way for natural selection to slow down the rate of genomic change without seriously risking extinction. Reducing reproductive output comes at a cost and there are only so many offspring a population can lose before it collapses (ReMine 2005). Second, selection only works on alleles that have a significant effect on fitness, and most mutations are expected to be neutral or nearly so. This subject was thoroughly reviewed by Sanford (2014), so it will not be discussed further here. The important thing to understand is that, if most mutations are selectively neutral, it is impossible to create a significant separation between the short-term mutation rate and the long-term accumulation of mutations. There is also a strong interaction between the nuclear and mitochondrial genomes. After discussing the joint mitochondrial-nuclear genotype, and after claiming that the mitochondrial-nuclear interaction is the unit of selection, Dowling et al. (2008) ask, “How much sequence variation in mtDNA is necessary to produce a phenotypic effect?” This number is not yet known, but the mitochondrial genome is not an independent entity. Nuclear effects further complicate the selective arena. Third, by invoking selection, all mitochondrial molecular clocks could suddenly be wrong. The timing of Mitochondrial Eve is partially based on the assumption of neutrality. If selective mutations occur, the rate of change could speed up (positive selection) or slow down (purifying selection) among the various lineages. Consider that humans live in radically different environments than the supposed source environment (i.e., the forests and plains of equatorial Africa) and there is little reason to believe there are no selective forces at play. Yet, those selective forces must be equivalent among all subpopulations, at all times, and in all places or the molecular clock hypothesis fails. Similar things could be said for the timing of Y Chromosome Adam. Thus, the data from Ding et al. (2021) might be pointing us in a very interesting direction. Fourth, some skeptics are arguing the wrong thing. Phylogenetic change is the rate at which species diverge. Yet, the mutations we are seeing in the mtDNA and Y chromosomes is the rate at which individuals within the population are diverging from each other. There is essentially no species-level divergence occurring within modern humans. The population size is too large and we have gone through a thousands-of-years-long exponential growth phase. These factors have prevented the fixation of any new mutations, so there has been no net change within the nuclear or mitochondrial genomes for quite some time. Even so, evolutionary estimates of the substitution rate within the human mitochondrial genome are on the order of one substitution every 2,400 (Rieux et al. 2014) to 3,500 (Soares et al. 2009) years. These dates are constrained by the accuracy of radiometric dating techniques, which are not part of this study, but see Carter (2022a). And yet, the estimated substitution rate for the individuals in the study of Rieux et al. (2014) were much more consistent than the substitution rates determined for the internal branches (e.g., dated demographic events) of a phylogenetic tree that included those individuals. There is a time-dependency to the data, so clearly one cannot assume a constant mutation or fixation rate through all human history. Population Genetics And yet, the substitution rate is also often misunderstood. The mathematics of population genetics tells us that the short-term, measurable mutation rate should approximate the phylogenetic rate. According to standard population genetics theory, the fixation rate (also called the substitution rate) of new neutral mutations should be directly proportional to the base mutation rate (µ). This can be found throughout the literature (e.g., Kimura 1983) and is part of any course in population genetics (e.g., the classic population genetics textbook of Hartl and Clark, 1997). The number of new mutations per generation is simply 2Nµ, where N is the population size and µ is the base mutation rate per individual. The “2” is included for diploid species (e.g., humans). For neutral mutations not affected by selection, the fixation rate is inversely proportional to its frequency in the population, or 1/(2N). Again, the “2” is included for diploid species. Multiplying these yields 2Nµ/(2N) = µ. Yet, the rate of substitution also depends on population structure. The time to fixation for new, neutral mutations is approximately 4Ne for diploid systems (e.g., nuclear variants) or 2Ne for haploid systems (e.g., mitochondrial variants). Here, “Ne” refers to the effective population size. The amount of subdivision within a population will affect the overall rate of change. Strongly divided populations behave as if they were smaller, so they have a lower “effective” population size. Yet, this can safely be ignored, as the effect is less than the orders-of-magnitude difference between the genealogical and the phylogenetic mutation rates. Also, the following calculations assume a well-mixed population with random mating (e.g., no population structure), so the modeled population sizes are equivalent to Ne. Thus, the substitution rate of neutral mutations in the population is equivalent to the initial mutation rate, by necessity. In other words, since most mutations are selectively neutral, genealogical mutation rates should be directly applicable to estimates of the timing of Y Chromosome Adam and Mitochondrial Eve, as well as any autosomal genes. The fact that Y chromosomes and mtDNA are haploid does not matter. The discussion so far has dealt with theoretical, neutral mutations only, so the question boils down to the rate of non-neutral mutations and the ability of selection to remove them from the population. This can be modeled. One purpose of this study was to provide such a model. Soares et al. (2009) examined this in detail, but from a purely phylogenetic perspective. After estimating the human-chimp split time, they applied a simple metric (number of mutations/evolutionary time) to generate a phylogenetic mutation rate, assuming in the process that humans and chimps did share a common ancestor and that this ancestor lived millions of years ago. They reasoned that newer mutations sit near the tips of the tree and older mutations are the ones shared by many sub branches. By examining the mutation spectra of ‘old’ vs ‘new’ mutations, they discovered that there were fewer missense (e.g., amino-acid-changing) mutations in the former. They reasoned that natural selection had removed many deleterious mutations and so the short-term genealogical mutation rate must be adjusted downward when estimating the timing of ancient genetic events. The rate they calculated was solely based on a presumed common ancestor with chimpanzees at least 6.5 MA. By making this assumption, they were able to estimate the split time for all Old-World monCARTER Genealogical vs. phylogenetic mutation rates 2023 ICC 171
RkJQdWJsaXNoZXIy MTM4ODY=