Search

Most of the human genome does not encode proteins but instead contains a vast array of non-coding sequences that play crucial roles in gene regulation. Understanding these regulatory sequences, especially in neuronal contexts, is essential for uncovering brain function and development complexities. In this work, I utilized high-throughput assays alongside the mouse as a model to examine neuronal non-coding regulatory sequences. In Chapter 1, I tested thousands of candidate regulatory elements using Massively Parallel Reporter Assays (MPRA). We further validated strong candidates using mouse transgenic assays to assess the enhancer activity in vivo. Our combined approach of MPRA and mouse transgenic assays revealed complementary information on enhancer activity, highlighting the strengths and limitations of each method. In Chapter 2, I focused on functionally characterizing the regulatory network of the oxytocin receptor. The oxytocin receptor is a key regulator of social behavior. We identified seven candidate regulatory elements using comparative and functional genomics tools. We further validated the enhancer activity of the strongest candidate regulatory element using stable mouse transgenic lines. We determined the candidate regulatory element to have enhancer activity in the mouse olfactory bulb at post-developmental stages. This comprehensive study underscores the intricate regulatory landscapes that govern neuronal functions and showcases the power of integrating high-throughput screening with in vivo validation to unravel these complexities.

Using machine learning (ML), we interrogated the function of all human-chimpanzee variants in 2,645 human accelerated regions (HARs), finding 43% of HARs have variants with large opposing effects on chromatin state and 14% on neurodevelopmental enhancer activity. This pattern, consistent with compensatory evolution, was confirmed using massively parallel reporter assays in chimpanzee and human neural progenitor cells. The species-specific enhancer activity of HARs was accurately predicted from the presence and absence of transcription factor footprints in each species. Despite these striking cis effects, activity of a given HAR sequence was nearly identical in human and chimpanzee cells. This suggests that HARs did not evolve to compensate for changes in the trans environment but instead altered their ability to bind factors present in both species. Thus, ML prioritized variants with functional effects on human neurodevelopment and revealed an unexpected reason why HARs may have evolved so rapidly.

The notochord is a defining feature of all chordates. The transcription factors Zic and ETS regulate enhancer activity within the notochord. We conduct high-throughput screens of genomic elements within developing Ciona embryos to understand how Zic and ETS sites encode notochord activity. Our screen discovers an enhancer located near Lama, a gene critical for notochord development. Reversing the orientation of an ETS site within this enhancer abolishes expression, indicating that enhancer grammar is critical for notochord activity. Similarly organized clusters of Zic and ETS sites occur within mouse and human Lama1 introns. Within a Brachyury (Bra) enhancer, FoxA and Bra, in combination with Zic and ETS binding sites, are necessary and sufficient for notochord expression. This binding site logic also occurs within other Ciona and vertebrate Bra enhancers. Collectively, this study uncovers the importance of grammar within notochord enhancers and discovers signatures of enhancer logic and grammar conserved across chordates.

The identification of causal genetic variants underlying human diseases and traits remains a major challenge in genomics. New progress towards this goal has been made possible by advancing high-throughput technologies in biology and the subsequent collection of large biological datasets. In my thesis work I use both large-scale experimental and computational approaches to identify genetic variants that alter the process of RNA splicing. I also identified a unique opportunity to utilize large-scale RNA-Seq data for variant discovery, particularly focusing on an understudied class of splicing events known as microexons. Microexons are extremely short exons that pose unique challenges for quantification. To address this, I developed an optimized computational pipeline described in Chapter 2. This pipeline was rigorously validated against long-read sequencing data to ensure its reliability. In Chapter 3, the functional importance of microexons is explored through analysis of multi-tissue RNA-Seq data sourced from the GTEx consortium. My analysis identified thousands of highly conserved microexons. These microexons align well with recent evidence indicating that disruption in microexon splicing is associated with neurological and muscular diseases. Finally in chapter 4, I expand on work using a massively parallel reporter assay (MPRA) to directly test thousands of sequence variants for their effect on splicing. Through a data science-driven approach, this thesis advances our understanding of the intricate landscape of splice-disrupting variants, producing new tools as well as novel insights into the functional importance of microexons in human biology and disease.

High blood pressure (BP) is the major risk factor for cardiovascular disease. Genome-wide association studies have identified genetic variants for BP, but functional insights into causality and related molecular mechanisms lag behind. We functionally characterize 4,608 genetic variants in linkage with 135 BP loci in vascular smooth muscle cells and cardiomyocytes by massively parallel reporter assays. High densities of regulatory variants at BP loci (i.e., ULK4, MAP4, CFDP1, PDE5A) indicate that multiple variants drive genetic association. Regulatory variants are enriched in repeats, alter cardiovascular-related transcription factor motifs, and spatially converge with genes controlling specific cardiovascular pathways. Using heuristic scoring, we define likely causal variants, and CRISPR prime editing finally determines causal variants for KCNK9, SFXN2, and PCGF6, which are candidates for developing high BP. Our systems-level approach provides a catalog of functionally relevant variants and their genomic architecture in two trait-relevant cell lines for a better understanding of BP gene regulation., We thank the Melé and Maass labs for intellectual input, Dr. Steven Erwood from Dr. Ronald Cohn’s lab for guidance in applying CRISPR prime editing, The Centre for Applied Genomics, The Structural & Biophysical Core Facility, and The Imaging Facility, The Hospital for Sick Children, Toronto, Canada, for assistance with high-throughput sequencing, luminescence detection, and imaging. We thank Dovetail Genomics, LLC, 100 Enterprise Way, Suite A101, Scotts Valley, CA 95066, USA, for generating Omni-C libraries and for the collaborative support throughout the project. W.O. was supported by a Fundació la Marató grant (ref. 321/C/2019), K.K. was supported by an OGS fellowship, J.W.L.B. was supported by a CGS-D fellowship, and D.F.L. was supported by an Ontario Genomics-CANSSI Ontario Postdoctoral Fellowship in Genome Data Science. This project was supported by Canada’s New Frontiers in Research Fund (NFRFE-2018-01305), the Canadian Institutes of Health Research (CIHR PJT 173542 [P.G.M.] and PJT 175034 [S.M., J.E.]), CIHR ENP 161429 under the frame of ERA PerMed (S.M.), the Ted Rogers Centre for Heart Research (S.M., J.E.), and the Heart and Stroke Foundation of Canada. J.E. holds a Canada Research Chair Tier 1 in Stem Cell Models of Childhood Disease, S.M. holds the Heart and Stroke Foundation of Canada & Robert M. Freedom Chair in Cardiovascular Science, M. Melé was supported by a Ramon y Cajal fellowship (RYC-2017-22249), and P.G.M. holds a Canada Research Chair Tier 2 in Non-coding Disease Mechanisms., Peer Reviewed, "Article signat per 14 autors/es: Winona Oliveros, Kate Delfosse, Daniella F. Lato , Katerina Kiriakopulos, Milad Mokhtaridoost, Abdelrahman Said, Brandon J. McMurray, Jared W.L. Browning, Kaia Mattioli, Guoliang Meng, James Ellis, Seema Mital, Marta Melé, Philipp G. Maass", Postprint (published version)

AU-rich elements (AREs) are 3' UTR cis-regulatory elements that regulate the stability of mRNAs. Consensus ARE motifs have been determined, but little is known about how differences in 3' UTR sequences that conform to these motifs affect their function. Here, we use functional annotation of sequences from 3' UTRs (fast-UTR), a massively parallel reporter assay (MPRA), to investigate the effects of 41,288 3' UTR sequence fragments from 4653 transcripts on gene expression and mRNA stability in Jurkat and Beas2B cells. Our analyses demonstrate that the length of an ARE and its registration (the first and last nucleotides of the repeating ARE motif) have significant effects on gene expression and stability. Based on this finding, we propose improved ARE classification and concomitant methods to categorize and predict the effect of AREs on gene expression and stability. Finally, to investigate the advantages of our general experimental design we examine other motifs including constitutive decay elements (CDEs), where we show that the length of the CDE stem-loop has a significant impact on steady-state expression and mRNA stability. We conclude that fast-UTR, in conjunction with our analytical approach, can produce improved yet simple sequence-based rules for predicting the activity of human 3' UTRs.

cis-regulatory elements (CREs) are non-coding segments of the genome which regulate the transcription of nearby genes. They can be broadly divided to two categories: 1) promoters, positioned directly upstream of their target gene, and 2) enhancers, positioned distally to their target gene. Enhancers are thought to be the main drivers of cell type-specific and state-specific transcription, and regulate gene expression by fine-tuning the rate of transcription, as opposed to the more binary (on or off) regulatory function that promoters typically have. Understanding how enhancers function is therefore crucially important to understanding how cells obtain and maintain certain fates and determine response to stimuli. Despite their importance, much is still unknown about the roles enhancers play in many biological processes, and how their sequence determines their regulatory function.The first part of this dissertation deals with single-cell chromatin accessibility data (e.g as produced by single-cell ATAC-seq) as a means for systemically studying heterogeneity of CREs, and specifically enhancers. In chapter 2 this is demonstrated in the innate immune system's response to vaccination: in a subset of cells, a distinct state of chromatin accessibility maintains long-term epigenetic changes that prime these cells to a different response to stimuli, and provides non-specific viral protection.However promising, the unique properties of this data modality poses significant challenges. These are addressed in chapter 3, which introduced PeakVI, a deep generative model that provides a comprehensive statistical framework for analyzing data generated by scATAC-seq assays. Recent advances in sequencing technologies now enable obtaining these measurements alongside gene expression measurements (i.e single cell RNA-seq), providing the ability to directly measure the relationship between the heterogeneity of the chromatin landscape and that of the transcriptional profile. Chapter 4 intr

AU-rich elements (AREs) are 3' UTR cis-regulatory elements that regulate the stability of mRNAs. Consensus ARE motifs have been determined, but little is known about how differences in 3' UTR sequences that conform to these motifs affect their function. Here, we use functional annotation of sequences from 3' UTRs (fast-UTR), a massively parallel reporter assay (MPRA), to investigate the effects of 41,288 3' UTR sequence fragments from 4653 transcripts on gene expression and mRNA stability in Jurkat and Beas2B cells. Our analyses demonstrate that the length of an ARE and its registration (the first and last nucleotides of the repeating ARE motif) have significant effects on gene expression and stability. Based on this finding, we propose improved ARE classification and concomitant methods to categorize and predict the effect of AREs on gene expression and stability. Finally, to investigate the advantages of our general experimental design we examine other motifs including constitutive decay elements (CDEs), where we show that the length of the CDE stem-loop has a significant impact on steady-state expression and mRNA stability. We conclude that fast-UTR, in conjunction with our analytical approach, can produce improved yet simple sequence-based rules for predicting the activity of human 3' UTRs.

The Neanderthal and Denisovan genomes enabled the discovery of sequences that differ between modern and archaic humans, the majority of which are noncoding. However, our understanding of the regulatory consequences of these differences remains limited, in part due to the decay of regulatory marks in ancient samples. Here, we used a massively parallel reporter assay in embryonic stem cells, neural progenitor cells, and bone osteoblasts to investigate the regulatory effects of the 14,042 single-nucleotide modern human-specific variants. Overall, 1791 (13%) of sequences containing these variants showed active regulatory activity, and 407 (23%) of these drove differential expression between human groups. Differentially active sequences were associated with divergent transcription factor binding motifs, and with genes enriched for vocal tract and brain anatomy and function. This work provides insight into the regulatory function of variants that emerged along the modern human lineage and the recent evolution of human gene expression.

The widespread adoption of genome-wide association studies (GWAS) has revolutionized the detection of genetic loci associated with complex traits. However, the majority of common susceptibility loci reside in poorly annotated noncoding genomic regions and are composed of many correlated polymorphisms due to linkage disequilibrium, obscuring identification of the causal variants and mechanisms underlying trait association. Thus, the functional annotation of noncoding variation is a major impediment to interpretation of genetic risk. Massively Parallel Reporter Assays (MPRA) are a novel experimental approach for the high-throughput functional characterization of noncoding genetic variation, yet remain to be systematically applied to any neurologic disorder. In this dissertation, I utilize MPRA to characterize variation associated with two neurodegenerative disorders that share tau-protein neuropathology, Alzheimer’s disease and Progressive Supranuclear Palsy. First, I describe the design and implementation of an MPRA to screen 5,706 noncoding variants derived from three GWAS for AD and PSP, identifying 320 regulatory polymorphisms comprising 27 of 34 tested loci. These results enable subsequent identification of novel putative risk genes including PLEKHM1 and APOC1 distributed across the complex 17q21.31 and 19q13.32 regions. In Chapter 3, I show that functional predictions from four popular computational algorithms for variant prioritization are discordant both with MPRA results and each other. In Chapter 4, I find that MPRA-defined functional variants preferentially disrupt predicted transcription factor binding sites that converge on enhancers with differential cell-type specific activity in PSP and AD, implicating a neuronal SP1-driven regulatory network in PSP pathogenesis. These analyses support a novel mechanism underlying noncoding genetic risk, whereby common genetic variants drive disease risk via their aggregate activity on specific transcriptional programs

Enhancers are cis-regulatory elements that play critical regulatory roles in modulating developmental transcription programs and driving cell-type-specific and context-dependent gene expression in the brain. The development of massively parallel reporter assays (MPRAs) has enabled high-throughput functional screening of candidate DNA sequences for enhancer activity. Tissue-specific screening of in vivo enhancer function at scale has the potential to greatly expand our understanding of the role of non-coding sequences in development, evolution, and disease. Here, we adapted a self-transcribing regulatory element MPRA strategy for delivery to early postnatal mouse brain via recombinant adeno-associated virus (rAAV). We identified and validated putative enhancers capable of driving reporter gene expression in mouse forebrain, including regulatory elements within an intronic CACNA1C linkage disequilibrium block associated with risk in neuropsychiatric disorder genetic studies. Paired screening and single enhancer in vivo functional testing, as we show here, represents a powerful approach towards characterizing regulatory activity of enhancers and understanding how enhancer sequences organize gene expression in the brain.

The widespread adoption of genome-wide association studies (GWAS) has revolutionized the detection of genetic loci associated with complex traits. However, the majority of common susceptibility loci reside in poorly annotated noncoding genomic regions and are composed of many correlated polymorphisms due to linkage disequilibrium, obscuring identification of the causal variants and mechanisms underlying trait association. Thus, the functional annotation of noncoding variation is a major impediment to interpretation of genetic risk. Massively Parallel Reporter Assays (MPRA) are a novel experimental approach for the high-throughput functional characterization of noncoding genetic variation, yet remain to be systematically applied to any neurologic disorder. In this dissertation, I utilize MPRA to characterize variation associated with two neurodegenerative disorders that share tau-protein neuropathology, Alzheimer’s disease and Progressive Supranuclear Palsy. First, I describe the design and implementation of an MPRA to screen 5,706 noncoding variants derived from three GWAS for AD and PSP, identifying 320 regulatory polymorphisms comprising 27 of 34 tested loci. These results enable subsequent identification of novel putative risk genes including PLEKHM1 and APOC1 distributed across the complex 17q21.31 and 19q13.32 regions. In Chapter 3, I show that functional predictions from four popular computational algorithms for variant prioritization are discordant both with MPRA results and each other. In Chapter 4, I find that MPRA-defined functional variants preferentially disrupt predicted transcription factor binding sites that converge on enhancers with differential cell-type specific activity in PSP and AD, implicating a neuronal SP1-driven regulatory network in PSP pathogenesis. These analyses support a novel mechanism underlying noncoding genetic risk, whereby common genetic variants drive disease risk via their aggregate activity on specific transcriptional programs

The Neanderthal and Denisovan genomes enabled the discovery of sequences that differ between modern and archaic humans, the majority of which are noncoding. However, our understanding of the regulatory consequences of these differences remains limited, in part due to the decay of regulatory marks in ancient samples. Here, we used a massively parallel reporter assay in embryonic stem cells, neural progenitor cells, and bone osteoblasts to investigate the regulatory effects of the 14,042 single-nucleotide modern human-specific variants. Overall, 1791 (13%) of sequences containing these variants showed active regulatory activity, and 407 (23%) of these drove differential expression between human groups. Differentially active sequences were associated with divergent transcription factor binding motifs, and with genes enriched for vocal tract and brain anatomy and function. This work provides insight into the regulatory function of variants that emerged along the modern human lineage and the recent evolution of human gene expression.

Enhancers are cis-regulatory elements that play critical regulatory roles in modulating developmental transcription programs and driving cell-type-specific and context-dependent gene expression in the brain. The development of massively parallel reporter assays (MPRAs) has enabled high-throughput functional screening of candidate DNA sequences for enhancer activity. Tissue-specific screening of in vivo enhancer function at scale has the potential to greatly expand our understanding of the role of non-coding sequences in development, evolution, and disease. Here, we adapted a self-transcribing regulatory element MPRA strategy for delivery to early postnatal mouse brain via recombinant adeno-associated virus (rAAV). We identified and validated putative enhancers capable of driving reporter gene expression in mouse forebrain, including regulatory elements within an intronic CACNA1C linkage disequilibrium block associated with risk in neuropsychiatric disorder genetic studies. Paired screening and single enhancer in vivo functional testing, as we show here, represents a powerful approach towards characterizing regulatory activity of enhancers and understanding how enhancer sequences organize gene expression in the brain.

To realize the promise of CRISPR-Cas9-based genetics, approaches are needed to quantify a specific, molecular phenotype across genome-wide libraries of genetic perturbations. We addressed this challenge by profiling transcriptional, translational, and posttranslational reporters using CRISPR interference (CRISPRi) with barcoded expression reporter sequencing (CiBER-seq). Our barcoding approach allowed us to connect an entire library of guides to their individual phenotypic consequences using pooled sequencing. CiBER-seq profiling fully recapitulated the integrated stress response (ISR) pathway in yeast. Genetic perturbations causing uncharged transfer RNA (tRNA) accumulation activated ISR reporter transcription. Notably, tRNA insufficiency also activated the reporter, independent of the uncharged tRNA sensor. By uncovering alternate triggers for ISR activation, we illustrate how precise, comprehensive CiBER-seq profiling provides a powerful and broadly applicable tool for dissecting genetic networks.

The integrative analysis of high-throughput reporter assays, machine learning, and profiles of epigenomic chromatin state in a broad array of cells and tissues has the potential to significantly improve our understanding of noncoding regulatory element function and its contribution to human disease. Here, we report results from the CAGI 5 regulation saturation challenge where participants were asked to predict the impact of nucleotide substitution at every base pair within five disease-associated human enhancers and nine disease-associated promoters. A library of mutations covering all bases was generated by saturation mutagenesis and altered activity was assessed in a massively parallel reporter assay (MPRA) in relevant cell lines. Reporter expression was measured relative to plasmid DNA to determine the impact of variants. The challenge was to predict the functional effects of variants on reporter expression. Comparative analysis of the full range of submitted prediction results identifies the most successful models of transcription factor binding sites, machine learning algorithms, and ways to choose among or incorporate diverse datatypes and cell-types for training computational models. These results have the potential to improve the design of future studies on more diverse sets of regulatory elements and aid the interpretation of disease-associated genetic variation.