Research Interests

I am interested in a variety of topics in evolutionary biology. My research has focused on two main areas: coalescent theory in the context of geographically structured populations, and models of intragenomic conflict and genomic imprinting.

In the area of geographic structure, I am currently developing a computational framework for the analysis of human genetic data. Information from archaeology, geology, and anthropology can be used to construct geographically explicit models of human demographic history. These models will then allow us to perform more sophisticated analyses of human genetic data, both in the context of basic questions about human history, and for more powerful approaches to association-mapping studies to identify disease genes.

My work in imprinting has led me to an interest in complex coevolutionary dynamics. In particular, I am interested in systems that are shaped by the interaction of their constituent parts, where those separate parts interact sometimes cooperatively, and sometimes antagonistically. I believe that contexts in which there is a dialectic relationship between cooperative and antagonistic processes are particularly fertile grounds for innovation.

Coalescent Theory in Geographically Structured Populations

Every species has come into existence coincident both in space and time with a pre-existing closely allied species.

                                    – A. R. Wallace 1855

Since the earliest days of evolutionary biology, geography has been recognized as an important factor shaping biological diversity.  The dispersal behavior of individuals determines the range of environments they encounter, who they mate with, and who they compete with for resources.  When the distances traveled by individuals are much smaller than the range of the population, the population can become geographically structured.  That is, individuals that are geographically close together will tend to be genetically more similar than individuals that are far apart.

Patterns of genetic diversity are therefore partly shaped by dispersal and geography.  This means that by studying present-day patterns of genetic diversity, we can gain an understanding of the dispersal history of the population.  Determining the dispersal behavior of natural populations is important for understanding natural processes such as speciation.  The dispersal behavior of pathogens and their carriers is also important for designing strategies of disease control.  For instance, the optimal strategy for using pesticides to control the spread of malaria depends on the dispersal behavior of the mosquitos that carry the disease.

Most genetic analyses assume very simple, often unrealistic, geographic models.  These models have the advantage of mathematical tractability, but often neglect geographic features that may have been important in establishing the patterns in genetic data.  My work has focused on in developing more sophisticated and realistic models of geographic structure, and new methods of data analysis that can account for the geographic complexities of real-world populations.

Genomic Imprinting

We inherit two copies of most genes, one from our mother, and another from our father. In most cases, it does not matter which copy came from which parent. That is, the gene's behavior – where and when it is expressed, and what protein it makes – is determined by its DNA sequence, which is copied from generation to generation, changing only when a mutation occurs. Some genes, however, retain a chemical memory of which parent they came from, and take on one of two different behaviors – one if the gene was inherited from the mother, and another if it came from the father.

Genes that exhibit this sort of parent-of-origin-specific behavior are referred to as "imprinted genes." The evolution of imprinting has been explained as the outcome of a conflict between the maternally derived and paternally derived genes within a single organism. This intragenomic conflict can, in some sense, be thought of as an extension of the battle of the sexes into the next generation. Males and females each modify the genes that they are passing on to their children in different ways. These conflicting modifications are attempts to alter the behavior of the offspring to their own benefit.

The best understood examples of imprinting involve genes that are expressed in fetal tissues and influence the distribution of maternal resources. The optimal level of demand for an offspring to place on its mother is determined by a tradeoff between the benefit to oneself, and the cost to the mother's other offspring, who share many of the same genes. For a maternally derived gene, demanding more resources from the mother acquires more resources, and a fitness benefit to oneself. Resources taken generate a fitness cost for the mother's other offspring, each of whom has a 50% chance of carrying the gene. For a paternally derived gene, a similar tradeoff applies, except that the mother's other offspring have less than a 50% chance of carrying the same gene, because some of them could have a different father. This means that the paternally derived allele will favor placing a greater demand on the mother than the maternally derived allele will.

Simple game-theoretic analysis predicts that imprinted genes that increase the demand on maternal resources will be maternally silenced, whereas those that decrease demand will be paternally silenced. This pattern has held up well as we have learned more about the functions of imprinted genes. My own research has focused on extending this sort of analysis to more complex, multi-way conflicts. It is possible to identify at least six distinct "genetic factions" that influence the establishment, propagation, and interpretation of the chemical marks associated with imprinted genes. Understanding the dynamic coevolution among these factions allows us to understand aspects of the molecular mechanisms of imprinting, and to describe the general coevolutionary principles that apply to partially allied, partially conflicting, complex systems.

Human Demographic History

The origin of modern humans has always been a contentious issue in evolutionary biology.  In recent decades, debates about where Homo sapiens arose and how they spread across the world centered around two major hypotheses.  The multi-regional, or “candelabra” model proposed that the human races evolved simultaneously in different regions of the world from their respective Homo erectus ancestors.  The “Out of Africa” or “Recent African Origins” model argued that Homo sapiens arose recently in Africa and then expanded across the world. 

The dispute remained unresolved until advances in DNA sequencing technology made it possible to examine human genetic diversity on a global scale, and it was discovered that the mitochondria and Y-chromosomes of all living humans share a recent common ancestry that strongly favors the Out of Africa model.

This result illustrates the potential power of DNA for studying human history, and recently, genetic tools have been applied to other questions about our ancestors.  Researchers have looked for evidence of the extent and scope of polygamy (reproductive skew), patterns of marital residence (matrilocality versus patrilocality), and whether the emerging Homo sapiens population interbred with other populations (such as Neanderthals) or drove them to extinction.  Unfortunately, this work has not led to the sort of revolutionary insights that came from the early studies of the mitochondria and Y-chromosome.  Analyses of different datasets support contradictory conclusions, and many inferences from genetic data are inconsistent with what we know about human history from other sources.

This confusion occurs because most research in human genetics is based on simple, equilibrium models that fail to capture critical aspects of human population dynamics.  These models are chosen for tractibility rather than realism, and while simple models are often valuable for heuristic analysis, they are not well suited to quantitative analyses of complex systems. 

Anatomically modern humans expanded across the globe within the last 100,000 years, and have undergone dramatic population growth.  Furthermore, many of the factors that affect genetic diversity, such as mating and residence patterns, are influenced by changing cultural practices.  For example, the emergence of agriculture is associated with a shift towards patrilocality.  Since agriculture arose at different times in different locations, and is still not practiced everywhere, the consequences of this shift can not be understood in terms of a model that assumes a population at equilibrium.  Fortunately, research from fields such as geology, archaeology, and anthropology provides a great deal of information that can be used to construct and constrain a more sophisticated, non-equilibrium model of human history.