Regulation of gene expression occupies a central role in the control of the flow of genetic information from genes to proteins. Regulatory events on multiple levels ensure that the majority of the genes are expressed under controlled circumstances to yield temporally controlled, cell and tissue-specific expression patterns. The combined set of expressed RNA transcripts constitutes the transcriptome of a cell, and can be analysed on a large-scale using both sequencing and microarray-based methods. The objective of this work has been to develop tools for analysis of the transcriptomes (methods), and to gain new insights into several aspects of the stem cell transcriptome (applications). During recent years expectations of stem cells as a resource for treatment of various disorders have emerged. The successful use of endogenously stimulated or ex vivo expanded stem cells in the clinic requires an understanding of mechanisms controlling their proliferation and self-renewal. This thesis describes the development of tools that facilitate analysis of minute amounts of stem cells, including RNA amplification methods and generation of a cDNA array enriched for genes expressed in neural stem cells. The results demonstrate that the proposed amplification method faithfully preserves the transcript expression pattern. An analysis of the feasibility of a neurosphere assay (in vitro model system for study of neural stem cells) clearly shows that the culturing induces changes that need to be taken into account in design of future comparative studies. An expressed sequence tag analysis of neural stem cells and their in vivo microenvironment is also presented, providing an unbiased large-scale screening of the neural stem cell transcriptome. In addition, molecular mechanisms underlying the control of stem cell self-renewal are investigated. One study identifies the proto-oncogene Trp53 (p53) as a negative regulator of neural stem cell self-renewal, while a second study identifies genes involved in the maintenance of the hematopoietic stem cell phenotype. To facilitate future analysis of neural stem cells, all microarray data generated is publicly available through the ArrayExpress microarray data repository, and the expressed sequence tag data is available through the GenBank. / QC 20100927

Studies of prophylactic salpingectomy specimens from BRCA1/2 mutation carriers, at risk for tubal and ovarian high-grade serous carcinoma (SerCa), have consistently revealed occult carcinomas and putative histological cancer precursors in the distal fallopian tube epithelium (FTE), supporting the FTE as the source of SerCa. In this thesis I molecularly characterized and compared non-malignant FTE from mutation carriers (FTEb) and control patients (FTEn) to identify alterations that may predispose to malignant transformation. Gene expression profiling of laser capture microdissected FTEn, FTEb and SerCa indicated that SerCa have similar molecular profiles whether of presumed ovarian or tubal origin, supporting the notion they share a common cell of origin within the FTE. Furthermore, FTEb samples obtained during the post-ovulatory luteal phase showed gene expression profiles closely resembling SerCa samples, suggesting that the luteal phase milieu may contribute to serous carcinogenesis. An initial hypothesis was that FTEb may respond differently to luteal progesterone compared to FTEn, via differential expression of progesterone receptor (PR) isoforms. However, similar relative isoform expression in FTEn and FTEb samples suggested that a luteal phase-associated factor other than progesterone directs gene expression changes in FTEb. The possibility that FTEb respond differently to ovulation-associated inflammatory cytokines that are locally elevated during the luteal phase was next investigated. Importantly, FTEb specimens previously found to cluster with SerCa based on their global gene expression profiles showed evidence of increased nuclear factor-κB (NFκB)-dependent (pro-inflammatory) signalling and diminished glucocorticoid receptor (GR)-dependent (anti-inflammatory) signalling. Furthermore, I demonstrate that disabled homolog 2 (DAB2), an adaptor molecule decreased in SerCa and FTE luteal samples, enhances both GR-mediated transactivation and suppression of NFκB signalling, implicating DAB2 as a crucial determinant of inflammatory signalling and ovarian cancer risk. Altogether, this thesis identifies gene expression changes in FTE from BRCA mutation carriers during the post-ovulatory luteal phase that parallel those detected in SerCa. The data support a proposed novel testable model for predisposing events contributing to SerCa that centres on an altered ability to quickly resolve the pro-inflammatory environment created by the ovulatory event.

Studies of prophylactic salpingectomy specimens from BRCA1/2 mutation carriers, at risk for tubal and ovarian high-grade serous carcinoma (SerCa), have consistently revealed occult carcinomas and putative histological cancer precursors in the distal fallopian tube epithelium (FTE), supporting the FTE as the source of SerCa. In this thesis I molecularly characterized and compared non-malignant FTE from mutation carriers (FTEb) and control patients (FTEn) to identify alterations that may predispose to malignant transformation. Gene expression profiling of laser capture microdissected FTEn, FTEb and SerCa indicated that SerCa have similar molecular profiles whether of presumed ovarian or tubal origin, supporting the notion they share a common cell of origin within the FTE. Furthermore, FTEb samples obtained during the post-ovulatory luteal phase showed gene expression profiles closely resembling SerCa samples, suggesting that the luteal phase milieu may contribute to serous carcinogenesis. An initial hypothesis was that FTEb may respond differently to luteal progesterone compared to FTEn, via differential expression of progesterone receptor (PR) isoforms. However, similar relative isoform expression in FTEn and FTEb samples suggested that a luteal phase-associated factor other than progesterone directs gene expression changes in FTEb. The possibility that FTEb respond differently to ovulation-associated inflammatory cytokines that are locally elevated during the luteal phase was next investigated. Importantly, FTEb specimens previously found to cluster with SerCa based on their global gene expression profiles showed evidence of increased nuclear factor-κB (NFκB)-dependent (pro-inflammatory) signalling and diminished glucocorticoid receptor (GR)-dependent (anti-inflammatory) signalling. Furthermore, I demonstrate that disabled homolog 2 (DAB2), an adaptor molecule decreased in SerCa and FTE luteal samples, enhances both GR-mediated transactivation and suppression of NFκB signalling, implicating DAB2 as a crucial determinant of inflammatory signalling and ovarian cancer risk. Altogether, this thesis identifies gene expression changes in FTE from BRCA mutation carriers during the post-ovulatory luteal phase that parallel those detected in SerCa. The data support a proposed novel testable model for predisposing events contributing to SerCa that centres on an altered ability to quickly resolve the pro-inflammatory environment created by the ovulatory event.

Advances in high-throughput genomic and proteomic technology have led to a growing interest in cancer biomarkers. These biomarkers can potentially improve the accuracy of cancer subtype prediction and subsequently, the success of therapy. However, identification of statistically and biologically relevant biomarkers from high-throughput data can be unreliable due to the nature of the data--e.g., high technical variability, small sample size, and high dimension size. Due to the lack of available training samples, data-driven machine learning methods are often insufficient without the support of knowledge-based algorithms. We research and investigate the benefits of using knowledge-based algorithms to solve clinical prediction problems. Because we are interested in identifying biomarkers that are also feasible in clinical prediction models, we focus on two analytical components: feature selection and predictive model selection. In addition to data variance, we must also consider the variance of analytical methods. There are many existing feature selection algorithms, each of which may produce different results. Moreover, it is not trivial to identify model parameters that maximize the sensitivity and specificity of clinical prediction. Thus, we introduce a method that uses independently validated biological knowledge to reduce the space of relevant feature selection algorithms and to improve the reliability of clinical predictors. Finally, we implement several functions of this knowledge-based method as a web-based, user-friendly, and standards-compatible software application.

Adult hippocampal neurogenesis is regulated at various levels and by various factors. Genetic influence is an important key determinant of adult neurogenesis and exerts its effects at all levels. In vivo studies have suggested that adult hippocampal neurogenesis is highly variable and heritable among different laboratory strains of mice. To dissect the genetic effect from other contributing factors, it is necessary to study adult neurogenesis under highly controlled environment conditions. We extracted adult hippocampal precursor cells (AHPCs) from 20 strains of the BXD set of recombinant inbred mice, cultured them and studied the effect of genetic background on neurogenesis. The BXD panel consists of mouse lines derived from an intercross between inbred parentals C57BL/6J and DBA/2J. Both of the parentals are fully sequenced and all the strains are well characterized in terms of genotypic and phenotypic characteristics. This allows us to use advanced genetic techniques to identify novel genomic loci and gene-gene interactions important in adult neurogenesis. Comparison of the AHPCs from 20 BXD strains, with respect to cell proliferation and neuronal and astrocytic differentiation in vitro, revealed a large variation for these traits across the strains. Proliferation, as measured by BrdU incorporation, showed over two- fold differences between the extremes. Similar differences were observed for neurogenic (4-fold) and astrogenic differentiation (2-fold). These three traits all showed strong heritability values indicating that the differences were mainly attributed to the genetic component. QTL mapping, with these phenotypic data, revealed that there was no major contribution from single loci controlling these traits. Instead, we found many loci with smaller effects associated with these traits. Gene expression profiling using RNA samples from proliferating cultures of the 20 BXD mice strains yielded two cis eQTL candidates that directly regulated proliferation, LRP6 and Chchd8. LRP6 is well known as a co-receptor of Wnt signaling, but the function of Chchd8 is not known. Further experimentation, using over expression and gene silencing demonstrated that LRP6 negatively regulates AHPCs proliferation. Thus, from this study using a system genetics approach, we were able to identify, LRP6 as a novel regulator of adult hippocampal neurogenesis.

Gene expression profiles are some of the most abundant sources of data about the cellular state of a collection of cells in an organism. Comparison of the expression profiles of multiple samples allows biologists to find associations between observations at the molecular level and the phenotype of the samples. A key challenge is to distinguish variation in expression due to biological factors of interest from variation due to confounding factors that can arise for unrelated technical or biological reasons. This thesis presents models that can explicitly adjust the comparison of expression profiles to account for specific types of confounding factors. One such confounding factor arises when comparing tissue-specific expression profiles across multiple organisms to identify differences in expression that are indicative of changes in gene function. When the organisms are separated by long evolutionary distances, tissue functions may be re-distributed and introduce expression changes unrelated to changes in gene function. We developed Brownian Factor Phylogenetic Analysis, a model that can account for such re-distribution of function, and demonstrate that removing this confounding factor improves tasks such as predicting gene function. Another confounding factor arises because current protocols for expression profiling require RNA extracts from multiple cells. Often biological samples are heterogeneous mixtures of multiple cell types, so the measured expression profile is an average of the RNA levels of the constituent cells. When the biological sample contains both cells of interest and nuisance cells, the confounding expression from the nuisance cells can mask the expression of the cells of interest. We developed ISOLATE and ISOpure, two models for addressing the heterogeneity of tumor samples. We demonstrated that modeling tumor heterogeneity leads to an improvement in two tasks: identifying the site of origin of metastatic tumors, and predicting the risk of death of lung cancer patients.

Gene expression profiles are some of the most abundant sources of data about the cellular state of a collection of cells in an organism. Comparison of the expression profiles of multiple samples allows biologists to find associations between observations at the molecular level and the phenotype of the samples. A key challenge is to distinguish variation in expression due to biological factors of interest from variation due to confounding factors that can arise for unrelated technical or biological reasons. This thesis presents models that can explicitly adjust the comparison of expression profiles to account for specific types of confounding factors. One such confounding factor arises when comparing tissue-specific expression profiles across multiple organisms to identify differences in expression that are indicative of changes in gene function. When the organisms are separated by long evolutionary distances, tissue functions may be re-distributed and introduce expression changes unrelated to changes in gene function. We developed Brownian Factor Phylogenetic Analysis, a model that can account for such re-distribution of function, and demonstrate that removing this confounding factor improves tasks such as predicting gene function. Another confounding factor arises because current protocols for expression profiling require RNA extracts from multiple cells. Often biological samples are heterogeneous mixtures of multiple cell types, so the measured expression profile is an average of the RNA levels of the constituent cells. When the biological sample contains both cells of interest and nuisance cells, the confounding expression from the nuisance cells can mask the expression of the cells of interest. We developed ISOLATE and ISOpure, two models for addressing the heterogeneity of tumor samples. We demonstrated that modeling tumor heterogeneity leads to an improvement in two tasks: identifying the site of origin of metastatic tumors, and predicting the risk of death of lung cancer patients.

Thesis (Ph.D. in Bioinformatics) -- University of Colorado Denver, 2008. / Typescript. Includes bibliographical references (leaves 141-150). Free to UCD Anschutz Medical Campus. Online version available via ProQuest Digital Dissertations;

Thesis (Ph. D.)--University of Oklahoma. / Bibliography: leaves 106-134.

Diss. (sammanfattning) Umeå : Umeå universitet, 2010.