Aaron Pfennig, Quantitative Biosciences Thesis Defense

In partial fulfillment of the requirements for the degree of
Doctor of Philosophy in Quantitative Biosciences
in the School of Biological Sciences

Aaron Pfennig

Defends his thesis:
Theoretical and empirical population genetics of admixture and introgression

Monday, June 17th, 2024
1:00 PM
Krone Engineered Biosystems Building (EBB), Children's Healthcare of Atlanta Seminar Room (EBB 1005)

Zoom link: https://gatech.zoom.us/j/92994408118
Meeting ID: 929 9440 8118

Thesis Advisor:
Dr. Joseph Lachance, School of Biological Sciences, Georgia Institute of Technology

Committee Members:
Dr. Annalise Paaby, School of Biological Sciences, Georgia Institute of Technology
Dr. I. King Jordan, School of Biological Sciences, Georgia Institute of Technology
Dr. Patrick McGrath, School of Biological Sciences, Georgia Institute of Technology
Dr. Amy Goldberg, Department of Evolutionary Anthropology, Duke University

Abstract:
Admixture and introgression are ubiquitous in nature and are crucial sources of genetic variation on which natural selection can act. They are also central aspects of human evolution. Given their significance, understanding the evolutionary dynamics at play during admixture and introgression is a prerequisite for comprehending how they have shaped present-day genetic variations and their biomedical implications in an era of precision medicine.
The first part of this thesis reviews human population history through the lens of admixture and introgression with a focus on African populations. Subsequently, commonly used models for estimating magnitudes of sex-biased admixture from autosomal and X chromosomal ancestry proportions are benchmarked using extensive neutral simulations. These models are highly sensitive to misestimated admixture proportions, requiring large sample sizes. Furthermore, they are confounded by violations of demographic assumptions; in particular, gene flow after the initial admixture and sex-specific assortative mating. As human population history is often more complex than assumed in any underlying demographic model, and given the social sensitivity of the topic, care should be taken when interpreting inferred magnitudes of sex-biased admixture with these models, and confidence intervals need to be provided when reporting specific numbers.

The second part of this thesis focuses on the mixing of populations that have been isolated for a relatively long time, i.e., introgression, and its effects on present-day humans. A new population genetic model that accounts for the fitness effects arising from a heterogeneous genetic background in hybrid genomes during introgression shows that it is important to account for the genetic background. Hybrid fitness effects arising from a heterogeneous genetic background, such as heterosis and Dobzhansky–Muller incompatibilities, can significantly hinder or boost the fixation probability of introgressed alleles. As Dobzhansky-Muller incompatibilities are predicted to accumulate faster than heterosis effects, this model predicts that hybrid fitness effects act as a genetic sieve in most cases of introgression.

Lastly, to test this hypothesis explicitly, the evolutionary fate of remaining Neanderthal ancestry in present-day genomes is assessed, leveraging 30,780 admixed genomes with predominantly recent African-like and European-like ancestry from the United States in All of Us. Such admixed genomes allow the direct testing of the evolutionary fate of Neanderthal ancestry, as the fitness of most archaic haplotypes has been re-assessed in an African genetic background during the 15 generations since admixture. The remaining Neanderthal ancestry appears to be effectively neutral on a genome level as the admixed genomes contain as much Neanderthal introgressed sequence as expected. Nevertheless, four and three loci are significantly depleted and enriched for Neanderthal ancestry. These regions overlap with genes that are associated with known phenotypes to be affected by Neanderthal introgression, such as spermatogenesis (FSCB) and innate immunity (SPINK gene family). Furthermore, the admixed genomes contain three novel introgression desert-like regions, i.e., large genome regions (>8 Mb) that are significantly depleted of Neanderthal ancestry. The population genetics of the three novel desert-like regions provide additional evidence for the occurrence of hybrid incompatibilities during human-Neanderthal introgression but do not rule out intrinsic negative selection.