An Apportionment of African Genetic Diversity Based on Mitochondrial, Y Chromosomal, and X Chromosomal Data

In an effort to better understand patterns of genetic variation in modern African populations, I surveyed nucleotide variability at four loci in five diverse sub-Saharan African populations. First, I analyzed the mitochondrial DNA (mtDNA) and the non-recombining portion of the Y chromosome (NRY), asking specifically if similar models of population size change could be fit to re-sequencing data from these two loci when examined in the same populations. Four tests of population growth were employed and results indicated that food-producing populations best fit a model of exponential growth for the mtDNA but not the NRY, and hunter-gathering populations best fit a model of constant population size for both mtDNA and the NRY. These results are likely due to sex-specific migration or differences in the effective population sizes of males and females.Next, I examined mtDNA and NRY population structure in these same populations, to assess the relative effects of migration and effective population size on patterns of mtDNA and NRY nucleotide variability. I used an Isolation with Migration (IM) model to disentangle estimates of effective population size and migration. Results indicated that levels of mtDNA population structure are higher than those of the NRY, and female migration tends to be unidirectional while that of males is largely bidirectional. I found that in food-producing populations, male migration rate estimates are in fact higher, not lower, than those of females, while estimates of male effective population size are strikingly small. I infered that males have experienced a period of population size reduction due to replacement, and that this most likely occurred during the Bantu expansions, approximately 5,000 years ago.Finally, I assessed population structure in these populations using a multilocus approach which estimated current and ancestral effective population sizes, migration rates, split times and fraction of the ancestral population that contributed to current populations. Current and ancestral effective population sizes ranged from ~5,000-8,000 individuals. Most populations showed an increase in size relative to the ancestral population. Population split times ranged from 17-142 thousand years (KYR); the Khoisan split times were the oldest and the Niger-Congo speaking populations' split times the most recent. Since the oldest population split times precede the dates for the earliest modern humans outside of Africa, I posited that modern humans likely evolved at a time when structured populations already existed in Africa