<p> The sheer complexity of evolutionary systems biology requires us to develop more sophisticated tools for analysis, as well as more probing and biologically relevant representations of the data. My research has focused on three aspects of evolutionary systems biology. I ask whether a gene&rsquo;s position in the human metabolic network affects the degree to which natural selection prunes variation in that gene. Using a novel orthology inference tool that uses both sequence similarity and gene synteny, I inferred orthologous groups of genes for the full genomes of 8 mammals. With these orthologs, I estimated the selective constraint (the ratio of non-synonymous to synonymous nucleotide substitutions) on 1190 (or 80.2%) of the genes in the metabolic network using a maximum likelihood model of codon evolution and compared this value to the betweenness centrality of each enzyme (a measure of that enzyme&rsquo;s relative global position in the network). Second, I have focused on the evolution of metabolic systems in the presence of gene and genome duplication. I show that increases in a particular gene&rsquo;s copy number are correlated with limiting metabolic flux in the reaction associated with that gene. Finally, I have investigated the proliferative cell programs present in 6 different cancers (breast, colorectal, gastrointestinal, lung, oral squamous and prostate cancers). I found an overabundance of genes that share expression between cancer and embryonic tissue and that these genes form modular units within regulatory, proteininteraction, and metabolic networks. This despite the fact that these genes, as well as the proteins they encode and reactions they catalyze show little overlap among cancers, suggesting parallel independent reversion to an embryonic pattern of gene expression.</p>

<p> Sex determination in leatherback sea turtles is directed primarily by the temperatures a clutch experiences during the middle third of development. Warmer temperatures tend to produce females will cooler temperatures yield males. Nest temperatures can vary spatially and temporally. During the 2010 and 2011 nesting seasons, this study estimated the hatchling sex ratio of leatherback sea turtles on Sandy Point National Wildlife Refuge (SPNWR), St. Croix, U.S. Virgin Islands. I measured sand temperatures from May- August and across the spatial range of leatherback nesting habitat. I spatially interpolated those temperatures to create maps that predicted temperatures for all nests incubating on SPWNR. Nest temperatures were also directly measured and compared with predicted nest temperatures to validate the prediction model. Sexes of dead-in-nest hatchlings and full term embryos were used to confirm the sex-temperature response. The model showed that microclimatic variation likely impacts the production of both sexes on SPNWR.</p>

<p> Plumage color evolution in birds has been the focus of theoretical and empirical research on sexual selection since Darwin. Many of the yellow, orange, and red hues seen in bird plumage are the result of carotenoid pigmentation. While a great number of recent studies have examined the functions of carotenoid-based plumage coloration in a single species, few have examined the evolutionary history of this trait in a comparative phylogenetic context. Using the New World blackbirds as a model clade, I focus on two questions that a comparative phylogenetic approach can uniquely address. First, what is the history of evolutionary change in carotenoid color that led to the colors seen in extant blackbird taxa? Second, by what proximate mechanisms have carotenoid pigments evolved? In Chapter 1, I present an ancestral state reconstruction of carotenoid-based plumage coloration across the Icterid phylogeny, based on reflectance measurements of museum skins. My results show robust evidence that red coloration was gained repeatedly from a yellow common ancestor. In Chapter 2, I used pigment biochemistry of meadowlark (<i>Sturnella</i>) and Cacique (<i>Cacicus</i>) feathers to test whether independent gains of red coloration are the result of parallel or convergent metabolic mechanisms. Meadowlarks have evolved red coloration using a different set of carotenoids than caciques, but the caciques have evolved the same set of carotenoids twice. This suggests that red coloration evolved by convergent evolution among different blackbird clades, but evolved by parallel evolution within the caciques. Lastly, in Chapter 3 I examine the relationship between color and carotenoid pigmentation in orioles, a blackbird clade in which orange has been gained at least twice independently from a yellow common ancestor. I found red-producing keto-carotenoids only in orange species and never in yellow species. This result is a striking contrast to our expectation for a continuous gradient of a carotenoid pigment concentration. These results suggest that repeated gains of C4-oxygenation ability best explain evolutionary changes in orange coloration in orioles. To summarize, I showed using phylogenetic comparative methods that blackbirds have repeatedly evolved towards redder carotenoid coloration. Using HPLC biochemistry, I showed that each of these gains of orange and red coloration is likely the result of a gain of C4-oxygenation ability. The prevalence of gains of orange and red coloration suggests that there may be a directional bias towards evolving longer-wavelength carotenoid plumage. The research presented in these chapters provides the phylogenetic framework necessary for future studies to examine the functional causes underlying the repeated evolution of carotenoid-based coloration.</p>

<p> Adult echinoderms form a mineralized skeleton, but only sea urchins and brittle stars form larval skeletons. In the sea urchin <i>Strongylocentrotus purpuratus</i>, the gene regulatory network (GRN) leading to skeleton formation has been characterized. This <i>S. purpuratus</i> GRN includes several members of the <i>ets</i> family, including <i> Erg, Ets1/2</i> and <i>Gabp</i>. The brittle star <i> Ophiocoma wendtii</i> forms an embryonic skeleton similar to <i> S. purpuratus</i>. The goal of this proposal is to see if expression of the <i>ets</i> family members is conserved as part of the skeletogenic GRN in <i>O. wendtii</i>. Four genes were identified in <i> O. wendtii</i>; homologous to <i>S. purpuratus Erg, Ets1/2, Ets4 </i> and <i>Gabp</i> based on phylogenetic analysis. The coding sequences of these <i>O. wendtii</i> genes were obtained and their temporal expression was determined. These results suggest that sea urchins and brittle stars share a GRN leading to skeleton formation that has been activated in the embryos of both.</p>

<p> <i>Arabidopsis thaliana</i> has undergone three whole genome duplications within its ancestry, and these events have dramatically affected its gene complement. Of the most recent whole genome duplication events (&alpha; event), there remain 11,452 conserved noncoding sequences (CNSs) that have been retained proximal to &alpha; duplicate gene pairs. As functional DNA elements are expected to diverge in sequence at a slower rate than nonfunctional DNA elements, the retained CNSs likely encode gene regulatory function. Within this dissertation I provide evidence for the regulatory role of CNSs within <i> Arabidopsis thaliana</i>. Using a collection of over 5,000 microarray RNA expression profiling datasets, I demonstrate that the presence of CNSs near &alpha; duplicate pairs is correlated with changes in average expression intensity (AEI), &alpha; duplicate pair co-expression, mRNA stability, and breadth of gene expression. The effects of CNSs on AEI, co-expression, and mRNA stability vary relative to their subgene position, because they are located in nontranscribed (5&rsquo;-upstream and 3&rsquo;-downstream) and transcribed (5&rsquo;- UTR, intronic and 3&rsquo;-UTR) regions. Modeling gene interactions through the generation of co-expression networks, I also demonstrate that a portion of CNSs participate in known gene regulatory networks. Collectively, this body of work demonstrates that CNSs regulate steady-state mRNA levels within Arabidopsis thailiana through both transcriptional and post-transcriptional mechanisms.</p>

<p> Flower shape, color and size are extensively studied to both identify and classify different angiosperm taxa. The availability of well-supported molecular phylogenies produced using complex models of sequence evolution, coupled with an understanding of the genes that regulate morphological form in model organisms, and new methods to infer gene expression patterns in diverse species now allow us to understand the genetic basis of morphological differences among closely related species. Studies in Plantaginaceae, Gesneriaceae, Fabaceae and Brassicaceae show the importance of <i>CYCLOIDEA (CYC), RADIALIS (RAD)</i> and <i>DIVARICATA (DIV)</i> in regulating flower shape, but also show divergence in gene function within flowering plants. Previous studies in the zygomorphic model species <i>Antirrhinum majus </i> (snapdragon) have shown that <i>AmCYC</i> is expressed in the adaxial (dorsal) petals of flowers where it activates <i>AmRAD </i>. This expression of <i>AmRAD</i> within adaxial petals represses <i>AmDIV</i> expression causing <i>AmDIV</i> to be restricted to abaxial (ventral) and lateral petals. Like <i>Antirrhinum </i>, traditional <i>Diplacus</i> flowers have distinct dorsal, ventral and lateral petal identities. However, within the clade actinomorphic flowers have evolved independently on two occasions: once in <i>D. pictus </i> and once in <i>D. mohaviensis</i>. mRNA reveal <i> DIV</i> expression to be conserved between <i>D. pictus</i> and snapdragon, whereas <i>CYC</i> and <i>RAD</i> expression, and presumably function, differ between the two species. <i>DpCYC</i> is expressed in a narrow portion on the upper lip of abaxial petals, whereas <i> DpRAD</i> is expressed within both lateral and abaxial petals. <i> D. pictus</i> flowers are characterized by a novel upturned abaxial petal which may be linked to localized <i>CYC</i> expression along the upper surface of the structure. This study sheds new light on the mechanisms regulating flower shape in an endemic Californian monkey flower and shows the importance of testing hypotheses from model species such as <i>Arabidopsis </i> and snapdragon in non-model taxa such as <i>D. pictus</i> to undercover the true variety of mechanisms driving morphological evolution.</p>

<p> Key to understanding the transition that vertebrates made from water to land is determining the developmental and genomic bases for the changes. New bioinformatic tools provide an opportunity to automate the discovery, broaden the number of, and provide an evidence-based ranking for potential candidate genes. I sought to explore this potential for the fin/limb transition, using the substantial genetic and phenotypic data available in model organism databases. Model organism data was used to hypothesize candidate genes for the fin/limb transition. In addition, 131 fin/limb candidate genes from the literature were extracted and used as a basis for comparison with candidates from the model organism databases. Additionally, seven genes specific to limb and 24 genes specific to fin were identified as future fin/limb transition candidates.</p>

<p> Lipid-producing microalgae are a feedstock for commercial products such as nutritional supplements, aquatic animal feed, and biofuels. Unlike most algal phyla, the diatoms (Bacillariophyta) characteristically produce storage lipids throughout their entire lifecycle. In this study, lipids were extracted via chloroform-methanol and quantified as percent dry weight, &mgr;g/mL, and pg/100 &mgr;m³ and then analyzed for a phylogenetic signal by comparing the variability between lineages to the variability within lineages for each metric. These ten taxa were then paired with data gathered from the literature and examined for a phylogenetic signal using previously described methods. In the first analysis, there was greater variability between than within lineages during stationary growth when using percent dry weight as a metric. In the second analysis, a statistically significant phylogenetic signal was detected for nutrient-deplete growth experiments when examining the genus-level phylogeny (P = 0.013).</p>

<p> Historical biogeography has a diversity of methods for inferring ancestral geographic ranges on phylogenies, but many of the methods have conflicting assumptions, and there is no common statistical framework by which to judge which models are preferable. Probabilistic modeling of geographic range evolution, pioneered by Ree and Smith (2008, <i>Systematic Biology</i>) in their program LAGRANGE, could provide such a framework, but this potential has not been implemented until now. </p><p> I have created an R package, "BioGeoBEARS," described in chapter 1 of the dissertation, that implements in a likelihood framework several commonly used models, such as the LAGRANGE Dispersal-Extinction-Cladogenesis (DEC) model and the Dispersal-Vicariance Analysis (DIVA, Ronquist 1997, <i> Systematic Biology</i>) model. Standard DEC is a model with two free parameters specifying the rate of "dispersal" (range expansion) and "extinction" (range contraction). However, while dispersal and extinction rates are free parameters, the cladogenesis model is fixed, such that the geographic range of the ancestral lineage is inherited by the two daughter lineages through a variety of scenarios fixed to have equal probability. This fixed nature of the cladogenesis model means that it has been indiscriminately applied in all DEC analyses, and has not been subjected to any inference or formal model testing. </p><p> BioGeoBEARS also adds a number of features not previously available in most historical biogeography software, such as distance-based dispersal, a model of imperfect detection, and the ability to include fossils either as ancestors or tips on a time-calibrated tree. </p><p> Several important conclusions may be drawn from this research. First, formal model selection procedures can be applied in phylogenetic inferences of historical biogeography, and the relative importance of different processes can be measured. These techniques have great potential for strengthening quantitative inference in historical biogeography. No longer are biogeographers forced to simply assume, consciously or not, that some processes (such as vicariance or dispersal) are important and others are not; instead, this can be inferred from the data. Second, founder-event speciation appears to be a crucial explanatory process in most clades, the only exception being some intracontinental taxa showing a large degree of sympatry across widespread ranges. This is not the same thing as claiming that founder-event speciation is the <i>only</i> important process; founder event speciation as the only important process is inferred in only one case (<i>Microlophus</i> lava lizards from the Galapagos). The importance of founder-event speciation will not be surprising to most island biogeographers. However, the results are important nonetheless, as there are still some vocal advocates of vicariance-dominated approaches to biogeography, such as Heads (2012, <i>Molecular Panbiogeography of the Tropics</i>), who allows vicariance and range-expansion to play a role in his historical inferences, but explicitly excludes founder-event speciation <i> a priori.</i> The commonly-used LAGRANGE DEC and DIVA programs actually make assumptions very similar to those of Heads, even though many users of these programs likely consider themselves dispersalists or pluralists. Finally, the inclusion of fossils and imperfect detection within the same likelihood and model-choice framework clears the path for integrating paleobiogeography and neontological biogeography, strengthening inference in both. </p><p> Model choice is now standard practice in phylogenetic analysis of DNA sequences: a program such as ModelTest is used to compare models such as Jukes-Cantor, HKY, GTR+I+G, and to select the best model before inferring phylogenies or ancestral states. It is clear that the same should now happen in phylogenetic biogeography. BioGeoBEARS enables this procedure. Perhaps more importantly, however, is the potential for users to create and test new models. Probabilistic modeling of geographic range evolution on phylogenies is still in its infancy, and undoubtedly there are better models out there, waiting to be discovered. It is also undoubtedly true that different clades and different regions will favor different processes, and that further improvements will be had by linking the evolution of organismal traits (e.g., loss of flight) with the evolution of geographic range, within a common inference framework. In a world of rapid climate change and habitat loss, biogeographical methods must maximize both flexibility and statistical rigor if they are to play a role. This research takes several steps in that direction. </p><p> BioGeoBEARS is open-source and is freely available at the Comprehensive R Archive Network (http://cran.r-project.org/web/packages/BioGeoBEARS/index.html). A step-by-step tutorial, using the <i>Psychotria</i> dataset, is available at PhyloWiki (http://phylo.wikidot.com/biogeobears). </p><p> (Abstract shortened by UMI.)</p>

<p> Human development and health involves the complex and coordinated regulation of gene expression across diverse tissues. Gene regulation is therefore an essential process in human biology. In the field of human genetics, this has only become more apparent as genomic technologies have made genome-wide surveys of genetic variation underlying human traits possible. In my thesis work, I studied the impact of variation in gene regulation on human traits from three distinct perspectives of human genetics. I first examined the contribution of gene regulation to human disease susceptibility by combining gene expression data with a genome-wide association study to identify novel asthma susceptibility candidate genes. I then studied the effects of depleting specific transcription factors from the cell on downstream gene expression by incorporating gene expression data (following cellular depletion of those factors) with genomic transcription factor binding data. Finally, I considered the role of gene regulation in human evolution by integrating RNA-seq data collected in human, chimpanzee, and rhesus macaque lymphoblastoid cell lines with promoter reporter assays conducted in the same lines. Throughout this work, I have synthesized multiple genomic data sets and multiple distinct sub-disciplines of human genetics in order to arrive at a unified view of the role of gene regulation in determining human traits.</p>