RNA-Seq (Whole Transcriptome Shotgun Sequencing) provides an ideal platform to study the complete set of transcripts for a specific developmental stage or physiological condition. It reveals not only expression-level changes, but also structural changes in the coding sequences, including gene rearrangements. In this dissertation, I present my contributions to the development of computational tools for the robust and efficient analysis of RNA-seq data to support cancer research. To automate the laborious and computationally intensive procedure of RNA-seq data management, I worked on the development of Hydra, an RNA-seq pipeline for the parallel processing and quality control of large numbers of samples. With user-friendly reports on quality control and running checkpoints, Hydra makes the data processing procedure fast, efficient and reliable. Here, I report my application of the pipeline to the analysis of patient-derived lymphoma xenograft samples, to show Hydra’s ability to detect abnormalities (e.g., mouse tissue contamination) in the sequencing data. Because fusions play an important role in carcinogenesis, fusion detection has become an important area of methodological research. Several computational methods have been developed to identify fusion transcripts from RNA-seq data. However, all these methods require realignment to the transcriptome, a computationally expensive task, unnecessary in many cases. Here, I present QueryFuse, a novel gene-specific fusion-detection algorithm for aligned RNA-seq data. It is designed to help biologists find and/or computationally validate fusions of interest quickly, and to annotate the detected events with visualization and detailed properties of the supporting reads. By focusing the fusion detection on read pairs aligned to query genes, we can not only reduce realignment time, but also afford to use a more accurate but computationally expensive local aligner. In the extensive evaluation I performed, I obtained comparable or better results compared with two widely adopted tools (deFuse and TophatFusion) on two simulated datasets, as well as on cell line datasets with known fusions. Finally, I contributed to the identification of a novel fusion event in lymphoma, with potential therapeutic implications in clinical samples. I validated this fusion in silico by my putative reference method before experimental validation.

During mouse development, the primordia of the posterior body including the trunk and tail tissues of the embryo forms largely from a bipotential cell population that resides in the posterior growth zone in vertebrate embryos. This bipotential cell population contains neuromesodermal progenitors (NMP) which are found in the tail bud which replaces the primitive streak after gastrulation and contributes to axial elongation by the formation of both the spinal cord and paraxial mesoderm derivatives. The three vertebrate Cdx genes, Cdx1, Cdx2 and Cdx4, encode transcription factors that play important roles in axial elongation since the triple Cdx mutant embryos fail to generate any tissue posterior to the occipital primordia. A comparison of Cdx mutant phenotypes suggests that Cdx2 is the most important contributor to axial elongation since Cdx2 heterozygous mutants exhibit foreshortened tails and Cdx2 conditional mutants exhibit axial truncation and complete loss of tail bud structures. Cdx2 target genes, such as Wnt3a, Cyp26a1 and T, are also essential for axial elongation. Cdx1 null mutants are viable and exhibit homeosis of cervical and anterior thoracic vertebrae, while Cdx4 null mutants are phenotypically normal. In addition, it has been shown that simultaneous loss of multiple copies of Cdx alleles disrupts axial elongation more severely than each single mutation which suggests there is overlapping function among the Cdx family. The genetic network underlying regulation of axial elongation by the Cdx family is not fully understood due in part to this functional overlap. In this thesis, I employed a conditional Cre-loxP system to derive conditional mutants lacking all Cdx functions. Additionally, Pax2-GFP transgenic mice where GFP is expressed under the control of Pax2 locus were used to enrich tail bud NMP cells for RNA-seq and ChIP-seq analysis for Cdx2. Using this approach, I revealed new target genes and pathways that are regulated by Cdx members and likely involved in axial elongation.

The focus for this research is to develop and apply statistical methods to analyze and interpret high-throughput biological data. We developed a novel correlation coefficient, shrinkage correlation coefficient (SCC), that fully exploits the similarity between the replicated microarray experimental samples. The methodology considers both the number of replicates and the variance within each experimental group in clustering expression data, and provides a robust statistical estimation of the error of replicated microarray data. Applying SCC-based hierarchical clustering to the replicated microarray data obtained from germinating spores of the fern Ceratopteris richardii, we discovered two clusters of genes with shared expression patterns during spore germination. This computational approach is not only applicable to DNA microarray analysis but is also applicable to proteomics data or any other high-throughput analysis methodology. The suppression of APY1 and APY2 in mutants expressing an inducible RNAi system resulted in plants with a dwarf phenotype and disrupted auxin distribution, and we used these mutants to discover what genes changed expression during growth suppression. We evaluated the gene expression changes of apyrase-suppressed RNAi mutants that had been grown in the light and in the darkness, using the NimbleGen Arabidopsis thaliana 4-Plex microarray, respectively. We compared the two sets of large-scale expression data and identified genes whose expression significantly changed after apyrase suppression in light and darkness, respectively. Our results allowed us to highlight some of the genes likely to play major roles in mediating the growth changes that happen when plants drastically reduce their production of APY1 and APY2, some more associated with growth promotion and others, such as stress-induced genes, more associated with growth inhibition. There is a strong rationale for ranking all these genes as prime candidates for mediating the inhibitory growth effects of suppressing apyrase expression, thus the NimbleGen data will serve as a catalyst and valuable guide to the subsequent physiological and molecular experiments that will be needed to clarify the network of gene expression changes that accompany growth inhibition. / text

The production of fully mature protein-coding transcripts is an intricate process that involves numerous regulation steps. The complexity of these steps provides the means for multilayered control of gene expression. Comprehensive understanding of gene expression regulation is essential for interpreting the role of gene expression programs in tissue specificity, development and disease. In this thesis, we aim to provide a better global view of the human transcriptome, focusing on its content, synthesis, processing and regulation using next-generation sequencing as a read-out. In Paper I, we show that sequencing of total RNA provides unique insights into RNA processing. Our results revealed that co-transcriptional splicing is a widespread mechanism in human and chimpanzee brain tissues. We also found a correlation between slowly removed introns and alternative splicing. In Paper II, we explore the benefits of exome capture approaches in combination with RNA-sequencing to detect transcripts expressed at low-levels. Based on our results, we demonstrate that this approach increases the sensitivity for detecting low level transcripts and leads to the identification of novel exons and splice isoforms. In Paper III, we highlight the advantages of performing RNA-sequencing on separate cytoplasmic and nuclear RNA fractions. In comparison with conventional poly(A) RNA, cytoplasmic RNA contained a significantly higher fraction of exonic sequence, providing increased sensitivity for splice junction detection and for improved de novo assembly. Conversely, the nuclear fraction showed an enrichment of unprocessed RNA compared to when sequencing total RNA, making it suitable for analysis of RNA processing dynamics. In Paper IV, we used exome sequencing to sequence the DNA of a patient with unexplained intellectual disability and identified a de novo mutation in BAZ1A, which encodes the chromatin-remodeling factor ACF1. Functional studies indicated that the mutation influences the expression of genes involved in extracellular matrix organization, synaptic function and vitamin D3 metabolism. The differential expression of CYP24A, SYNGAP1 and COL1A2 correlated with the patient’s clinical diagnosis. The findings presented in this thesis contribute towards an improved understanding of the human transcriptome in health and disease, and highlight the advantages of developing novel methods to obtain global and comprehensive views of the transcriptome.

High-throughput sequencing has greatly influenced the amount of data produced and biological questions asked and answered. Sequencing approaches have also enabled rapid development of related technological fields such as single-cell and spatially resolved expression profiling. The introductory parts of this thesis give an overview of the basic molecular and technological apparatus needed to analyse the transcriptome in cells and tissues. This is succeeded by a summary of present investigations that report recent advancements in RNA profiling. RNA integrity needs to be preserved for accurate gene expression analysis. A method providing a low-cost alternative for RNA preservation was reported. Namely, a low concentration of buffered formaldehyde was used for fixation of human cell lines and peripheral blood cells (Paper I). The results from bulk RNA sequencing confirmed gene expression was not negatively impacted with the preservation procedure (r2&gt;0.88) and that long-term storage of such samples was possible (r2=0.95). However, it is important to note that a small population of cells overexpressing a limited amount of genes can skew bulk gene expression analyses making them sufficient only in carefully designed studies. Therefore, gene expression should be investigated at the single cell resolution when possible. A method for high-throughput single cell expression profiling termed microarrayed single-cell sequencing was developed (Paper II). The method incorporated fluorescence-activated cell sorting, sample deposition and profiling of thousands of barcoded single cells in one reaction. After sample attachment to a barcoded array, a high-resolution image was taken which linked the position of each array barcode sequence to each individual deposited cell. The cDNA synthesis efficiency was estimated at 17.3% while detecting 27,427 transcripts per cell on average. Additionally, spatially resolved analysis is important in cell differentiation, organ development and pathological changes. Current methods are limited in terms of throughput, cost and time. For that reason, the spatial transcriptomics method was developed (Paper III). Here, the barcoded microarray was used to obtain spatially resolved expression profiles from tissue sections using the same imaging principle. The mouse olfactory bulb was profiled on a whole-transcriptome scale and the results showed that the expression correlated well (r2=0.94-0.97) as compared to bulk RNA sequencing. The method was 6.9% efficient, reported signal diffusion at ~2 μm and accurately deconvoluted layer-specific transcripts in an unbiased manner. Lastly, the spatial transcriptomics concept was applied to profile human breast tumours in three dimensions (Paper IV). Unbiased clustering revealed previously un-annotated regions and classified them as parts of the immune system, providing a detailed view into complex interactions and crosstalk in the whole tissue volume. Spatial tumour classification divulged that certain parts of the tumour clearly classified as other subtypes as compared to bulk analysis providing useful data for current practice diagnostics. The last part of the thesis discusses a look towards the future, how the presented methods could be used, improved upon or combined in translational research. / <p>QC 20170109</p>

Hemorrhagic fever viruses cause severe infections characterized by a hyperactive immune response that often leads to multiorgan failure and death. Current diagnostic tests are based on detecting viral proteins and nucleic acids in the blood. These are late-stage events during infection, which makes it impossible to perform a diagnosis before they are present in the collected sample. In this thesis, I explore an alternative approach using the transcriptional changes of circulating immune cells during the early stages of infection to identify unique markers of viral infection. The main advantage of this method is that it can be used to identify highly pathogenic viruses before standard detection methods become effective. I initially used RNA sequencing data to compare the host patterns of expression of macaques infected with either Lassa virus or Marburg virus, two related hemorrhagic fever viruses. I identified a set of genes that quickly become upregulated after a viral infection and remain highly expressed throughout the entire disease course, irrespective of the specific virus that caused the infection. I was also able to identify a set of biomarker genes that follow unique patterns of expression depending on the type of infection. I used an independent dataset to validate the potential of these genes to be used as biomarkers of infection. Additionally, I compared these results to the patterns of expression of macaques infected with Ebola virus, looking at multiple experimental conditions, tissues and routes of infection. Finally, I validated the host patterns of expression using two independently generated datasets corresponding to infection by different strains of arenaviruses and filoviruses. Studying the host immune response has the potential to improve the diagnosis of viral hemorrhagic fevers and other diseases. It can also accelerate our efforts to understand the underlying molecular mechanisms that lead to pathogenesis and severe disease.

Lamprey adults may be parasitic or non-parasitic, but the genetic basis of life history type is unknown. Although external differences between types are not apparent until metamorphosis, previous studies have suggested histological differences during ovarian differentiation. This study examined potential differences between parasitic chestnut lamprey Ichthyomyzon castaneus and non-parasitic northern brook lamprey I. fossor before, during, and after ovarian differentiation, using histological examination of the gonad, quantitative reverse-transcriptase PCR, and RNA sequencing. There were no observable differences in the timing or nature of ovarian differentiation. However, there was evidence of differential expression of individual genes associated with growth, apoptosis, and fecundity and of sets of genes associated with energy and lipid metabolism. The sequence resources developed in this project will be useful for future examination of the genetic basis of lamprey life history type and of the genes controlling sex differentiation in these ancient vertebrates.

Lamprey adults may be parasitic or non-parasitic, but the genetic basis of life history type is unknown. Although external differences between types are not apparent until metamorphosis, previous studies have suggested histological differences during ovarian differentiation. This study examined potential differences between parasitic chestnut lamprey Ichthyomyzon castaneus and non-parasitic northern brook lamprey I. fossor before, during, and after ovarian differentiation, using histological examination of the gonad, quantitative reverse-transcriptase PCR, and RNA sequencing. There were no observable differences in the timing or nature of ovarian differentiation. However, there was evidence of differential expression of individual genes associated with growth, apoptosis, and fecundity and of sets of genes associated with energy and lipid metabolism. The sequence resources developed in this project will be useful for future examination of the genetic basis of lamprey life history type and of the genes controlling sex differentiation in these ancient vertebrates.

Important physiological changes occur during lactation to allow for nourishment of the offspring. Specific neuronal groups within the arcuate nucleus of the hypothalamus influence prolactin (PRL) secretion, metabolism and fertility during lactation. Our overall goal was to identify gene expression changes in the arcuate nucleus during lactation and examine the roles of PRL and ovarian hormones in regulating expression of select genes. We evaluated transcriptome changes in the arcuate nucleus during lactation using RNA-sequencing. Thirty-seven differentially expressed genes, including neuropeptides, signaling molecules, receptors and enzymes, were identified between suckled and pup-deprived groups. Selected genes were evaluated by qRT-PCR in ovary-intact and ovariectomized lactating models, which included non-lactating, suckled and 24hr pup-deprived lactating groups. The mRNA expression of tyrosine hydroxylase (Th), kisspeptin (Kiss1), and neurokinin B (Tac3) was decreased, whereas mRNA expression of proenkephalin (Penk), parathyroid hormone 2 receptor (Pth2r), insulin-like growth factor binding protein 3 (Igfbp3), membrane progesterone receptor beta (Paqr8), suppressor of cytokine signaling 2 (Socs2) and cytokine-inducible SH2 domain-containing protein (Cish) was increased in suckled lactating rats. In 24hr pup-deprived dams, mRNA expression of Pth2r, Igfbp3, Paqr8, Socs2 and Cish was decreased and Th was increased, as compared to suckled rats. The mRNA expression of Kiss1 and Tac3 was increased and Penk was decreased after 72hr, but not 24hr, pup deprivation suggesting gene expression of these neuropeptides is slow to return to non-lactating levels after removing the suckling stimulus. Tyrosine hydroxylase (TH) protein and enkephalin (ENK) peptide expression was examined by immunohistochemistry. Lactating rats had increased ENK in the median eminence and decreased TH in the median eminence and arcuate nucleus as compared to virgin ovariectomized rats. ENK co-localization with TH in the arcuate nucleus was more predominant in lactating rats. Penk, Igfbp3, Pth2r, Cish, and Socs2 mRNA expression was decreased after 72hr bromocriptine treatment in suckled rats, suggesting that these genes are PRL-regulated. In contrast, gene expression of Th, Tac3 and Kiss1 were increased and Paqr8 was decreased with 72hr pup-deprivation, but expression of these genes, were not altered with bromocriptine treatment, indicating that these genes are regulated by a non-PRL component of the suckling stimulus. The mRNA expression of Kiss1, Socs2 and Igfbp3 was increased and Penk was decreased in ovariectomized as compared to ovary intact lactating rats, suggesting that ovarian hormones influence the expression of these genes during lactation. Our data show gene expression changes in the arcuate nucleus that may contribute to increased PRL secretion (Th, Penk and Pth2r), decreased PRL receptor signaling (Cis and Socs2), reduced fertility (Kiss1 and Tac3), increased metabolism (Igfbp3) and support a role for progesterone membrane actions (Paqr8). The expression of some genes appeared to be selectively regulated by ovarian hormone input and/or PRL feedback.

Establishing a definitive cell cycle progression has been one of the fundamental aims of cellular biology. Its importance lies in gaining insight into the basic processes of life as well as the functions of mutant cell cycle pathways in promoting cancer by replication deficiencies and loss of checkpoint control. Currently used methods to control cell cycle and synchronize cells, function by halting cell cycle progression. Such harsh methods are detrimental to the cell and insufficient to provide an accurate reflection of the cell cycle. This study focused on replicating and confirming the efficiency of a technique developed by Helmstetter, called the “Baby Machine,” that can produce new born cells with little to no perturbations. Using this in conjunction with a short pulse RNA labelling technique, called Bru-seq, allowed the capture and RNA sequencing of synchronized cells and its nascent RNA. Here we show the first glimpse into the transcriptional profile of newly divided cells as well as novel rapid exon splicing and transcription read-through processes.