The interacting forces of mutation, recombination, natural selection, and genetic drift shape the heritable differences among individuals and species. We are interested in mining the information carried by extant variation to learn about these genetic and evolutionary processes. To this end, work in the group combines some data generation, with large-scale data analysis and modeling. Currently, our main foci are to (i) understand how natural selection shapes patterns of genetic variation in humans and other species; (ii) elucidate the causes and consequences of variation in recombination rates in humans and in non-model organisms; and (iii) learn about the determinants of germline mutation rates and their evolution in primates.

(i) The effects of natural selection on genetic variation

Natural selection

from Hernandez et al. 2011

How do species adapt to their environment: does it involve most of the genome or only a small fraction; large changes in response to major environmental shifts or primarily fine-tuning? In principle, comprehensive answers to these questions can be obtained by analyzing genetic variation data within and between species.

Until recently the efforts to address these questions and to identify specific targets of adaptation have been guided by the “classic selective sweep” model, in which a new, strongly beneficial mutation increases in frequency to fixation in the population. This model captures only one of many possible modes of selection, however, and makes strong, and often implicit, assumptions about the nature of adaptation. Recent genomic data sets have made it feasible to test such assumptions. Analyzing the pilot data from the 1000 Human Genome Project, for instance, led us to conclude that classic sweeps were a rare mode of adaptation over the past ~250,000 years of human evolution (Hernandez et al. 2011 Science), consistent with findings from Jonathan Pritchard’s group (Coop et al. 2009 PLoS Genetics). Although additional cases of classic sweeps undoubtedly remain to be found, these results suggest that many human adaptations may have taken a different form, motivating a shift in focus.

In particular, we have become interested in characterizing and identifying instances of balanced polymorphisms, i.e., cases where selection leads to the maintenance of two or more variants in the population in the face of stochastic loss. Together with Carole Ober (University of Chicago), we showed that A and B blood types were examples of ancient balanced polymorphisms that persisted for tens of millions of years across primates. More recently, in collaboration with Gil McVean (Oxford), we conducted a genome-wide scan for balanced polymorphisms shared identical by descent between humans and chimpanzees, uncovering at least six cases outside of the MHC.

Examples of lab publications on this topic are Teshima et al. 2006 Genome Research; Hernandez et al. 2011 Science; Wilson et al. 2011 PLoS Genetics; Segurel et al. 2012 PNAS and Leffler et al. 2013 Science.

(ii) Causes and consequences of variation in recombination

Recombination in humans

from Baudat et al. 2010

Recombination is a fundamental meiotic process that helps to align chromosomes, ensure proper disjunction and maintain genome integrity. These roles impose a number of constraints on the number and placement of recombination events on each chromosome. In humans, errors in the recombination process can lead to aneuploidy and chromosomal rearrangements, highly deleterious outcomes.

In spite of its essential role, however, recombination is a highly variable phenotype, with modifiers of the broad-scale recombination rate segregating in natural populations. In humans at least, this variation has consequences for fertility: mothers with a higher mean recombination rate have slightly more children. At a finer-scale, extensive modeling work in evolutionary biology suggests that local modifiers of recombination can be selected for because of their effects on population dynamics, independent of the role of recombination in meiosis. Finally, cross-species comparisons indicate that local recombination rates are labile. In particular, we found that human and chimpanzee genomes differ markedly in the locations of recombination hotspots, indicating that a dramatic reshuffling of the recombination landscape has occurred over a short evolutionary time scale.

These observations raise a number of questions, including: How much variation in recombination rates exists in natural populations, and over what genetic scales does it exert its effects? How is the recombination process so tightly regulated in the face of extensive variability? What precise constraints on the recombination process are imposed by its roles in meiosis and possibly by other roles? How can a process as essential as recombination evolve so rapidly? To understand the evolutionary dynamics of recombination, we are studying the genetic basis for rate variation as well as the selective pressures imposed by the roles of recombination in meiosis and in evolution.

Lab publications on the topic include Coop et al. 2008 Science; Fledel-Alon et al. 2009 PLoS Genetics; Baudat et al. 2010 Science; Fledel-Alon et al. 2011 PloS One; Auton et al. 2012 Science; Kermany et al. 2014 Bioinformatics; and Singhal et al. 2015 Science.

(iii) Determinants of germline mutation rates and their evolution

Recent work in our group aims to learn about properties of germline mutation in humans and close evolutionary relatives. The revolution in sequencing technologies has made it feasible to identify de novo mutations in transmissions from parents to offspring, providing an unprecedented opportunity to learn about the genesis and properties of germline mutations. When recent pedigree studies are considered jointly, however, and alongside results from other methodologies, it becomes clear that the pieces of the puzzle do not fit together. We consider these gaps in our understanding in terms of three sets of interwoven questions: (i) On a mechanistic level, what proportion of mutations is introduced through mistakes in the replication process versus non-replicative, “spontaneous” errors? (ii) In terms of variation among individuals, why do mutation rates depend so strongly on sex and age? (iii) From an evolutionary perspective, how do mating systems and life history traits shape the mutation rate of a species? To address them, we combine the analysis of genome sequence data from pedigrees, phylogenetic analyses in primates, and mathematical modeling.

Lab publications on this topic include Segurel, Wyman & Przeworski 2014 Annual Reviews of Genomics and Human Genetics; Gao et al. 2016 PLoS Biology; and Moorjani, Amorim et al. 2016 PNAS.

For more on population genetics and some of the work going on in the group.