Stockholm Neonatal Project (SNP)

<p>A major challenge in human genetics is to understand the genetic variation underlying common diseases. In this thesis, I focus on forces creating differences between individuals and genomic regions, methods for characterizing genomic variation, and the association between genomic and phenotypic variation. Genetic markers are widely used to locate genes associated with different phenotypes. In my first paper, I describe novel algorithms for automatic genotype determination of microsatellite markers, a procedure which is currently both time-consuming and error prone. </p><p>The co-segregation of genetic markers in a population leads to non-random association of alleles at different loci - linkage disequilibrium (LD). LD varies throughout the genome and differs between populations due to factors such as their demographic history. In my second paper, I discuss the increased power, for mapping of human traits, that results from studying a population with appreciable levels of LD such as is found in the Swedish Sami population. </p><p>Lately, large-scale analyses of single nucleotide polymorphisms (SNPs) have become available and efforts have been made to identify a set of SNPs, which captures most of the genome variation in a population (tagSNPs). In my third paper, I describe the limitations of this approach when applied to data from an independent population sample of randomly ascertained SNPs. The transferability of tagSNPs between populations is poor, presumably due to variation in allele frequencies and the bias towards common SNPs used in most studies. </p><p>The level of genomic variation is influenced by population structure, recombination and mutation rate, as well as natural selection. During the exodus from Africa, humans have adapted to new environmental conditions. In my fourth paper, I describe a new method for identifying genomic regions carrying signatures of recent positive selection and apply this to an available dataset of millions of SNPs.</p>

The domestic dog was probably the first domesticated animal, and the only one to spread to all continents in ancient times. The dog is one of the most phenotypically diverse animals, a result of human selection throughout dog history. Studies of the genetic origins and early spread of domestic dogs is important to gather information about biological and cultural mechanisms behind domestication but also to investigate early human history. The step from a hunter and gatherer lifestyle to farming is one of the most important steps in human history. In this thesis I will present work aimed at understanding both domestic dog origins and dispersal. In order to be able to investigate dog origins based on a second haploid chromosome we identified 14,437 bp of Y-chromosomal DNA sequence. Based on this we show that dogs in Asia south of Yangtze River (ASY) has the highest genetic diversity and was founded from a large number of wolf founders confirming earlier mtDNA results. Early dog dispersal is tightly coupled to human history with the dog brought along as a cultural item. We have for the first time investigated the dog dispersal into Polynesia and Australia and our data can be used as evidence for a more complex settlement of Polynesia than earlier indicated from archaeological and linguistic studies. Analysis of Y-chromosome SNPs in Australian dingoes confirms earlier mtDNA genetic studies that the dingo is part of the domestic dog phylogeny and was founded from a small population of domestic dogs. We have also for the first time investigated the dog population on Madagascar and our data strongly indicates a mainland African origin for the Madagascan dogs. Finally, we have investigated the American dog population sampled from throughout the continent and also for the first time included putative indigenous breed dogs such as Chihuahua and Pero Sín Pelo del Peru, and the free-ranging Carolina dogs from southern USA. Our data clearly indicates a primarily Old World origin for the indigenous breed dogs and also for the free-ranging Carolina dogs in USA. We can also for the first time present evidence for continuity between the ancient and extant dog population with e.g. exclusive sharing of a haplotype between a modern sample of Chihuahua and an ancient Mexican sample. / QC 20120510

Imagine if you could combine the power and capabilities ofan entire laboratory in the palm of your hand. Advances inmicrofluidic chip technology promise to integrate andminiaturize multiple lab processes into a single palm-sizeddevice. The advantages of these lab-on-a-chip devices,sometimes also referred to as micro total analysis systems(µTAS), compared with conventional bench-scale systems arenumerous and wide ranging and include: less reagentconsumption, low manufacturing costs, increased performance,faster analysis, high sample throughput, integration andautomation possibilities, and disposability. However,microfluidic devices also present challenges such as theinterfacing to the macro world and detection limits. In this thesis the focus has been to develop novel discretemicrofluidic components for biotechnology and organic chemicalapplications with the goal to integrate them to formlab-on-chips. A flow-through filter-chamber device has beendesigned, manufactured and evaluated for chemical analysis onbeads. Passive liquid handling has been integrated on the chipin the form of hydrophobic valves at the inlet channels. Anarray format has also been developed to allow parallel analysisof multiple samples. The filter-chamber functions well forsingle nucleotide analysis using pyrosequencing. Initialevaluations on catalyst screening in the filter-chamber devicehas been performed. The suitability of valve-less micropumps for biochemicalapplications is presented. Fluids encountered in variousbiochemical methods, including living cells, that areproblematic for other micropumps have been pumped with goodperformance. This thesis also introduces expandablemicrospheres as a novel component in microfluidics includingapplications such as one-shot valves, micropositioning andsurface enlargement. A novel technique for bead immobilization in microfluidicdevices based on surface chemistry is presented in this thesis.Beads for both biochemical assays and organic chemistry havebeen self-sorted and self-assembled in line patterns as narrowas 5 µm on both structured and unstructured substrates.This method will greatly facilitate the generation of screeningplatforms, for example. To develop a microfluidic device for catalysis-on-chip,ligands for asymmetric catalysis have successfully beenimmobilized in silicon channels by consecutive microcontactprinting, which is a novel technique presented in thisthesis. <b>Keywords:</b>microfluidics, beads, microspheres, silicon,filter-chamber, flow-through, bead trapping, DRIE, passivevalves, fluorocarbon, microfluidic array, adhesive bonding,valve-less micropump, microcontact printing, PDMS,self-assembly, self-sorting, DNA, SNP, pyrosequencing,allele-specific extension, expandable microspheres, catalysis,chiral ligand, monolayer, miniaturization, lab-on-a-chip,µTAS.

A major challenge in human genetics is to understand the genetic variation underlying common diseases. In this thesis, I focus on forces creating differences between individuals and genomic regions, methods for characterizing genomic variation, and the association between genomic and phenotypic variation. Genetic markers are widely used to locate genes associated with different phenotypes. In my first paper, I describe novel algorithms for automatic genotype determination of microsatellite markers, a procedure which is currently both time-consuming and error prone. The co-segregation of genetic markers in a population leads to non-random association of alleles at different loci - linkage disequilibrium (LD). LD varies throughout the genome and differs between populations due to factors such as their demographic history. In my second paper, I discuss the increased power, for mapping of human traits, that results from studying a population with appreciable levels of LD such as is found in the Swedish Sami population. Lately, large-scale analyses of single nucleotide polymorphisms (SNPs) have become available and efforts have been made to identify a set of SNPs, which captures most of the genome variation in a population (tagSNPs). In my third paper, I describe the limitations of this approach when applied to data from an independent population sample of randomly ascertained SNPs. The transferability of tagSNPs between populations is poor, presumably due to variation in allele frequencies and the bias towards common SNPs used in most studies. The level of genomic variation is influenced by population structure, recombination and mutation rate, as well as natural selection. During the exodus from Africa, humans have adapted to new environmental conditions. In my fourth paper, I describe a new method for identifying genomic regions carrying signatures of recent positive selection and apply this to an available dataset of millions of SNPs.

Stockholm Neonatal Project (SNP)

DNA copy number variations (CNVs), which involve the deletion or duplication of subchromosomal segments of the genome, have become a focus of genetics research. This dissertation develops Bayesian HMMs for finding CNVs from single nucleotide polymorphism (SNP) arrays. A Bayesian framework to reconstruct the DNA copy number sequence from the observed sequence of SNP array measurements is proposed. A Markov chain Monte Carlo (MCMC) algorithm, with a forward-backward stochastic algorithm for sampling DNA copy number sequences, is developed for estimating model parameters. Numerous versions of Bayesian HMMs are explored, including a discrete-time model and different models for the instantaneous transition rates of change among copy number states of a continuous-time HMM. The most general model proposed makes no restrictions and assumes the rate of transition depends on the current state, whereas the nested model fixes some of these rates by assuming that the rate of transition is independent of the current state. Each model is assessed using a subset of the HapMap data. More general parameterizations of the transition intensity matrix of the continuous-time Markov process produced more accurate inference with respect to the length of CNV regions. The observed SNP array measurements are assumed to be stochastic with distribution determined by the underlying DNA copy number. Copy-number-specific distributions, including a non-symmetric distribution for the 0-copy state (homozygous deletions) and mixture distributions for 2-copy state (normal), are developed and shown to be more appropriate than existing implementations which lead to biologically implausible results. Compared to existing HMMs for SNP array data, this approach is more flexible in that model parameters are estimated from the data rather than set to a priori values. Measures of uncertainty, computed as simulation-based probabilities, can be determined for putative CNVs detected by the HMM. Finally, the dissertation concludes with a discussion of future work, with special attention given to model extensions for multiple sample analysis and family trio data.

The search for genetic regions associated with complex diseases, such as cancer or Alzheimer's disease, is an important challenge that may lead to better diagnosis and treatment. The existence of millions of DNA variations, primarily single nucleotide polymorphisms (SNPs), may allow the fine dissection of such associations. However, studies seeking disease association are limited by the cost of genotyping SNPs. Therefore, it is essential to find a small subset of informative SNPs (tag SNPs) that may be used as good representatives of the rest of the SNPs. Several informative SNP selection methods have been developed. Our experiments compare favorably to all the prediction and statistical methods by selecting the least number of informative SNPs. We proposed algorithms for faster prediction which yielded acceptable trade off. We validated our results using the k-fold test and its many variations.

DNA copy number variations (CNVs), which involve the deletion or duplication of subchromosomal segments of the genome, have become a focus of genetics research. This dissertation develops Bayesian HMMs for finding CNVs from single nucleotide polymorphism (SNP) arrays. A Bayesian framework to reconstruct the DNA copy number sequence from the observed sequence of SNP array measurements is proposed. A Markov chain Monte Carlo (MCMC) algorithm, with a forward-backward stochastic algorithm for sampling DNA copy number sequences, is developed for estimating model parameters. Numerous versions of Bayesian HMMs are explored, including a discrete-time model and different models for the instantaneous transition rates of change among copy number states of a continuous-time HMM. The most general model proposed makes no restrictions and assumes the rate of transition depends on the current state, whereas the nested model fixes some of these rates by assuming that the rate of transition is independent of the current state. Each model is assessed using a subset of the HapMap data. More general parameterizations of the transition intensity matrix of the continuous-time Markov process produced more accurate inference with respect to the length of CNV regions. The observed SNP array measurements are assumed to be stochastic with distribution determined by the underlying DNA copy number. Copy-number-specific distributions, including a non-symmetric distribution for the 0-copy state (homozygous deletions) and mixture distributions for 2-copy state (normal), are developed and shown to be more appropriate than existing implementations which lead to biologically implausible results. Compared to existing HMMs for SNP array data, this approach is more flexible in that model parameters are estimated from the data rather than set to a priori values. Measures of uncertainty, computed as simulation-based probabilities, can be determined for putative CNVs detected by the HMM. Finally, the dissertation concludes with a discussion of future work, with special attention given to model extensions for multiple sample analysis and family trio data.

Les cytokines sont d'importants médiateurs dans la physiopathologie des lymphomes hodgkiniens (LH). A partir d'une cohorte de 464 patients, nous avons évalué l'impact pronostique de onze SNPs parmi les gènes de cytokines : IL10 (rs1800890, rs1800896, rs1800871, rs1800872), TNFA (rs1800629) ; IL6 (rs1800795) ; IL1B (rs16944) ; ILRN (rs419598) ; INFG (rs2430561) ; IL12 (rs3212227) ; CCL17 (rs223828). Le génotypage du SNP de l'IL12 montre une distribution différente de celle attendue dans la population générale selon le test de Hardy-Weinberg. Ce résultat suggère que les variations génétiques de l'IL12 pourraient être impliquées dans la susceptibilité au LH. Les patients porteurs du génotype IL10-1082AA présentent un taux de rémission complète au traitement initial supérieur aux patients présentant un autre génotype (95% vs. 88% P = .02). Pour les patients de stade avancé III-IV, le taux de survie globale à 6 ans est statistiquement différent entre les génotypes IL10-592AA/CC/AC et IL10-819TT/CC/CT (100%, 94%, 78%, P = .03). Ce résultat est retrouvé pour les patients porteurs de LH n'exprimant pas l'EBV. Pour les LH EBV négatif, le taux de survie sans progression à 6 ans est différent en fonction du génotype du TNFA-308AA/GG/AG (100%, 84%, 68%, P = .03). Il n'a été pas retrouvé de corrélation entre les génotypes et les dosages plasmatiques de l'IL-10, TNFA, IL-1RA, IL-6. Cette étude montre que le ''fond génétique immun'' est important à prendre en considération pour définir le pronostic des patients. Le rôle des SNPs de l'IL10 et du TNFA dans les LH EBV négatif devra être confirmé ainsi que l'influence des variations génétiques de l'IL12 dans la susceptibilité au LH.