It is a microbial world we live in: microbes outnumber other organisms by several orders of magnitude, and they have great importance for the environment. However, environmental microbes are notoriously difficult to grow in the laboratory, and using culture independent techniques is necessary to expand our view. In this thesis, I apply metagenomics and single-cell genomics to environmental samples from ancient human remains and lakes. First, I used metagenomics to learn about bacteria from a Neanderthal’s bone and the gut of Ötzi, a frozen natural mummy. Both were exploratory studies where the main question was what kind of bacteria are present. I found out that Streptomyces dominated this particular Neanderthal fossil, and the DNA lacked the damage that is often seen in ancient samples. Ötzi's gut sample was dominated by Clostridia and fungi belonging to Basidiomycota. Second, ten single-cell amplified genomes of freshwater Alphaproteobacterium LD12 and three metagenomes from Swedish lakes were sequenced. Comparative metagenomics allowed hypothesizing about which functions are important for microbe proliferation in freshwater, pointing to osmoregulation and transport proteins and, possibly, to different strategies of metabolizing sugars. I also focused on SAR11 sister-groups in oceans and lakes. Phylogenies and sequence evolutionary distance estimates indicated the existence of microclusters within LD12, showing variation in abundance between lakes. The most striking difference was the relative amount of recombination compared to mutation, the estimated r/m ratio. SAR11 marine and their freshwater cousins are found at the opposite extremes of the r/m range, lowest and highest, respectively. The genetic background or sequence diversity did not explain the observed dramatic difference, so it is possibly connected to environmental adaptation or population dynamics. In addition, I have spent a substantial amount of effort benchmarking available metagenomic methods, for example fragment recruitment of metagenomes to reference genomes. In conclusion, my exploratory metagenomic studies have shed some light on the bacteria present in ancient human remains; comparative metagenomics has suggested the importance of substrate preference on functional differences between lakes and oceans; finally, single-cell genomes have allowed some insight into molecular evolutionary processes taking place in the freshwater LD12 bacterium.

Denna projektrapport är avsedd att vara ett hjälpmedel för SiCell, en del av SciLifeLab Uppsala som ska bli Europas första plattform för enkelcellgenomik till hösten 2013. SiCell har bett projektgruppen om undersökningar gällande ett Laboratory Information Management System, LIMS. På svenska ett datahanteringsystem för laboratorier. Ett sådant system skulle kunna effektivisera SiCells verksamhet. Undersökningarna har resulterat i en kravspecifikation som ett LIMS för SiCell ska uppfylla och en sammanställning av tillgängliga mjukvaror som bäst uppfyller dessa krav. Screensaver, MISO och Gnomex är de tre gratisprogram med öppen källkod som hamnade högst upp i listan. Inget av dem uppfyller alla krav men med modifieringar av programmerare tros detta vara möjligt. SiCell bad också om lägre prioriterade undersökningar av några av de metoder som används inom plattformen. Cellysering, Alternativa amplifieringsmetoder och transkriptomik har undersökts av projektgruppen. Detta resulterade i en sammanställning av vilka alternativ som finns och vad som är under utveckling inom respektive område.

A fundamental challenge in biology is to understand the complex gene regulatory networks which control tissue development in the mammalian embryo, and maintain homoeostasis in the adult. The cell fate decisions underlying these processes are ultimately made at the level of individual cells. Recent experimental advances in biology allow researchers to obtain gene expression profiles at single-cell resolution over thousands of cells at once. These single-cell measurements provide snapshots of the states of the cells that make up a tissue, instead of the population-level averages provided by conventional high-throughput experiments. The aim of this PhD was to investigate the possibility of using this new high resolution data to reconstruct mechanistic computational models of gene regulatory networks. In this thesis I introduce the idea of viewing single-cell gene expression profiles as states of an asynchronous Boolean network, and frame model inference as the problem of reconstructing a Boolean network from its state space. I then give a scalable algorithm to solve this synthesis problem. In order to achieve scalability, this algorithm works in a modular way, treating different aspects of a graph data structure separately before encoding the search for logical rules as Boolean satisfiability problems to be dispatched to a SAT solver. Together with experimental collaborators, I applied this method to understanding the process of early blood development in the embryo, which is poorly understood due to the small number of cells present at this stage. The emergence of blood from Flk1+ mesoderm was studied by single cell expression analysis of 3934 cells at four sequential developmental time points. A mechanistic model recapitulating blood development was reconstructed from this data set, which was consistent with known biology and the bifurcation of blood and endothelium. Several model predictions were validated experimentally, demonstrating that HoxB4 and Sox17 directly regulate the haematopoietic factor Erg, and that Sox7 blocks primitive erythroid development. A general-purpose graphical tool was then developed based on this algorithm, which can be used by biological researchers as new single-cell data sets become available. This tool can deploy computations to the cloud in order to scale up larger high-throughput data sets. The results in this thesis demonstrate that single-cell analysis of a developing organ coupled with computational approaches can reveal the gene regulatory networks that underpin organogenesis. Rapid technological advances in our ability to perform single-cell profiling suggest that my tool will be applicable to other organ systems and may inform the development of improved cellular programming strategies.

No description available.

Deep-sea anoxic brine pools are one of the most extreme ecosystems on Earth, which are characterized by drastic changes in salinity, temperature, and oxygen concentration. The interface between the brine and overlaying seawater represents a boundary of oxic-anoxic layer and a steep gradient of redox potential that would initiate favorable conditions for divergent metabolic activities, mainly methanogenesis and sulfate reduction. This study aimed to investigate the diversity of Bacteria, particularly sulfate-reducing communities, and their ecological roles in the interfaces of five geochemically distinct brine pools in the Red Sea. Performing a comprehensive study would enable us to understand the significant role of the microbial groups in local geochemical cycles. Therefore, we combined culture-dependent approach and molecular methods, such as 454 pyrosequencing of 16S rRNA gene, phylogenetic analysis of functional marker gene encoding for the alpha subunits of dissimilatory sulfite reductase (dsrA), and single-cell genomic analysis to address these issues. Community analysis based on 16S rRNA gene sequences demonstrated high bacterial diversity and domination of Bacteria over Archaea in most locations. In the hot and multilayered Atlantis II Deep, the bacterial communities were stratified and hardly overlapped. Meanwhile in the colder brine pools, sulfatereducing Deltaproteobacteria were the most prominent bacterial groups inhabiting the interfaces. Corresponding to the bacterial community profile, the analysis of dsrA gene sequences revealed collectively high diversity of sulfate-reducing communities. Desulfatiglans-like dsrA was the prevalent group and conserved across the Red Sea brine pools. In addition to the molecular studies, more than thirty bacterial strains were successfully isolated and remarkably were found to be cytotoxic against the cancer cell lines. However, none of them were sulfate reducers. Thus, a single-cell genomic analysis was used to study the metabolism of uncultured phyla without having them in culture. We analysed ten single-cell amplified genomes (SAGs) of the uncultivated euryarchaeal Marine Benthic Group E (MBGE), which contain a key enzyme for sulfate reduction. The results showed the possibility of MBGE to grow autotrophically only with carbon dioxide and hydrogen. In the absence of adenosine 5’-phosphosulfate reductase, we hypothesized that MBGE perform sulfite reduction rather than sulfate reduction to conserve energy.

The self-organisation of cells into complex tissues relies on the tight regulation of molecular processes governing their behaviour. Understanding these processes is a central questions in cell biology. In recent years, technological breakthroughs in single-cell sequencing experiments have enabled us to probe these processes with unprecedented molecular detail. However, biological function relies on collective processes on the mesoscopic and macroscopic scale, which do not necessarily obey the rules that govern it on the microscopic scale. Insights from these experiments on how collective processes determine cellular behaviour consequently remain severely limited. Methods from nonequilibrium statistical physics provide a rigorous framework to connect microscopic measurements to their mesoscopic or macroscopic consequences. In this thesis, by combining for the first time the possibilities of single-cell technologies and tools from nonequilbrium statistical physics, we develop theoretical frameworks that overcome these conceptual limitations. In particular, we derive a theory that maps measurements along the linear sequence of the DNA to mesoscopic processes in space and time in the cell nucleus. We demonstrate this approach in the context of the establishment of chemical modifications of the DNA (DNA methylation) during early embryonic development. Drawing on sequencing experiments both in vitro and in vivo, we find that the embryonic DNA methylome is established through the interplay between DNA methylation and 30-40 nm dynamic chromatin condensates. This interplay gives rise to hallmark scaling behaviour with an exponent of 5/2 in the time evolution of embryonic DNA methylation and time dependent, scale-free connected correlation functions, both of which are predicted by our theory. Using this theory, we successfully identify regions of the DNA that carry DNA methylation patterns anticipating cellular symmetry breaking in vivo. The primary layer determining cell identity is gene expression. However, read-outs of gene-expression profiling experiments are dominated by systematic technical noise and they do not provide “stochiometric” measurements that allow experimental data to be predicted by theories. Here, by developing effective spin glass methods, we show that the macroscopic propagation of fluctuations in the concentration of mRNA molecules gives direct information on the physical mechanisms governing cell states, independent of technical bias. We find that gene expression fluctuations may exhibit glassy behaviour such that they are long-lived and carry biological information. We demonstrate the biological relevance of glassy fluctuations by analysing single-cell RNA sequencing experiments of mouse neurogenesis. Taken together, we overcome important conceptual limitations of emerging technologies in biology and pioneer the application of methods from stochastic processes, spin glasses, field and renormalization group theories to single-cell genomics.

En dépit de leur importance, la caractérisation des communautés bactériennes dans l’environnement reste encore très incomplète. Les principales raisons sont, d’une part, la difficulté d’appréhender la totalité de la communauté bactérienne quand plus de 99% des bactéries demeurent récalcitrantes à la culture in vitro et ne peuvent donc être étudiées par les approches classiques de microbiologie. D’autre part, la métagénomique, censée contourner cette méthode de culture en s’intéressant à l’ensemble des génomes extraits des milieux d’études, demeure elle aussi imparfaite du fait de limitations techniques (biais d’extraction de l'ADN, de clonage, de PCR, de séquençage et d’assemblage des génomes etc.) et conceptuelles, inhérentes à la complexité et l’hétérogénéité des environnements. Pour compenser les limites de chacune de ces techniques, des méthodes de tri cellulaire appliquées en conjonction avec les deux premières pourraient aider à un meilleur décryptage de la diversité microbienne. Basée sur la sélection spécifique (taxonomique et/ou fonctionnelle) et l’isolement direct des cellules bactériennes ciblées à partir d’un échantillon environnemental complexe, l’étude est restreinte à une population spécifique, voire à une cellule isolée. Pourront alors être appliquées les approches classiques de mise en culture ou d’extraction de l’ADN pour une étude restreinte à l’ADN ou l’ARN, leur répétition sur les différentes populations devant à terme (lointain) approcher l’exhaustivité. C’est dans ce contexte que s’est positionné ce travail de thèse visant dans un premier temps à mettre au point un nouvel outil de tri cellulaire basé sur l’intégration de micro-aimants permanents dans un canal microfluidique. A partir de ce système de tri magnétique miniaturisé, offrant de nombreux avantages (dispositif portable, peu coûteux, nécessitant de faibles volumes réactionnels et potentiellement intégrable en « laboratoire sur puce »), une technique d’isolement sélectif de cellules bactériennes marquées magnétiquement a alors été développée. Ciblées sur des critères taxonomiques après hybridation in situ avec des sondes d’acides nucléiques biotinylés complémentaires d’une région spécifique du gène 23S rRNA, des cellules bactériennes ont été marquées magnétiquement après réaction de la sonde avec des nanoparticules magnétiques fonctionnalisées par des molécules de streptavidine. Les premiers résultats montrent l’établissement d’une méthode de tri suffisamment spécifique et sensible pour piéger les cellules marquées diluées (0,04%) au sein d’une suspension, à des niveaux compatibles avec l’isolement futur de populations d’intérêt à partir de communautés d’environnements complexes. Sur un principe comparable, l’approche a été adaptée à l’étude des transferts horizontaux de gènes in situ. Les applications d’un tri cellulaire grâce au marquage par des nanoparticules magnétiques et l’emploi de micro-aimants intégrés dans des microsystèmes fluidiques semblent donc très prometteuses pour le développement de la microbiologie environnementale. / Despite their importance, bacterial communities in the environment remain poorly characterized. On the one hand, it is difficult to gain knowledge of the community as a whole because over 99% of bacteria are recalcitrant to in vitro culture, rendering classic microbiological approaches imposible to carry out. On the other hand, metagenomics, which can be used to circumvent culture-based approaches by extracting all the genomes from a given environment, is also problematic given the associated technical limitations (biases related to DNA extraction, cloning, PCR, genome sequencing and assembling etc.), and conceptual difficulties related to the complexity and the homogeneity of the environments. In order to overcome some of the limitations of these approaches, bacterial cell selection methods have been developed and can be used to improve our understanding of microbial diversity. Based on taxonomic and/or functional selection and the direct isolation of bacterial cells from an environmental sample, bacterial cell selection can be used to reduce microbial community complexity by targeting specific populations, or even an isolated cell. A variety of classic approaches such as cultivation or DNA/RNA extraction can then be carried out. This cycle can theoretically be repeated until all members of the community are characterized. The aim of this doctoral thesis was to design a novel cell selection tool based on the permanent integration of micro-magnets into a microfluidic canal. In conjunction with a new miniaturized magnetic selection system that provides several advantages over larger systems (portable, low cost, requiring smaller reaction volumes and can be potentially integrated on “laboratory on a chip” systems), a method for selective bacterial cell isolation using magnetic labeling was developed. The bacterial cells were targeted based on taxonomic criteria; biotin-labeled probes were developed for a specific region of the 23S rRNA gene. Following in situ hybridization with the probes, baceterial cells were labeled with streptavidin-functionalized magnetic nanoparticles. First results showed that the tool was specific and sensitive enough to trap labeled and diluted (0,04%) cells from a suspension at levels that are comparible to populations of interest found in complex environmental communities. This tool has also been adapted to study in situ horizontal gene transfer as well. The application of a cellular selection tool that labels targets with magnetic nanoparticles coupled to fluidic microsystems with integrated nano-magnets looks very promising for future studiesin environmental microbiology.

Our understanding of microbial evolution is largely dependent on available genomic data of diverse organisms. Yet, genome-sequencing efforts have mostly ignored the diverse uncultivable majority in favor of cultivable and sociologically relevant organisms. In this thesis, I have applied and developed cultivation independent methods to explore microbial diversity and obtain genomic data in an unbiased manner. The obtained genomes were then used to study the evolution of mitochondria, Rickettsiales and Haloarchaea. Metagenomic binning of oceanic samples recovered draft genomes for thirteen novel Alphaproteobacteria-related lineages. Phylogenomics analyses utilizing the improved taxon sample suggested that mitochondria are not related to Rickettsiales but rather evolved from a proteobacterial lineage closely related to all sampled alphaproteobacteria. Single-cell genomics and metagenomics of lake and oceanic samples, respectively, identified previously unobserved Rickettsiales-related lineages. They branched early relative to characterized Rickettsiales and encoded flagellar genes, a feature once thought absent in this order. Flagella are most likely an ancestral feature, and were independently lost during Rickettsiales diversification. In addition, preliminary analyses suggest that ATP/ADP translocase, the marker for energy parasitism, was acquired after the acquisition of type IV secretion systems during the emergence of the Rickettsiales. Further exploration of the oceanic samples yielded the first draft genomes of Marine Group IV archaea, the closest known relatives of the Haloarchaea. The halophilic and generally aerobic Haloarchaea are thought to have evolved from an anaerobic methanogenic ancestor. The MG-IV genomes allowed us to study this enigmatic evolutionary transition. Preliminary ancestral reconstruction analyses suggest a gradual loss of methanogenesis and adaptation to an aerobic lifestyle, respectively. The thesis further presents a new amplicon sequencing method that captures near full-length 16S and 23S rRNA genes of environmental prokaryotes. The method exploits PacBio's long read technology and the frequent proximity of these genes in prokaryotic genomes. Compared to traditional partial 16S amplicon sequencing, our method classifies environmental lineages that are distantly related to reference taxa more confidently. In conclusion, this thesis provides new insights into the origins of mitochondria, Rickettsiales and Haloarchaea and illustrates the power of cultivation independent methods with respect to the study of microbial evolution.

Single-cell genomics (SCG) are methods for investigating heterogeneity between biological cells, among these is Smart-seq which sequences from RNA molecules. A more recent version of this method is Smart-seq3xpress which is currently in the process of being automated by the Sandberg lab at Karolinska Institutet. As part of this automated lab system, microwell plates are moved by a robot arm between molecular biology instuments. The purpose of this project was to create and integrate an image-based classification solution to validate the placement of these plates. This was done by building upon the VGG-16 convolutional neural network (CNN) model and specialising it through transfer learning to train models which classify microwell plate placement as correct or incorrect. These models were then integrated into the automated lab pipeline so that the system could self-correct or warn lab personnel of misplacement, removing the need for constant human supervision. / Enskild cellgenomik (eng. single-cell genomics) är metoder för att undersöka heterogenitet mellan biologiska celler, bland dessa metoder är Smart-seq vilken sekvenserar från RNA molekyler. En nyare version av denna metod är Smart-seq3xpress vilken nu håller på att automatiseras av Sandberglabbet vid Karolinska Institutet. Som del av detta automatiserade labbsystem förflyttas mikrobrunnplattor av en robotarm mellan molekylärbiologiska mätinstrument. Syftet med detta projekt var att skapa samt integrera en bildbaserad klassificeringslösning för att säkerställa placeringen av dessa plattor. Detta gjordes genom att bygga på djupinlärningsmodellen VGG-16 och specialisera den med överförd inlärning för att kunna träna modeller vilka klassificerar om mikrobrunnplattornas placeringar är korrekta eller inkorrekta. Sedan integrerades dessa modeller som en del av det automatiserade labbsystemet sådan att systemet kunde självkorrigera eller varna labbpersonal vid felplaceringar, och därmed ta bort behovet av konstant mänsklig tillsyn.