Lipoprotein lipase deficiency is a malfunkction of a key enzyme for lipoprotein metabolism. Lipoprotein lipase deficiency can cause serious episodes of pancreatitis. The deficiency is caused by pathology in the LPL gene. There are many variants in the LPL gene and quite a lot of them are pathogenic.The first gene therapy in Europe was licensed for the treatment of lipoprotein lipase deficienciency using a gain-of-function variant of the LPL gene. A pattern of examination was established to including MLPA, Light SNiP, quantitative real time PCR and sequencing in order to find pacients for the treatment. Three pacients suitable for the gene therapy Alipogene tiparvovec (AAV1-LPL S447X) were found amongst a group of patients with history of pancreatitis.

Doctor of Philosophy / Department of Agronomy / Guihua Bai / Guorong Zhang / Fusarium head blight (FHB) is a devastating fungal disease in wheat, reducing not only grain yield but also quality. The pathogen produces the mycotoxin deoxynivalenol (DON) that induces severe toxicological problems in human and animals. Using host resistance has been the most efficient way to control the disease. To identify quantitative trait loci (QTLs) for FHB resistance in Chinese landrace Haiyanzhong (HYZ), a recombinant inbred lines (RILs) population derived from a cross between HYZ and Wheaton was developed. The RILs were evaluated for percentage of symptomatic spikelets (PSS) in three greenhouse experiments, and genotyped using simple sequence repeats (SSRs) and single nucleotide polymorphism (SNPs) developed from genotyping-by-sequencing (GBS). Eight QTLs were identified for type II (PSS) resistance on chromosomes 5A, 6B, 7D, 2B (2), 3B, 4B, and 4D, with 5A as the major QTL. Ten SNPs closely linked to 5A, 6B, and 2B QTLs were successfully converted to Kompetitave allelic specific PCR (KASP) assays. To identify common QTLs across different populations, we constructed high-density GBS-SNP maps in an additional four RIL populations derived from the Chinese landraces, Wangshuibai (WSB), Baishanyuehuang (BSYH), Huangfangzhu (HFZ), and Huangchandou (HCD) and conducted meta-analysis of the QTLs for FHB resistance using a consensus map developed from the five populations. We identified six MQTLs on chromosomes 3BS (2), 3A, 3D, 2D, and 4D and 23 tightly linked GBS-SNPs to the MQTLs. These GBS-SNPs were successfully converted to KASPs. The KASPs linked to MQTLs can be used for pyramiding these QTL in breeding programs. To quickly reduce FHB damage in U.S. hard winter wheat (HWW), we transferred Fhb1, a major QTL with stable effects on FHB resistance, from Ning7840 into three adapted HWW cultivars Overland, Jagger, and Overley, by marker-assisted backcross (MAB), and assessed the effect of Fhb1 on FHB resistance in these different backgrounds. The results showed that Fhb1 can significantly lower FHB severity, Fusarium-damaged kernel (FDK), and DON accumulation in the all the three HWW backgrounds. Some of the selected lines showed high levels of FHB resistance, but agronomically similar traits as recurrent parents, can be used as resistant parents to improve HWW FHB resistance.

Submitted by MARINA MORTATI DIAS null (marina.mortati@gmail.com) on 2016-08-07T23:35:50Z No. of bitstreams: 1 Tese_Marina Mortati Dias.pdf: 1708247 bytes, checksum: 534b59570229fd27eb89d883f8e64df7 (MD5) / Approved for entry into archive by Juliano Benedito Ferreira (julianoferreira@reitoria.unesp.br) on 2016-08-10T13:49:51Z (GMT) No. of bitstreams: 1 dias_mm_dr_jabo.pdf: 1708247 bytes, checksum: 534b59570229fd27eb89d883f8e64df7 (MD5) / Made available in DSpace on 2016-08-10T13:49:51Z (GMT). No. of bitstreams: 1 dias_mm_dr_jabo.pdf: 1708247 bytes, checksum: 534b59570229fd27eb89d883f8e64df7 (MD5) Previous issue date: 2016-07-27 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Características reprodutivas, como a idade à puberdade, são economicamente relevantes para sistemas de produção de carne. A ocorrência de prenhez precoce aos 16 meses (OPP) é considerada um indicador da idade à puberdade e pode ser utilizada como critério de seleção. A busca por mutações potencialmente causais em genes candidatos pode ajudar a melhorar a acurácia de predição genômica se forem inseridas em painéis de marcadores genéticos. O objetivo desse estudo foi identificar polimorfismos potencialmente causais em genes candidatos associados a características reprodutivas em novilhas, por meio de dois ensaios. No primeiro, foram sequenciadas regiões dos genes PPP3CA e FABP4 em 380 amostras de DNA provenientes de novilhas Nelore, por meio de equipamento MiSeq® (Illumina, Inc). Nestas sequências foram detectadas duas mutações na região do gene PPP3CA e 13 no gene FABP4. Dentre elas, 14 eram SNP e uma era deleção, sendo esta última homozigota em todas as amostras sequenciadas, o que nos leva a acreditar que possa ser uma mutação fixada na raça Nelore. Um haplótipo constituído por quatro SNP no gene FABP4 não teve efeito significativo ao nível de p<0,05 sobre a característica OPP. Todos os SNP que passaram pelo controle de qualidade também foram analisados, porém nenhum teve efeito significativo sobre a característica OPP (p>0,05). No segundo ensaio, foram analisadas sequências de amostras de RNA de novilhas Brangus com o objetivo de identificar polimorfismos nas regiões de 62 genes candidatos associados a características reprodutivas. Detectou-se 1157 SNP nas regiões de 46 daqueles genes. Esses SNP foram comparados com SNP detectados a partir de sequências de amostras de RNA de outras quatro raças (Angus, Brahman, Nelore e Holandês). Cento e setenta e dois SNP foram coincidentes em todas as raças. A partir de sequências de RNA de Brangus também foram detectados 160 novos transcritos na região de 42 dos genes candidatos e cinco genes não anotados que sobrepõem a região de genes candidatos. A detecção de novos transcritos e genes é fundamental para enriquecer a anotação do genoma bovino. As mutações detectadas podem ser utilizadas para a customização de painéis de marcadores, e aplicações de seleção genômica. / Fertility traits such as age at puberty are economically important for meat production system. Occurrence of early pregnancy (OPP) is considered a good indicator of age at puberty and can be used as selection criteria. The search for causal mutations in candidate genes can helpful to improve the accuracy of genomic prediction if used to design low-density chips. The objective of this study was to identify polymorphism in candidate genes associated with puberty in heifers. DNA sequences were obtained from 380 Nellore heifers from exons sequencing of PPP3CA and FABP4 genes through MiSeq® equipment (Illumina, Inc.). From these sequences were found two SNP in the region of PPP3CA gene and 13 variations in the gene FABP4. Among these 15 variations 12 were SNP and one was a deletion, and this deletion was detected in all samples sequenced, which leads us to conclude that this may be a variation between Nellore and Hereford. One haplotype consisting of four SNP located in the FABP4 and nine SNP, that were analyzed separately, had not a significant effect at the p <0.05 on OPP characteristic. Also were analyzed RNA sequences from Brangus heifers in 62 candidate genes associated with puberty with the aim to discovery SNP in these regions. Were detected 1157 SNP in the region of 46 gene. These SNP were compared with SNP detected from RNA sequences from four breeds, Angus, Brahma, Nellore and Holstein. One hundred and seventy-two SNP were matched in all breeds. From Brangus RNA sequences were also detected 160 new transcripts in the region of 42 candidate genes and five genes not annotated that overlap the region of candidate genes. Detection of new transcripts and genes is essential to enrich the bovine genome annotation. The mutations detected can be used for the customization of low density chips and future genomics selection strategies.

Submitted by VALENTINA ROLDAN MONTES null (valentiniya@hotmail.com) on 2016-11-14T17:33:47Z No. of bitstreams: 1 dissertação_Valentina_Roldan.pdf: 1303188 bytes, checksum: 91f87b135d433e48cad11782f8d4380d (MD5) / Approved for entry into archive by Juliano Benedito Ferreira (julianoferreira@reitoria.unesp.br) on 2016-11-21T13:30:26Z (GMT) No. of bitstreams: 1 roldanmonter_v_me_jabo.pdf: 1303188 bytes, checksum: 91f87b135d433e48cad11782f8d4380d (MD5) / Made available in DSpace on 2016-11-21T13:30:26Z (GMT). No. of bitstreams: 1 roldanmonter_v_me_jabo.pdf: 1303188 bytes, checksum: 91f87b135d433e48cad11782f8d4380d (MD5) Previous issue date: 2016-10-18 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Considerando a importância das doenças que afetam o desempenho produtivo dos animais na indústria leiteira em todo o mundo é necessário implementar ferramentas moleculares que auxiliem na identificação e controle destas doenças. Quando ocorre alguma infecção em um organismo superior, existe aumento do número de células de defesa e o sistema imune inato proporciona uma linha de defesa contra os patógenos. Os “Toll Like Receptors” (TLR) são proteínas da membrana que desempenham um papel chave na imunidade, reconhecendo patógenos e, posteriormente, ativando as respostas. O presente estudo foi realizado para identificar SNPs no gene TLR4 bubalino e analisar suas associações com características de importância produtiva, incluindo a contagem de células somáticas (CCS). Foram utilizadas amostras de DNA de 130 búfalas da raça Murrah. A região codificante do gene TLR4 foi amplificada através de reações de PCR e posteriormente sequenciada. Os polimorfismos encontrados tiveram suas frequências alélicas e genotípicas calculadas e verificadas quanto ao equilíbrio de Hardy-Weinberg, além de serem utilizados para os estudos de associação. Foram identificados 13 polimorfismos do tipo SNP para as regiões sequenciadas do gene TLR4, sendo que a maioria encontra-se em região codificante. Encontrou-se associação significativa, com porcentagem de gordura dos Snps g514>C/T (P=0,0040) e g536>A/T (P=0,0035). As associações para CCS demostrou-se altamente significantes (p=<0,001) para todos os Snps (g322>G/A, g514>C/T, g536>A/T, g8338>A/C, g8341>A/G, g8342>T/G, g8343>G/A, g8345>A/G, g8413>A/G, g8428>G/A, g8438>A/C, g8578>G/T e g8582>A/C). Sugere-se que os Snp do gene TLR4 possam ser utilizados como marcadores moleculares em búfalos, já que foram verificadas suas associações com características como porcentagem de gordura e proteína, e contagem de células somáticas. / Molecular markers might be developed to investigate genetic variants associated to the disease and assist selection process in order to identify resistant animals. When the mammary gland is infected, there is an increase in the defense cells, also called somatic cells. It is a defense line of the immune system against pathogens. The “tool like receptors” TLR are membrane proteins that have an important role in the immunity, recognizing pathogens and activating adequate responses. The present study aimed to investigate the association of TLR4 gene SNPs with productive characteristics and SCC in buffaloes. The DNA was extracted from hair follicules of 130 Murrah buffaloes. The fragments were amplified by Polymerase Chain Reaction (PCR) and sequenced. Thirteen SNPs were found. Allelic and genotypic frequencies were calculated as well as the adhesion to Hardy-Weinberg equilibrium and the linkage disequilibrium (r2) and the association to the characteristic. For SCC was tested methodology linear generalized mixed model, assuming Poisson distribution. Bonferroni correction was applied for the number of SNPs. Thirteen SNP polymorphisms were identified in coding region of the TLR4 (g322>G/A, g514>C/T, g536>A/T, g8338>A/C, g8341>A/G, g8342>T/G, g8343>G/A, g8345>A/G, g8413>A/G, g8428>G/A, g8438>A/C, g8578>G/T, g8582>A/C). The SNPs g514>C/T and g536>A/T had significant association whit %G. All the SNPs were associated (p=0.001) whit the CCS. Other authors also reported the association of TRL4 SNPs to the trait. The results show that the SNPs of TLR4 gene can be used as molecular markers in buffaloes.

Made available in DSpace on 2014-12-17T15:18:09Z (GMT). No. of bitstreams: 1 ThayseAS.pdf: 421769 bytes, checksum: da81d1c89c2fb11516509aaaad4a2595 (MD5) Previous issue date: 2008-04-08 / Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico / Despite advances in vaccine development and therapy, bacterial meningitis (BM) remains a major cause of death and long-term neurological disabilities. As part of the host inflammatory response to the invading pathogen, factors such as reactive oxygen species are generated, which may damage DNA and trigger the overactivation of DNA repair mechanisms. It is conceivable that the individual susceptibility and outcome of BM may be in part determined by non synonymous polymorphisms that may alter the function of crucial BER DNA repair enzymes as PARP-1, OGG-1 and APE-1. These enzymes, in addition to their important DNA repair function, also perform role of inflammatory regulators. In this work was investigated the non synonymous SNPs APE-1 Asn148Glu, OGG-1 Ser326Cys,PARP-1 Val762Ala, PARP-1 Pro882Leu and PARP-1 Cys908Tyr in patients with bacterial meningitis (BM), chronic meningitis (CM), aseptic meningitis (AM) and not infected (controls). As results we found increased frequency of variant alleles of PARP-1 Val762Ala (P = 0.005) and APE-1 Asn148Glu (P=0.018) in BM patients, APE-1 Asn148Glu in AM patients (P = 0.012) and decrease in the frequency of the variant allele OGG-1 Ser326Cys in patients with CM (P = 0.013), regarding the allelic frequencies in the controls. A major incidence of individuals heterozygous and/ or polymorphic homozygous in BM for PARP-1 Val762Ala (P= 0.0399, OD 4.2, 95% IC 1.213 -14.545) and PARP-1 Val762Ala/ APE-1 Asn148Glu (P = 0.0238, OD 11.111, 95% IC 1.274 - 96.914) was observed related to what was expected in a not infected population. It was also observed a major incidence of combined SNPs in the BM patients compared with the control group (P=0.0281), giving evidences that SNPs can cause some susceptibility to the disease. This combined effect of SNPs seems to regulate the principal cytokines and other factors related to BM inflammatory response and point the importance of DNA repair not only to repair activity when DNA is damaged, but to others essential functions to human organism balance. / Apesar dos avan?os no desenvolvimento de vacinas e terapias, a meningite bacteriana (MB) continua sendo uma das principais causas de morte e seq?elas neurol?gicas causadas por uma doen?a infecciosa. Como parte da resposta inflamat?ria ao pat?geno invasor, fatores como esp?cies reativas de oxig?nio (ROS) s?o geradas, podendo causar danos no DNA e ativar seus mecanismos de repara??o. ? poss?vel que a susceptibilidade individual do hospedeiro ? MB possa ser em parte determinada por polimorfismos n?o-sin?nimos (SNPs) que possivelmente alterem a fun??o de enzimas de reparo de DNA da via BER como PARP-1, OGG-1 e APE-1. Estas enzimas, al?m da importante fun??o na corre??o de danos no DNA, tamb?m desempenham papel de reguladores inflamat?rios. Neste trabalho foram investigados os SNPs n?o-sin?nimos APE-1 Asn148Glu, OGG-1 Ser326Cys, PARP-1 Val762Ala, PARP-1 Pro882Leu e PARP-1 Cys908Tyr em pacientes com meningite bacteriana (MB), meningite cr?nica (MC), meningite ass?ptica (MA) e n?o infectados (controles). Como resultado, foi encontrado um aumento na freq??ncia dos alelos variantes de PARP-1 Val762Ala (P = 0.005) e APE-1 Asn148Glu (P=0.018) em pacientes com MB, de APE-1 Asn148Glu em pacientes com MA (P = 0.012) e diminui??o da freq??ncia do alelo variante OGG1 Ser326Cys (P = 0.013) em pacientes com MC em rela??o ?s freq??ncias al?licas observada nos controles para estes polimorfismos. Foi observado um maior n?mero de indiv?duos heterozigotos e/ou homozigotos polim?rficos para os gen?tipos polim?rficos PARP-1 Val762Ala (P= 0.0399, OD 4.2, 95% IC 1.213 -14.545) e PARP-1 Val762Ala/ APE-1 Asn148Glu (P = 0.0238, OD 11.111, 95% IC 1.274 - 96.914) no grupo MB em rela??o ao que se era esperado dentro de uma popula??o n?o-infectada. Observou-se tamb?m uma maior incid?ncia de SNPs combinados em pacientes com MB quando comparado ao grupo controle. Estas rela??es trazem evid?ncias de que os SNPs analisados causam alguma susceptibilidade ? doen?a. O efeito combinado destes SNPs parece influenciar a regula??o das principais citocinas e de outros fatores relacionados com a resposta inflamat?ria a MB, mostrando a import?ncia da ativa??o de enzimas de reparo de DNA n?o somente quando o DNA ? danificado, mas para outras fun??es essenciais ao equil?brio do organismo humano.

Next-generation sequencing is revolutionising in genetics, where base-by base information for the whole genome is available for a large sample of individuals. This type of data is becoming commonly used and will continue to be in the near future. One of the first questions arising is the identification of novel variants and subsequently genotype calling of the individuals in the sample. However, given the cost of sequencing, so far most projects are sequencing individuals in low to medium coverage. In this thesis, we present two distinct methods for SNP and genotype calling from low-coverage sequencing data, TreeCall and MVNcall, that combine sequencing and Linkage Disequilibrium (LD) information. We begin by describing the pipeline for next-generation sequencing analysis and existing methods for SNP and genotype calling using low-coverage sequencing information. Subsequently, we present the two novel LD-based methods for SNP and genotype calling. The two methods developed assume a study design where the individuals are both genotyped and sequenced at low-coverage. The genotypes are used to construct a haplotype scaffold, where the LD information is extracted, either by the construction of genealogical trees (TreeCall), or the approximation of a windows of contiguous SNPs of the scaffold by a multivariate normal distribution (MVNcall). Both methods have been applied on real datasets from the 1000 Genomes project and compared to other LD-based methods applied on the same datasets, mainly in terms of genotype calling and phasing. Whereas TreeCall gives lower genotype concordance rates than the other methods, MVNcall provides the highest genotype concordance rates for a dataset with a small sample size (Lowcoverage pilot of the 1000 Genomes project). Applying the MVNcall on a larger dataset (Phase 1 of the 1000 Genomes project), it achieves an overall genotype discordance rate of 0.58%, whereas SNPTools achieves an overall genotype discordance rate of 0.57%, Thunder 0.56%, and BEAGLE 0.61% (comparison based on Axiom chip). The main advantage of MVNcall is in terms of phasing accuracy, where by using a haplotype scaffold, and especially in the case where the haplotype scaffold is phased using pedigree information, it provides accurate haplotypes. MVNcall is also extended to incorporate trio information for genotype calling. Experiment on a deeply sequenced trio leads to an accurate set of haplotypes of the trio with switch error rates as low as ~0.28 for the parents and ~0.12 for the offspring.

Background: The ovarian response to follicle stimulating hormone (FSH) stimulation in assisted conception cycles is variable. Although it would be beneficial to predict accurately the response of patients to FSH, to date no robust predictors of ovarian performance have been identified. Recently, there have been a number of studies on the effect of single nucleotide polymorphisms (SNP) in the FSH receptor gene and its predictive value in the patients undergoing ovarian stimulation. Several reports have shown that two common SNPs at positions 307 and 680 in exon 10 of the FSH receptor gene are associated with ovarian response in in-vitro fertilization (IVF). Some authors have shown predictability of ovarian response to FSH stimulation in patients with different alleles, while others have refuted this finding. Until now, there is no clear clinical benefit in screening FSHR genotypes before IVF treatment. Objective: 1) To study the association between ovarian response and FSHR gene polymorphisms2) To study the association between FSHR gene polymorphisms and markers of ovarian reserve, including Anti Mullerian Hormone, Antral Follicle Count, Follicle Stimulating Hormone.Design: Prospective observational studyMethodology: 421 patients attending a tertiary reproductive medicine unit undergoing first cycle of IVF treatment were recruited into the study. Blood tests were taken on day 2 or 3 of the cycle for assessment of hormones and for DNA extraction. The SNP genotyping was done using Taqman analysis. Non-parametric tests were done to compare the various outcome parameters in patients with different genotypes.Results: FSHR p.Asn680Ser was not predictive of ovarian response. There was no evidence of any difference in basal FSH, AMH or AFC between the patients with different FSHR genotypes, with or without an adjustment for age or BMI. On subgroup analysis, there was no evidence that FSHR p.Asn680Ser genotypes are associated with PCOS, high AMH levels or response to clomiphene citrate. FSHR gene polymorphism was also not related to oocyte maturity or fertilization rate.Conclusions: FSHR p.Asn680Ser was not shown to be predictive of ovarian response, although clinically relevant differences cannot be ruled out. There may be an effect size but smaller than that detected for the power of this study. Other genetic markers may be relevant in the prediction of response to ovarian stimulation.

The objective was to determine how the interaction of implant strategies with the galanin receptor 2 genotype would influence feeding behavior, production efficiencies, carcass characteristics, and meat quality in finishing steers. Angus steers were selected based on GALR2-c.-199T>G genotype (n = 36 TT, 38 TG, and 19 GG). Calves were blocked by body weight and fed a standard feedlot ration, blood and BW were collected every 28 d. Steers were randomly assigned to an implant strategy of Revalor-S (1×) or Revalor-S (2×). Intake and feeding behavior data were individually recorded. There was an effect of genotype on DMI but not feed efficiency. Treatment interactions were observed for several meat quality attributes but not carcass characteristics. Altering implant strategy does not appear to interact with the GALR2-c.-199T>G genotype to alter production or carcass characteristics.

Hedgehogs from the genus Erinaceus are an important model organism for studying the postglacial recolonisation of Europe and the processes that take place in the secondary contact zones of their areas of distribution. In this study, five individuals of white-breasted hedgehog (Erinaceus roumanicus), four individuals of western hedgehog (Erinaceus europaeus) and one estimated hybrid were analysed. Geographical distribution of individuals used in the study covers the region of the Central Europe, however in the further research expansion of analsysed individuals will be needed and the whole Palearct should be sampled. The main goal was to implement novel methods in research of hedgehogs, which will enable to map the population-genomic structure of the genus Erinaceus in western Palearct. The method RADSeq (Restriction site associated DNA sequencing) enables to obtain polymorphic markers, e.g., SNPs which we used (Single Nucleotide Polymorphisms) across the genome. In this work it was analyzed 16382 SNPs. Using the binary data which indicates the presence and absence of SNPs for each species, hypotheses raised under classical analyzes of genetic markers from previous studies have not been fully confirmed. In further research it will be necessary to verify possible occurrence of biases connected with...

Diese kumulative Dissertation umfaßt drei Publikationen, die im Folgenden kurz vorgestellt werden. Die erste Publikation befasst sich mit der Fragestellung aus der Populationsgenetik, ob Isolatpopulationen für die Erforschung genetischer Ursachen von Krankheiten oder quantitativen Phänotypen besser geeignet sind als nicht isolierte Populationen. Man erwartet aufgrund homogenerer Umwelteinflüsse, geringerer Anzahl kausaler genetischer Varianten und insbesondere durch homogenere Bereiche im Genom Vorteile bei der Identifikation genetischer Ursachen in Isolatpopulationen. Am Beispiel der Sorben, die einen gewissen Isolatcharakter aufweisen, sollte deshalb untersucht werden, inwieweit sich diese von einer deutschen populationsbasierten Studie wie KORA genetisch unterscheiden und welche Bedeutung die Unterschiede für genetische Assoziationsanalysen haben. In der ersten Publikation wird gezeigt, daß die Sorben Merkmale genetischer Isolation aufweisen, die nicht auf eine stärkere Verwandtschaftsstruktur der Studienpopulation gegenüber KORA zurückzuführen sind. Die Merkmale genetischer Isolation sind moderat, trotzdem ist der slawische Ursprung erkennbar. Daraus läßt sich schließen, daß die Sorben ursprünglich genetisch isoliert waren, jedoch die genetische Isolation verloren geht. Trotz Unterschiede in der SNP-Korrelationsstruktur durch ein im Mittel höheres Kopplungsungleichgewicht zwischen benachbarten SNPs ist kein klarer Vorteil bei der Power von SNP-Assoziationsanalysen zu erwarten. Die Verwandtschaftsstruktur der Sorben kann aber bei unkorrigierten SNP-Assoziationsanalysen zu einer Varianzinflation des Effektschätzers führen und die Power des Tests in komplexer Weise beeinflussen. Es sollte daher in einer weiteren Publikation geklärt werden, wie die Verwandtschaftsstruktur der Studienpopulation und die Heritabilität eines Phänotyps die Varianz des Effektschätzers und die Power des Tests tatsächlich beeinflussen. In der zweiten Publikation wird der Einfluß der Verwandtschaftsstruktur auf SNP-Assoziationsanalysen im Detail untersucht. Verwandtschaften in einer Studienpopulation führen zu korrelierten Phänotypen, was die Annahme unabhängiger Beobachtungen des einfachen linearen Modells verletzt. Aus empirischen Studien war zudem bekannt, daß eine stärkere Verwandtschaftsstruktur der Studienpopulation und eine größere Heritabilität des Phänotyps den Fehler erster Art eines unkorrigierten Tests vergrößern. Der Einfluß der Verwandtschaftsstruktur auf die Power wurde in empirischen Studien unterschiedlich beurteilt. Zudem wird genomic control häufig dazu verwendet, eine Inflation der Teststatistik durch Verwandtschaft zu korrigieren, jedoch führt genomic control zu einer Power-Reduktion. Auch diese empirischen Beobachtungen sollten erklärt werden. In der zweiten Publikation wird analytisch gezeigt, wie die Verwandtschaftsstruktur und die Heritabilität des Phänotyps mit der Varianzinflation des Effektschätzers und der Teststatistik zusammenhängen. Während der Fehler erster Art mit größerer Varianzinflation steigt, wird die Power in komplexer Weise beeinflußt. Ob die Power bei Varianzinflation größer oder kleiner wird, hängt von der Stärke des genetischen Effekts und vom Signifikanzniveau des Tests ab. Zudem konnten weitere empirische Beobachtungen aus der Literatur analytisch erklärt werden, zum Beispiel daß der Erwartungswert des Effektschätzers nicht durch Verwandtschaft beeinflußt wird, die empirische Varianz des Effektschätzers bei Verwandtschaft deflationiert ist und daß die Allelfrequenz des SNP nur einen geringen Einfluß auf die Varianzinflation hat. Weiterhin kann genomic control im Allgemeinen nicht für die Korrektur von Varianzinflation durch Verwandtschaft empfohlen werden. Obwohl der Fehler erster Art durch genomic control eingehalten wird, führt die Methode zu einem starken Power-Verlust in Abhängigkeit der Varianzinflation. Zur Bestimmung der Varianzinflation wurde eine Näherungsformel analytisch hergeleitet, die nur die Verwandtschaftsstruktur und die Heritabilität des Phänotyps benötigt. Aus der Publikation folgt, daß eine Varianzinflation kleiner als 1,05 keinen relevanten Einfluß auf den statistischen Test hat und die Verwendung des einfachen linearen Modells in diesem Fall angemessen ist. Ist die Varianzinflation größer, müssen Methoden wie beispielsweise gemischte Modelle im Rahmen einer SNP-Assoziationsanalyse verwendet werden, welche explizit die Verwandtschaftsstruktur berücksichtigen. In der dritten Publikation wird ein weiteres Paradigma der Statistik betrachtet. Eine SNP-Assoziationsanalyse kann neben klassischen Methoden auch mit bayesianischen Methoden erfolgen. Bayesianische Methoden bieten dabei die Möglichkeit, SNP- und Phänotyp-Korrelationen zu berücksichtigen und so die Modellanpassung gegenüber der klassischen Analyse zu verbessern. Am Beispiel einer Kinderstudie sollte nach dem Einfluß bestimmter SNPs ausgewählter Kandidaten-Gene (SORT1, HMGCR, MLXIPL, FADS2, APOE, MAFB) auf Lipidkonzentrationen von HDL-C (high density lipoprotein cholesterol), LDL-C (low density lipoprotein cholesterol), TC (total cholesterol) und TG (triglyceride) gesucht werden, um auf genetische Ursachen für Parameter des Stoffwechsels in der frühen Entwicklung schließen zu können. In der dritten Publikation wurde zunächst eine klassische SNP-Assoziationsanalyse durchgeführt und ein Zusammenhang von SORT1 und APOE mit LDL-C und TC identifiziert. Darauf wurde in einer bayesianischen Analyse der mehrdimensionale Phänotyp aus HDL-C, LDL-C und TG modelliert, wodurch explizit die Phänotyp-Korrelationsstruktur berücksichtigt wurde. Für die einzelnen Lipidkonzentrationen wurde eine plausible Auswahl von Einflussfaktoren bestehend aus genetischen Varianten, Alter, Geschlecht und BMI unter Berücksichtigung verschiedener genetischer Modelle bestimmt. Dadurch wurden sowohl die Ergebnisse aus der klassischen Analyse bestätigt, als auch weitere Kandidaten, beispielsweise ein Zusammenhang zwischen MLXIPL und TG, gefunden. Ein wichtiges Ergebnis dieser Arbeit war zudem die Präsentation der bayesianischen Modellergebnisse in einfacher Form. Für die bayesianische Analyse wurden gegenüber der klassischen Analyse einige Vorteile festgestellt, die zukünftig weiter untersucht werden sollen. Dazu zählt die Berücksichtigung von Korrelationsstrukturen im bayesianischen Modell, die zu einer verbesserten Identifikation von Phänotyp-Genotyp-Beziehungen führen kann. Weiterhin lassen sich die bei der bayesianischen Modellauswahl identifizierten genetischen Effekte über alle Modelle mitteln, in denen die entsprechenden Variablen eingeschlossen wurden. Dadurch fallen die empirischen Varianzen der Effekte meist kleiner aus als die zugehörigen Varianzen der Beta-Schätzer aus der klassischen Analyse. Zuletzt werden bei der klassischen Analyse meist aufgrund einzelner Fehlwerte in Phänotypen, SNPs oder Kovariablen ganze Fälle verworfen, was zu einer erheblichen Fallzahlreduktion führen kann. Diese Fehlwerte können bei der bayesianischen Analyse als zusätzliche Parameter modelliert werden, wodurch die ursprüngliche Fallzahl erhalten bleibt. Diese Aspekte sollen zukünftig in Simulationsstudien untersucht werden, in denen der Einfluß von Korrelationsstrukturen, Effektstärken und Fehlwerten auf die Identifikation genetischer Effekte im Vergleich zur klassischen Analyse betrachtet wird.