Multi-omics reveals abundant mitochondrial genomic and circadian expression variation in the Drosophila Genetic Reference Panel

Modern biology rapidly generates a wealth of data,which can only be analysed computationally.Despite the variety of available data types,most of them aim to provide an answer to one question: how does genotypic diversity translates into phenotypes? This question became central for this thesis where the computational analysis,via integration of genomics and transcriptomics,became the main method to unravel the full picture.Two biological processes were examined with this question in mind: mitochondrial genomics and circadian genomics and transcriptomics.First,to uncover the relationship between mitochondrial genetics and phenotypes,we produced a high quality catalogue of mitochondrial variants across 169 Drosophila Genetics Reference Panel (DGRP) lines.The high coverage and quality of the variant calls was achieved via developed mitochondrial DNA (mtDNA)-enriching method.The reliability of variant detection was supported by the low contamination rate of nuclear mitochondrial DNA fragments detected de-novo by a computational method developed in this thesis.Within the coding part of the mtDNA,231 variants were detected,most of them being SNPs.Based on the computed mitochondrial haplotypes,we revealed population structure despite common perception of its absence within the DGRP.Haplotype-phenotype associations with publicly available phenotypes showed a significant association with food intake in males which was further confirmed experimentally.At the molecular level,haplotypes did not reveal a strong association with mitochondrial protein complex quantity or activity.However,the activity showed greater variability between genotypes than haplotypes,therefore indicating a possible buffering effect.Furthermore,numerous mito-nuclear genome incompatibilities were detected computationally,but didn't affect tested phenotypes.Secondly,to determine the diversity of the circadian transcriptome across tissues and the influence of genetic diversity thereon,a vast dataset of 778 transcriptomes was produced.The first part of the dataset was an RNA-seq time series across four tissues of the w- reference fly strain.This dataset was used to reveal a high degree of tissue specificity of the circadian clock transcriptome,where the majority of the genes cycled restrictively in one tissue.Out of the 14 genes cycling in all tissues,7 weren't previously associated with the circadian rhythm.The evaluation of novel genes through the activity of flies with gene knockdowns revealed their functional relevance.Next,a cross tissue gene regulatory network indicated on a potential mechanism of tissue specific circadian expression: it is achieved through collaboration of core circadian transcription factors,such as clock,with other cycling and non-cycling TFs.The second part of the dataset consisted of static tissue specific transcriptomes of 141 DGRP lines.Tissue specific physiological time was computationally estimated and revealed a high variability across the samples suggesting underlying genetic components.For one of the samples the physiological time differed >10h from the harvesting time and displayed an abolished molecular rhythm,yet preserved its rhythmic behaviour.The genetic component underlying this observation was a newly identified cry allele disrupting the light sensing pathway.As a whole,this thesis demonstrates the utility of complex multi-omics approaches in modern biology and reveals new genetic determinants for food intake and circadian rhythm defects

The phenotypic impact of naturally occurring genetic and molecular mitochondrial variation in a Drosophila Genetic Reference Population

Roel Paulus Josephus Bevers

Genome-wide association (GWA) studies aim to dissect the relationship between genotype and phenotype. So far, the focus has been on nuclear genetic determinants as variation in the mitochondrial genome is often low in the study cohorts. Despite this observation, mitochondrial genomic variants are increasingly linked to metabolic and neurological disorders. Thus, there is much to gain from studying how mitochondrial variation interacts with nuclear variation and/or environment, and how this subsequently affects phenotypes. In my thesis I address these topics using Drosophila melanogaster as a model system, and specifically the Drosophila Genetic Reference Panel (DGRP). The DGRP consists of 200 sequenced and genetically diverse Drosophila populations and has been extensively used to study a wide-range of quantitative traits. However, in all of these studies, researchers have focussed on the influence of nuclear genetic variants in the absence of a high-resolution map of mitochondrial variants. To address this problem, we have developed a high-throughput method to specifically sequence mitochondrial DNA that enabled us to characterize mitochondrial genomic variation to an unprecedented depth. On average, each DGRP line had 25 mitochondrial variants (1 in 600 bp) and that most of the variants detected were unique for each line. However, we also found that some variants are more prevalent and show they can be used to construct mitochondrial haplotypes. We then used these haplotypes to test whether mitochondrial variation affected previously published phenotypes. We detected significant associations with 12 out of 259 publicly accessible phenotypes and found that these were mostly related to metabolism and responses to environmental cues. We selected one of the significant phenotypes, food intake in males, to verify our findings. We demonstrated the effect of mitochondrial haplotypes on the phenotype by swapping mitochondria from ¿high food intake¿ lines with mitochondria from ¿low food intake¿ lines and vice versa. We observed that, largely independent of the nuclear background, DGRP lines with a ¿high food intake mitochondria¿ consumed more food and concomitantly, DGRP lines with the ¿low food intake mitochondria¿ consumed less food. Furthermore, we investigated the influence of natural genetic variation was on the response to a high-fat diet. We designed a control and a high-fat diet and fed this to >100 DGRP lines and measured the lifespan on each diet. On average, lifespan was decreased by 48% on a high-fat diet, however this ranged from an increase of 2% to a decrease of 60% accentuating the complexity of the underlying genetic architecture. Our dataset was relatively small and limited our genome-wide and mitochondrial haplotype association analyses. Nonetheless, our GWA analysis detected variants related to genes involved in lipid metabolism and mitochondria but we did not detect significant associations with mitochondrial haplotypes. This suggests that while mitochondrial function may be important for the response to a high-fat diet, it is not dependent on the mitochondrial haplotype. Taken together, the work described in my thesis demonstrates the significance of accounting for mitochondrial haplotypes in phenotype association analyses. By doing so, it can open up new insights into genotype-to-phenotype relationships and gene regulation.

EPFL2018

Computational analysis of ultraconserved non-coding DNA elements in vertebrate genomes

Slavica Dimitrieva Janeva

Recent advancements in DNA sequencing technologies, accompanied by parallel development of sophisticated computational methods, offer an unprecedented opportunity to explore and study the genetic material of many organisms. However, despite the tremendous progress in the field of genomics, our understanding of the genetic information encoded by the DNA is far from complete. Whole-genome comparisons of vertebrate species have uncovered the existence of non-coding DNA sequences that exhibit exceptionally high levels of similarity over hundreds of base pairs across distant species, referred to as ultraconserved non-coding elements (UCNEs). The existence of such sequences represents one of the biggest mysteries in current biology. Many UCNEs exhibit even stronger conservation than protein coding regions, but to date no molecular mechanism has been described that would require such a high degree of conservation over such long sequences. This thesis focuses on computational analysis of UCNEs across vertebrate genomes, with an ultimate goal to provide insights into the functioning of these elements and to unravel the reasons for their extreme conservation. Using computational approaches we studied the salient characteristics of UCNEs, and explored their genomic environment and their organization across genomes. Like previous studies, we observed that UCNEs are organized as large clusters around essential developmental genes, forming genomic regulatory blocks. Then we sought to understand the reasons behind this strong clustering of UCNEs. The clustering could reflect functional cooperativity between UCNEs. Alternatively, if each UCNE acts independently, their high concentration near developmental genes could merely reflect the extreme regulatory complexity of these genes. In a special setting of genomic context analysis, we analyzed the fate of UCNEs in teleost fish genomes that were subjected to an additional round of whole-genome duplication. We found that in most cases all UCNEs of a block were retained in one copy only, but together on the same chromosome. Conversely, the corresponding target genes were often retained in two copies, one completely devoid of UCNEs. Our results suggested that UCNEs of a cluster function in a highly cooperative manner. We propose that a multitude of cooperative cis-interactions between UCNEs is the reason for their extreme sequence conservation. Our work presents a novel hypothesis about the reasons for UCNE conservation and their mode of action, which is yet to be confirmed experimentally. The results from our study are made accessible through a new web resource called UCNEbase (http://ccg.vital-it.ch/UCNEbase). UCNEbase provides informa- tion about the genomic organization of UCNEs and their association with developmental regulatory genes, and enables all data to be explored in a genome browser environment. As part of a side project, we developed a new sparsified version of algorithms for computational prediction of RNA secondary structures and introduced a new program sibRNAfold. We performed a thorough analysis of the time complexity of sparsified RNA folding algorithms covering a large parameter space and empirically demonstrated that sparse RNA folding has cubic time complexity rather than quadratic, as claimed previously.

EPFL2013

Rhythmic and clock-dependent chromatin loops regulate circadian gene expression

Jérôme Mermet

In mammals the circadian clock drives daily behavioural and physiological changes that resonate with environmental cues, which can be observed, for example, in the intricate timing of rest during the night and activity during the day in humans. The circadian clock consists of a network of hierarchical clocks in which the central pacemaker is located in the suprachiasmatic nucleus, which is sensitive to external light and propagates time to the peripheral organ-based clocks. Within organs, clocks tick in each individual cell and are molecularly encoded in the genome as a network of dynamic feedback loops. By mediating the expression of the appropriate gene products at the correct moment of the day, the molecular clocks op-timize physiological functions of virtually every cell in our body, allowing us to synchronize with daily fluc-tuations in the environment. In mammalian cells the chromatin organization in the nucleus consists of a topological network, in which chromatin interactions between distal regulatory elements such as enhancers and target genes are essen-tial to regulate gene expression. Important questions remain: is the complex chromatin folding dynamic and if so on which genomic and temporal scales? In fact, dynamic rearrangements of the chromatin have been reported on multiple genomic and time scales, from the entire genome to enhancer-promoter con-tacts, across developmental transitions, cellular differentiation, and transcription cycles. With this per-spective in mind, the circadian oscillator provides a unique model to study the dynamics of genomic inter-actions in the context of fluctuating transcription. Here, we explored chromatin contacts of core-clock and clock-controlled gene promoters in the mouse. We aimed at evaluating the putative dynamics of chromatin contacts, at estimating their conservation across tissues, and at exploring the contribution of the circadian clock to this interaction network. To achieve this goal, we performed 4C-sequencing assays at two circadian time points in the liver and kidney of WT and clock-deficient animals. Overall, we observe that chromatin contacts are largely conserved across circadian time points and genetic backgrounds, and are tissue-specific. Our experiments reveal that the promoters of core-clock and clock-controlled genes contact distal genomic regions containing en-hancer chromatin signatures, suggesting a functional role of these distal sites in target gene regulation. Our data also suggest a clock-driven module of rhythmic genes that are colocalized in the nucleus. However, we also observe dynamic contacts between oscillating gene promoter and enhancer-like ge-nomic regions. In this thesis, we discuss in detail two examples of dynamic chromatin looping that are ob-served at the Cry1 and Gys2 loci. Our data reveal not only a dynamic coupling between the chromatin loop and the transcriptional state but also that the dynamic of chromatin looping depends on a functional clock. Furthermore, genome editing experiments using CRISPR-Cas9 tools reveal that the Cry1 distal site is essential for the transcriptional regulation of Cry1 and contributes to the circadian clock robustness both in-vivo and in cultured cells. These results demonstrate how local chromatin rearrangements take place in a 24-hour time-scale and set chromatin looping between gene promoters and distal regions as a new reg-ulatory layer for circadian gene expression.

EPFL2017