for the species as a whole.
The carnivorous marsupial once roamed over all of Australia, but now only exists on the island of Tasmania. Tasmanian devils have so little genetic variability that for the last hundred or so years they have exactly the same alleles with only rare differences. This indicates a “bottleneck” event occurring in which there was a severe reduction in the population with the resulting loss of genetic variability. The problem for the devils is that there is a form of cancer transmitted by bites that is threatening their complete extinction. Normally, a recipient animal could fight off the cancerous cells but the devils are all so genetically similar that the cancer cells do not trigger an immune response.
In humans, by contrast, there is much genetic variability. That is why there can be such a problem with donor organs; a close match has to be found. For the Tasmanian devils any one could be an organ donor for another (or, sadly, a tumor donor). This example also illustrates how long it takes a population to rebuild its genetic diversity—many thousands of generations. The implications are clear; Tasmanian devils experienced a severe bottleneck in the distant past, humans did not.
There have been some theories that have proposed bottleneck events for humans. The controversial Toba catastrophe theory, presented in the late 1990s to early 2000s, suggested that a bottleneck of the human population occurred about 70,000 years ago, proposing that the human population was reduced to perhaps 10,000–30,000 individuals when the Toba supervolcano in Indonesia erupted and triggered a major environmental change. Parallel bottlenecks were proposed to exist among chimpanzees, gorillas, rhesus macaques, orangutans and tigers. The hypothesis was based on geological evidence of sudden climate change and on coalescence evidence of some genes (including mitochondrial DNA, Y-chromosome DNA and some nuclear genes). However, subsequent research, especially in the 2010s, appeared to refute both the climate argument and the genetic argument. Recent research shows the extent of climate change was much smaller than believed by proponents of the theory. In addition, coalescence times for Y-chromosomal and mitochondrial DNA have been revised to well above 100,000 years since 2011.
Since the question of estimating population size from genetic diversity is the key question that determines the geneticist’s claim humans could not originate from just two people, Dennis takes some time to sketch out the methods used that support that conclusion. One simple way is to select a few genes and see how many alleles of that gene are in the present day population. Taking into account the mutation rate and the mathematical probability of new mutations spreading in a population or being lost, computer calculations indicate an ancestral population of right around 10,000. In fact, to generate the number of alleles we see in the present day from a starting point of just two individuals, the mutation rate would have to be far in excess of what we observe for any animal.
The astute skeptic could point out the assumption is that the mutation rate has remained constant; but what if it was higher in the distant past? However, we have other ways to measure ancestral population sizes that do not depend on mutation frequency. These methods provide an independent check on results using allele diversity. One such method is known as “linkage disequilibrium”. The basic idea is that if two genes are located close to each other on the same chromosome, the alleles present at both locations tend to be inherited together.
(Note: the following discussion is taken pretty much straight from the book– I just couldn’t figure out how to condense it. But I’m not going to block quote it for readability.) In Figure 3-1, the long line represents a chromosome, and the hash marks across it show us where the two genes in question are located. Geneticists even us the Latin word for location (locus, pronounced “low-cuss”) as a synonym for ‘gene’. (Latin makes us sound smarter, I guess). If we could zoom in on the diagram, we would see a long DNA molecule with two regions that are translated into proteins (the two genes). The different alleles at either locus would have slight sequence differences, giving us four possible combinations for these two loci (plural for locus, pronounce “low-sigh”). The four possible combinations are “AB”, “Ab”, “aB”, and “ab”.
During cell divisions that make gametes (i.e. eggs or sperm), there is a process of mixing and matching of alleles to make new combinations. For example, suppose an individual had one chromosome with the A and B alleles and another with the a and b. During gamete formation, it is possible to produce gametes that are “recombinant”—in this case, ones that have either and “Ab” or “aB” combination. Recombination requires a process of precise chromosome breakage and rejoining called “crossing over”—something you might recall from high school biology (Figure 3-2)
The key point to understand is that the closer together two loci are on a chromosome, the less likely it is that a crossover event will happen between them. The further apart two are loci are, the more likely it is that a crossover will recombine them. What this means is that alleles of loci close together tend to be inherited as sets.
Let’s work through an example of how this plays out in practice. Consider an extended family represented by a pedigree. This is the type of diagram geneticists use to trace alleles through large families. Females are represented by circles and males by squares. Connecting horizontal lines indicate parents, vertical lines are offspring. Generations are labeled with Roman numerals and individual are labeled with Arabic numerals. In this way we can refer to any individual in the pedigree (Figure 3-3).
Now consider a large pedigree where we know the allele combinations of everyone represented (Figure 3-4). For example, individual I-4 has one chromosome with the “AB” alleles linked together, and one chromosome with the “ab” as a set. We can represent her chromosome set, then, as “AB/ab” – the shorthand geneticists’ use. We can then use this convention or other individuals in the pedigree. For example, the daughter of I-1 and I-2 might have an “AB/ab” combination. If these two loci are very closely linked together, it is highly unlikely that crossing over will occur. Thus she would have inherited her “aB” set from her mother, and “AB” set from her father.
Likewise, her husband, II-3, would have inherited “Ab” from his dad, and “ab” from his mom. Their children (generation III) similarly would inherit these sets without crossing over. Looking at the combinations carried by these children then allows us to infer things about their ancestors. If these two loci are very close to each other, we might not expect them to recombine over hundreds of generations or more. Thus it’s reasonable to infer that these four combinations come from four distant ancestors.
The trick is that we can now do this for tens of thousands of loci across the whole human genome. As we have sequenced the DNA of more and more individuals in different people groups around the globe, we’ve simply been asking the questions: Based on the number of allele combinations that we observe in this population, how many ancestors do we need to invoke in order to explain what we observe? In this case, rather than estimating mutation frequency, the calculations require knowing how often crossing over happens between two loci. This is also something we can measure directly in humans and other animals, and there is a well-characterized relationship between chromosome distance between two loci and crossing over frequency. We’ve now done this sort of analysis for millions of pairs of loci (you read that right—millions) for each chromosome pair in our genome (all 23 pairs). And what is the final tally after crunching all that data and counting up ancestors. The results indicate that we come from an ancestral population of about 10,000 individuals—the same result we obtained when using allele diversity alone.
• • •
Other posts in the series: