<div>In the past several decades, the advent of high-throughput biotechnologies for genomics study provides appealing opportunities for us to understand the complex gene interaction inside biological systems, attracting many researches in constructing gene regulatory networks (GRNs). Motivated by the promise of the genetical genomics</div><div>study, our research group has recently focused on representing gene regulatory networks using structural equation models and further revealing system-wide gene regulations.This dissertation presents two recent works along this direction.</div><div><br></div><div><div>Firstly, we conducted thorough theoretical analysis of the recently proposed Two-Stage Penalized Least Squares (2SPLS) method for constructing large systems of structural equation models. We establish the estimation and prediction error bounds for results at both stages of 2SPLS as well as its variable selection consistency. Specically, a bounded eigenvalue assumption is imposed to ensure the consistency properties of the ^l2-penalized regressions at the first stage. At the second stage, the estimation and</div><div>variable selection consistency of the ^l1-penalized regressions are obtained by assuming a restricted eigenvalue condition and a variant of irrepresentable condition, which are both commonly employed in the current literature. We will show that the 2SPLS estimator works not only for fixed dimensions but also diverging dimensions which can grow to infinity with the sample size but at an appropriate rate.</div></div><div><br></div><div><div>Secondly, we developed a novel statistical method to identify structural differences between two cognate networks characterized by structural equation models. We</div><div>propose to reparameterize the model to separate the differential structures from common structures, and then design an algorithm with calibration and construction stages to identify these differential structures directly. The calibration stage serves to obtain consistent prediction by building the^l2 regularized regression of each endogenous</div><div>variables against pre-screened exogenous variables, correcting for potential endogeneity issue. The construction stage consistently selects and estimates both common and</div><div>differential effects by undertaking ^l1 regularized regression of each endogenous variable against the predicts of other endogenous variables as well as its anchoring exogenous</div><div>variables. Our method allows for easy parallel computation. Theoretical results are obtained to establish non-asymptotic error bounds of predictions and estimates at both stages. Our studies on simulated data demonstrated that the proposed method performed much better than independently constructing networks. A real data set</div><div>was analyzed to illustrate the applicability of our method.</div></div>

Afin de mieux comprendre les mécanismes impliqués dans l'initiation des racines coronaires chez le riz nous avons recherché des gènes régulés par le facteur de transcription CROWN ROOTLESS 1 (CRL1) qui contrôle leur initiation en réponse à l'auxine. Différentes approches de transcriptome ont été mises en œuvre. La première a consistée à rechercher les gènes différenciellement exprimés dans les bases de tige du mutant crl1 par rapport au sauvage. La seconde a consisté à rechercher des gènes régulés par l'auxine de manière CRL1-dépendant. La troisième a consisté à rechercher des gènes dont l'expression est induite dans les bases de tige du mutant crl1 après expression conditionnelle ectopique du gène CRL1. Parmi les gènes identifiés des expériences de qRT-PCR nous ont permis de valider 11 gènes comme étant induit par l'auxine de manière CRL1 dépendant et des expérience d'hybridation in situ, 10 gènes qui s'expriment de manière spécifique dans les primordium de racine coronaires. La majorité de ces gènes codent pour des facteurs de transcription et des éléments de transduction de signaux. Certains sont des modulateurs de la stucture de la chromatine d'autre des transporteurs d'auxine. Ces résultats éclairent sur les cibles régulées par le facteur de transcription CRL1 et apportent des éléments nouveaux sur les mécanismes moléculaires impliqués dans la régulation de l'initiation des racines coronaires chez le riz. / In order to understand better the mechanisms involved in crown root initiation in rice we researched the genes regulated by the CROW ROOT LESS 1 (CRL1) transcription factor that controls their initiation in response to auxin. Several transcript profiling approaches have been used. The first was to look for the genes differentially expressed in crl1 stem bases relatively to the wild type. The second one was to research genes that are CRL1-dependant auxin responsive. The last one consisted to research genes that are up-regulated in crl1 stem bases just after the inducible ectopic expression of CRL1. Among identified genes RT-qPCR experiments allowed to validate 11 CRL1-dependant auxin responsive genes and in situ hybridization experiments ten genes that are specifically expressed in crown root primordia. Most of these genes encodes transcription factors or components of transduction signal patways. Some of them encode chromating modulling factors or auxin transporters. These results give new knowledge about the gene regulatory network acting down-stream CRL1 and about the molecular mechanisms involved in crown root initiation in rice.

This dissertation develops and analyzes differential equation-based mathematical models and efficient numerical methods and algorithms for genetic regulatory network identification. The primary objectives of the dissertation are to design, analyze, and test a general variational framework and numerical methods for seeking its approximate solutions for reverse engineering genetic regulatory networks from microarray datasets using the approach based on differential equation modeling. In the proposed variational framework, no structure assumption on the genetic network is presumed, instead, the network is solely determined by the microarray profile of the network components and is identified through a well chosen variational principle which minimizes a biological energy functional. The variational principle serves not only as a selection criterion to pick up the right biological solution of the underlying differential equation model but also provide an effective mathematical characterization of the small-world property of genetic regulatory networks which has been observed in lab experiments. Five specific models within the variational framework and efficient numerical methods and algorithms for computing their solutions are proposed and analyzed in the dissertation. Model validations using both synthetic network datasets and real world subnetwork datasets of Saccharomyces cerevisiae (yeast) and E. Coli are done on all five proposed variational models and a performance comparison vs some existing genetic regulatory network identification methods is also provided. As microarray data is typically noisy, in order to take into account the noise effect in the mathematical models, we propose a new approach based on stochastic differential equation modeling and generalize the deterministic variational framework to a stochastic variational framework which relies on stochastic optimization. Numerical algorithms are also proposed for computing solutions of the stochastic variational models. To address the important issue of post-processing computed networks to reflect the small-world property of underlying genetic regulatory networks, a novel threshholding technique based on the Random Matrix Theory is proposed and tested on various synthetic network datasets.

As food scares have hastened the growth of safety and quality standards around the world, certification schemes to assure various attributes of foods have proliferated in the global marketplace. High-value food commodities produced in the global south for export have been the subject of such schemes through third-party environmental certifications, providing regulatory and verification mechanisms welcomed by global buyers. As certification becomes more common, re-localization in the current global context can also mean the projection of place onto a food commodity to highlight its origin or attributes secured by transparent verification mechanisms. However, environmental food certification is often criticized for its inapplicability in the context of the global south, due to the extensive documentation requirements and high costs. The key question here is the process for small-scale producers in the global south to navigate increasing international regulation of food safety and quality. This dissertation examines (1) how the environmental standards (as defined by the global north) were translated in the rural global south through international certification schemes, and (2) what the implications are at the local level, especially where producers had not yet integrated into conventional global markets before the introduction of certification. The dissertation also analyzes the influence of such certification in determining the development trajectories of rural society in the global south. A case study is used to examine newly-introduced certified organic shrimp production in Ca Mau Province in Vietnam’s Mekong Delta. The selected shrimp production site is the first pilot organic shrimp project in Vietnam working with an international third-party certification scheme. It is located in rural Vietnam where, as in other parts of Southeast Asia, an accelerated process of agrarian transition is underway. Whereas elsewhere the trend with intensified regulation has been the consolidation of large-scale farms and the exclusion of small-scale farms from international agrofood markets, this case study demonstrates comparative advantages of small-scale farms over large-scale farms in producing sensitive high-value crops. This dissertation employs two main analytical approaches. The first approach is to examine the network of actors and the flow of information, payment and shrimp at the production level using environmental regulatory network (ERN). In contrast to chain analyses, which can be useful in identifying linear structure of supply chains for global commodities, ERN can capture the interrelatedeness of actors in the network built around environmental certification for agrofood products. The second analytical lens is that of agrarian transition. Countries experiencing agrarian transition at present are doing so in a very different international context from countries that accomplished their transitions in the past. Results of this research indicate that technical and financial constraints at the time of initial certification are not the primary obstacles to farmers getting certified, since the extensive farming method employed at the study site is organic by default. In spite of this, many farmers unofficially withdrew from the organic shrimp project by simply shifting their marketing channel back to a conventional one. Inefficient flows of information and payments, and a restrictive marketing channel within the environmental regulatory network that does not take into account local geographical conditions and farming practices, all contributed to limiting the farmers’ capacity and lowering their incentives to get involved in the network. The analysis also indicates that, by influencing those agrarian transition processes, food standards and certification based on values developed in the global north may modify, reshape and/or hold back agrarian transition processes in agricultural sectors of developing countries. The potential benefits of environmental certification are enhanced rural development, by generating opportunities for small-scale farmers to connect to global niche markets. The findings of this dissertation highlighted that such certification schemes or their environmental regulatory networks need to ensure information sharing and compensation for farmers. As an empirical finding, this dissertation also captures where ecological credibility and market logic meet: the success of this kind of certification depends on finding a balanced point where standards are ecologically (or ethically) credible to the level that does not attract too much criticism for being green washing, but not too unrealistic to become a disincentive for farmers to participate.

Epithelial-mesenchymal transitions (EMTs) are fundamental and indispensable to embryonic morphogenesis throughout the animal kingdom. At the onset of gastrulation in the sea urchin embryo, micromere-derived primary mesenchyme cells (PMCs) undergo an EMT process to ingress into the blastocoel, and these cells later become the larval skeleton. Much has been learned about PMC specification in sea urchin embryos. However, much less is known about how states of the sequentially progressing PMC gene regulatory network (GRN) controls the EMT process during PMC ingression. Transcriptional regulators such as Snail and Twist have emerged as important molecules for controlling EMTs in many model systems. Sea urchin snail and twist genes were cloned from Lytechinus variegates, and each has been experimentally connected to the PMC regulatory network; these experiments demonstrate several requirements for PMC ingression, and in doing so, begin to illustrate how a gene regulatory network state controls morphogenesis. Functional knockdown analyses of Snail with morpholino-substituted antisense oligonucleotides (MASO) in whole embryos and chimeras demonstrated that Snail is required in micromeres for PMC ingression. Investigations also show that Snail downregulates cadherin expression as an evolutionarily conserved mechanism, and Snail positively regulates a required endocytic clearance of epithelial membrane molecules during EMT. Perturbation experiments indicate that Twist has accessory roles in regulating PMC ingression, and possibly plays a maintenance role in PMC specification network state. In addition, Twist also functions in the post-EMT network state, particularly in directing PMC differentiation and skeletogenesis. The recently annotated sea urchin genome accelerates the discovery of new genes and holds strong promise of mapping out a complete canvas of the micromere-PMC gene regulatory network. Using the genome resources we successfully cloned several newly identified PMC genes, and found most of them to be expressed in micromeres just prior to ingression of the nascent PMCs. Current experiments focus on the roles of these genes in preparing for, executing, and/or controlling the mesenchymal behavior following PMC ingression. The functions and inter-relationships of these genes will greatly augment our understanding of how a gene regulatory network state controls a crucial morphogenetic event. / Dissertation

<p>Many biological processes are effectively modeled as networks, but a frequent assumption is that these networks do not change during data collection. However, that assumption does not hold for many phenomena, such as neural growth during learning or changes in genetic regulation during cell differentiation. Approaches are needed that explicitly model networks as they change in time and that characterize the nature of those changes.</p><p>In this work, we develop a new class of graphical models in which the conditional dependence structure of the underlying data-generation process is permitted to change over time. We first present the model, explain how to derive it from Bayesian networks, and develop an efficient MCMC sampling algorithm that easily generalizes under varying levels of uncertainty about the data generation process. We then characterize the nature of evolving networks in several biological datasets.</p><p>We initially focus on learning how neural information flow networks change in songbirds with implanted electrodes. We characterize how they change in response to different sound stimuli and during the process of habituation. We continue to explore the neurobiology of songbirds by identifying changes in neural information flow in another habituation experiment using fMRI data. Finally, we briefly examine evolving genetic regulatory networks involved in Drosophila muscle differentiation during development.</p><p>We conclude by suggesting new experimental directions and statistical extensions to the model for predicting novel neural flow results.</p> / Dissertation

No description available.

Many vertebrate organs form via the sequential, reciprocal exchange of signaling molecules between juxtaposed epithelial (E) and mesenchymal (M) tissues. For example, the instructive signaling potential for tooth development (odontogenesis) resides in the dental epithelium at the initiation-stage, and subsequently shifts to the dental mesenchyme one day later at the bud-stage. However, the properties of the gene regulatory networks (GRNs) that control the signaling dynamics during epithelial-mesenchymal (E-M) interactions in organogenesis are largely unknown. This dissertation describes an interdisciplinary effort between developmental and systems biology to elucidate the E-M GRN that controls early odontogenesis. The results provide a molecular mechanism for the longstanding paradigm of sequential, reciprocal E-M tissue interactions in development. We generated large-scale spatiotemporal gene expression data for the developing mouse tooth. Surprisingly, the shift in signaling molecule expression from E to M is accompanied by a striking concordance in genome-wide expression changes in both E-M compartments as development proceeds. We hypothesized that since diffusible signaling molecules can act on either E or M independent of their tissue site of synthesis, signaling molecules are uniquely able to simultaneously synchronize and couple the transcriptional dynamics and hence the developmental progression of E and M. To identify the unifying mechanism behind concordant E and M genome-wide expression changes in the face of the discordant expression changes in signaling molecule expression, we developed a novel probabilistic technique that integrates regulatory evidence from microarray gene expression data and the literature to determine the E-M GRN for early tooth development. This GRN contains a uniquely configured E-M Wnt/Bmp feedback circuit in which the Wnt and Bmp signaling pathways in E cross-regulate the expression of Wnt and Bmp4 signaling molecules, whereas both pathways jointly regulate Bmp4 expression in M. We validated the Wnt/Bmp feedback circuit in vivo using compound genetic mutations in mice that either short-circuit or break the circuit, and used mathematical modeling to show how the structure of the Wnt/Bmp feedback circuit can account for reciprocal signaling dynamics. Collectively, these results provide a simple mechanistic framework for how simultaneous signal transduction in E-M compartments can account for the signaling dynamics in organogenesis.

Differential gene expression is an essential component of the programs that give rise to specific cellular fates and functions. This differential regulation occurs primarily at the transcriptional level and is controlled by complex regulatory networks governed by the action of transcription factors at specific DNA regulatory elements. Transcription factors rarely act alone, often functioning through combinatorial interactions with other transcription factors, co-factors and chromatin-remodeling proteins. Defining these protein-protein interactions is an essential component to understanding transcription factor function and consequently, the cell as an integrated network.

The basis of developmental biology lies in the idea of when and how cells decide to divide or to differentiate. Previous studies have established several signaling pathways that determine cell fate decisions, including Notch, Wingless, Hedgehog, Bone morphogenetic protein, and Fibroblast growth factor. Signaling converges on transcriptional factors that regulate gene expression. In mouse embryonic stem cells, I explored how pluripotency and differentiation are regulated through opposing actions of beta-catenin-mediated canonical Wnt signaling, and the mechanisms underlying Sonic hedgehog signaling in generating progenitor cells in the ventral neural tube.