Menu:

Past Research Projects

 

Characterizing Meiotic Drive in Mimulus Species Hybrids

Fall 2006 - Spring 2008
Plant Evolutionary Genomics
With: Dr. Lila Fishman
University of Montana

I worked as a research assistant with Lila Fishman at the University of Montana.  In the Fishman lab, we use the genus Mimulus (monkey flowers) to investigate population and speciation genomics. Mimulus has been a long-standing system for studying the ecology and evolution of flowering plants, but with the development of new genomic resources, including a whole genome sequence, Mimulus offers a unique opportunity to study the underlying genetic architecture of well-characterized ecological and evolutionary processes. 

I worked to characterize the mechanism and evolutionary significance of meiotic drive behavior in Mimulus hybrids.  Meiotic drive is a specific violation of Mendel’s law of random segregation, wherein parental alleles are unequally represented in the progeny.  As opposed to male meiosis, where all four gametic products share the same fate as carriers of parental DNA for the next generation, female meiosis has intrinsic asymmetry: only one of the four gametic products is destined to become the female gametophyte (egg).  Female meiosis can therefore be considered a competitive arena, where certain mutations may confer a heightened ability for chromosomes to make it into the egg and contribute their DNA to the next generation. While meiotic drive has its roots in the biochemistry of meiosis, the genetic ramifications of drive may be seen 1) at the protein level through the natural history of selection on the kinetochore components, 2) at the organismal level through evolutionary trade-offs in life history, and 3) at the population level through a drive-associated haplotypes.


In our system, we have observed what appears to be extreme centromere-based meiotic drive in crosses between two species of Mimulus. Centromere drive has been theorized to play an important role in generating genomic incompatibilities that can lead to speciation, yet up to this point, no known examples exist.  My goals were to establish whether drive is in fact centromeric, characterize how drive manifests in natural populations, and examine how a history of drive may have affected protein evolution in the kinetochore.  To get at these questions, I used cytogenetics, linkage mapping, and sequence comparison, as well as bioinformatics to mine the nascent Mimulus genome build.

 

An Initial Analysis of Genome Synteny Between Daphnia pulex and magna

Summer 2006
Daphnia Comparative Genetics
With: Dr. W.Kelley Thomas
Hubbard Center for Genome Studies, University of New Hampshire

            Working with Dr. Thomas, I learned the theory and practice of building large-insert DNA libraries to support genome assembly.  The Thomas lab participates in the Daphnia Genome Consortium, a group working to assemble and analyze the genome of Daphnia pulex, a model system for ecological and evolutionary genomics.  In an effort to find a productive species for comparative genomics within Daphnia, my project was to construct a BAC and Fosmid library for Daphnia magna, a distant relative of pulex, exhibiting extreme morphological stasis and a different ecological distribution. 
            Toward this end, I first learned the theory behind library construction, which involves comparing the estimated size of the genome, the desired amount of genome coverage and the efficiency of clone production.  After we settled on a method, I extracted intact genomic DNA and analyzed its size and quality using pulse-field gel electrophoresis.  Next, genomic DNA was partially digested and size-selected for transformation.  After transformation, clones were assessed for insert size and consistency.  Our final product was a Fosmid library, consisting of ~8,000 clones with an average insert size of 42 Kb.  This library provides ~ 1.6 times coverage of the magna genome. 
            Ideally, the genomes underlying a productive comparative system should be a mix of tractable similarities and striking differences.  A simple yet informative means of assessing this potential is through a comparison of genome synteny.  Using synteny as a heuristic to explore the productivity of the magna/pulex system, I end-sequenced 50 Fosmid clones and BLASTed the resulting sequence-pairs against the pulex genome.  The data showed a wide variety of syntenic relationships, including extremes of conservation and divergence, suggesting that the magna/pulex system is likely to be highly informative for understanding the evolution, development and adaptation in Daphnia.

 

The Role of 4 Gustatory Neuron Populations in Drosophila Courtship and Courtship Conditioning

Fall 2004-Spring 2006
Drosophila Neurogenetics and Behavior
With: Dr. Kathleen Siwicki
Swarthmore College

I spent the last two years at Swarthmore working in the Siwicki lab, proud father to thousands of fruit flies.  Our lab studies the neural underpinnings of Drosophila courtship conditioning, a phenomenon where male flies learn to suppress courtship toward virgin females after an unfulfilling sexual experience with an already-mated female.  Although courtship is multi-modal, this type of learning and memory is thought to be dependent on hydrocarbon pheromones detected by the male’s gustatory and olfactory systems.  Traditionally, courtship conditioning has been formalized as an associative learning process.   In keeping with this framework, previous work in the Siwicki lab focused on attempting to identify the chemicals constituting the conditioned stimulus (CS) based on a correlative approach: comparing male courtship memory after sexual experience with flies expressing genetically altered pheromone blends. 

For my thesis project, I wanted to take a more direct approach at testing the role taste plays in courtship conditioning.  Using the GAL4-UAS system, genetically defined populations of gustatory receptor neurons (GRNs), the cells responsible for the initial transduction of tastants, were rendered non-functional with a temperature change.  In total, four GRN populations, defined through the expression of putative pheromone receptors, were rendered non-functional in male flies during courtship and courtship conditioning assays and male courtship behavior was quantified.

In addition to providing initial evidence for previously uncharacterized pheromone detecting GRN populations, my results suggest that courtship conditioning may not be a classic example associative learning, where male flies modify their behavior based on a learned change in meaning of a single CS.  Alternatively, I suggest that the learned suppression of male courtship occurs when a top-down negative experience dampens the aphrodisiac-drive of certain gustatory input pathways; when these gustatory aphrodisiac pathways are made unavailable by preventing neuronal activity during the male’s initial negative sexual experience, they may fail to be dampened and thus can continue to drive courtship in future encounters with females. 

 

Adapting Methods of Biological Phylogenetics for Inferring Language Evolution

Summer 2005 - Fall 2005
Linguistic Phylogenetics
With: Dr. Søren Wichmann
Max Planck Institute for Evolutionary Anthropology

Inferring the evolutionary relationships for languages is the one of the main goals of historical linguistics.  Historical linguistics, however, is a relatively old fashioned discipline: lists of cognate words are gathered, shared sound changes are extracted, and simple languages trees are built.    This strategy, known as the  “comparative method,” has been the backbone of historical linguistics for decades.  Unfortunately, the half-life of words is rather short, even with respect to the age of most language families.  Linguists interested in more ancient evolutionary relationships, or attempting to assess general principles of language evolution, must turn to alternative data and methods of analysis; a process of methodological expansion only in its infancy. 

I spent the summer of 2005 in the linguistics department of the MPI-EVA, working as the biologist on a small team of linguistis interested in applying the sophisticated, computationally-driven methods of biological phylogenetics to a recently compiled database of language structures.  The database, called the World Atlas of Linguistic Structures (WALS), is an unprecedented survey of the structural features covering thousands of languages.  While WALS was not built with phylogenetics in mind, as a comparative resource, the database offered a unique chance to explore the possibility of using structures, rather than words, in conjunction with new phylogenetic methods to model evolutionarily “deep” relationships.

My goal was to performance-evaluate methods of biological phylogenetics on structural data.  In practice, this meant surveying the phylogenetic techniques available, thinking through the assumptions of the algorithms, figuring out how to code and input the data, and finally determining the meaning and accuracy of the results.  For my sample, I chose what I thought would be a closest to a model system: the Austronesian languages of island South East Asia and the Pacific.  The natural history of this language family is especially appropriate for phylogenetics.  The languages were distributed through one massive migratory sweep and the island lifestyle has kept many speakers isolated, preventing linguistic hybridization. Additionally, resources for the Austronesian languages are relatively abundant, including a database of comparative vocabulary. 

After spending about a month fishing through old grammars to fill out my dataset, I set to work coding and analyzing the data.  I treated the language data as one would discreet, morphological characters in biology.  I built several datasets varying in data type (structures, words or both) and coding strategy.  I also converted a “known” phylogeny based on the comparative method to a form amenable for similarity comparison with the experimental trees that I generated.  Next I used distance, parsimony, and Bayesian methods to build trees and networks out of the datasets, comparing the results from different datasets within a method (a measure of precision) and to the “known” tree (a measure of accuracy).  While the outcome was broadly informative about the application of these methods, my most important result was the complete recapitulation of the known phylogeny of my sample using Bayesian analysis with a dataset composed of both structures and words.  My hope is that these results represent a strong  “proof of concept” for modeling language evolution with computational methods and may help linguists apply biological methods in a productive and standardized way.   

 

Molecular Evolution of the Bobber Heat Shock Protein in Plants

Spring 2006
Constraints on Gene Evolution
With: Dr. Nick Kaplinsky and Dr. Colin Purrington
Swarthmore College
           
I investigated evolutionary constraints on the Bobber gene as my seminar project for the course “Analysis of Adaptation.”  Bobber codes for a putative heat shock protein essential for development in Arabidopsis.  I wanted to know how selection has shaped the protein within the Kingdom Plantae.  To address this question, I mined plant genomic databases for Bobber express sequence tags (ESTs).  I then drove an alignment of Bobber cDNA based on peptide sequence.  Once I had aligned homologous codons, I analyzed the ratio of synonymous (Ks) to non-synomous (Kn) mutations along the gene using a sliding window analysis from the program K-estimator.  The results showed two distinct trends, corresponding to each of the two protein domains: one domain exhibited a high degree of conservation (low Kn/Ks ratio) while the other domain was more evolutionarily variable (a Kn/Ks ratio approaching 1).  The domains also exhibited different phylogenetic signals.  Together, these data highlight modularity in protein evolution.  

 

Assessing the Role of a Unique Odor in the Behavior of the Colonial Seabird Aethia cristatella (Crested Auklet)

Summer 2004
Behavioral Ecology of Avian Olfaction
With: Dr. Julie Hagelin
Swarthmore College

I spent the summer of 2004 on St. Lawrence Island, Alaska, with a small team investigating the role olfactory communication plays in Aethia cristatella (Crested Auklet) social behavior.  Crested Auklets are shore birds, breeding in large colonies during the arctic summer. At the height of the breeding season the birds have a strong citrus-like smell that we hypothesized plays some communicative role during mate-selection.  We monitored behavior during in-field odor manipulations to assay what if any social role the odor might have.  We also collected blood, feather and odor samples from Auklets that we trapped.  Our initial results suggest that the odor plays more of a role when the Auklets are in the dark rocky crags they use for nests, rather than during the more conspicuous aboveground displays important for mate selection.