University of Maryland, Baltimore
SiGN Project Overviews12001- Development of a Method for Joint Modeling of Genetic and Clinically Obtained Neuroimaging Data to Discover Genetic Associations with Neuroimaging PhenotypesIn initial studies of imaging genetics, imaging is used as an endo-phenotype to perform na?ve, univariate modeling of genetic data. Relevant genetic markers that may affect variability in particular traits are then proposed. Current research avoids the intricacy of both the genetic and imaging data by largely ignoring complex genetic structure and interactions, while summarizing imaging data through a few parameters. As a result, groups of genetic factors working together to produce a greater effect than any individual marker may be missed. This proposal will test the following hypotheses:A high-throughput automated segmentation approach can accurately measure the volumes of neuroanatomical structures from clinical MRIs.Application of a statistical model for the interaction of genetics with brain anatomy, explicitly accounting for structure patterns present in both data domains can yield novel genetic associations.Our long-term aim is to develop methods that identify clusters of genetic markers working together to affect phenotypes related to brain structure. Our first step will be to develop methods for automated segmentation of brain structures from the clinical MRIs within SiGN. Because clinical scans are not obtained in a standardized format or orientation, they are not compatible with the standard approaches to automated segmentation such those employed in the popular free-surfer program. Our next goal is to develop a statistical model for the interaction of genetics with brain anatomy, explicitly accounting for structure patterns present in both data domains. We will expand current epistasis modeling methods, such as co-clustering of genetic markers with phenotypic traits, and will develop models for components of brain regions. Further, we will explore methods such as canonical correlation analysis to identify combined features. The goal of this framework is to determine connections between clusters of SNPs simultaneously with properties of brain anatomy, and to further determine how these connections behave with respect to different individual traits. We propose developing these methods using those subjects in NINDS SiGN for whom MRI and genome-wide genotyping data are already available, and validating them in the larger dataset that becomes available once genotyping is completed at CIDR. 12002- Classification of Stroke Subtypes by Race-Ethnicities in the Stroke Genetics Network with Causative Classification SystemIdentification of gene variants of stroke subtypes is important for the development of tailored ischemic stroke therapies among various ethnic groups. Valid and reliable determination of ischemic stroke subtype is essential for achieving this goal and to standardize a classification scheme across multi-center studies and different populations. Causative Classification System for Ischemic Stroke (CCS) is a novel computerized sub classification tool developed to improve reliability and accuracy of classifying stroke types. The CCS algorithm relies on both phenotypic and causative stroke variables. Distinct Race-Ethnic subgroups and risk factor profiles may predispose to specific classification of strokes in SiGN. Hypothesis: Due to genetic and environmental factors, different race-ethnicities may have different proportions of stroke subtypes as reflected in phenotypic and etiological classification in the CCS system.Analysis: A comparison of the CCS classification by the race-ethnic groups adjusted for risk factor profiles needs to be performed.12003- Systematic, Centralized, and Study-specific Phenotyping in a Large-scale Stroke Genetics Study: An Analysis from the NINDS Stroke Genetics Network (SiGN)The Stroke Genetics Network (SiGN) funded by the NINDS aims to identify genetic risk factors in ischemic stroke using whole-genome association studies (GWAS). High quality phenotyping is crucial to successful application of GWAS. As a heterogeneous disorder, stroke poses specific challenges. The Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification is a broadly used, but its validity is challenged especially when performed by multiple investigators with differing interpretations of the system. The Causative Classification System for Ischemic Stroke (CCS) system is a new, web-based, and computerized algorithm that integrates clinical, diagnostic, and etiologic stroke characteristics in an evidence-based manner (ccs.mgh.harvard.edu) to generate subtypes.We will perform pair-wise and cross category comparisons of adjudicated phenotypes by the TOAST and the CCS system, the site-specific and the central adjudication, and across all 4 approaches. We will analyze sources of variance.12004- Etiologic Stroke Subtypes in the NINDS Stroke Genetics NetworkThe aim is to describe the phenotyping process in the SiGN study. The goal is to generate a master paper on methodology to be used as a reference document by all other SiGN papers. This paper will provide a description of CCS, modifications done to the original CCS for the purpose of adopting it for retrospective data collection, the training and certification process, and the re-adjudication process. The paper will also provide results on distribution of stroke subtypes and reliability of CCS.12005- (Withdrawn by Author) Pitfalls in Etiologic Stroke Classification: Lessons Learnt from the NINDS Stroke Genetics Network Study12006/07- The Impact of Diagnostic Stroke Tests on Final Etiologic Subtype Assignment (12006 combined with 12007)It is debated that, instead of one-fits-all approach, stroke evaluation should be customized at an individual patient level according to the underlying phenotype. Nevertheless, such an approach, although appealing from the cost-effectiveness perspective, requires knowledge of diagnostic yield of each individual stroke test. This issue is further complicated by the fact that diagnostic yield is not a constant factor; it changes as a function of underlying etiologic phenotype. The SiGN study collected cardiac and vascular evaluation data as well as phenotypic CCS subtypes where abnormal evaluation findings were organized in major etiologic categories. In this project, we aim to quantify the yield of cardiac and vascular evaluations for identifying an alternative etiology in patients with a given CCS-phenotype. More specifically, we will determine how often: 1-cardiac and vascular evaluations reveal an abnormality in patients with a typical lacunar infarct, 2- cardiac evaluation reveals an abnormality in patients with large artery atherosclerosis (causing ≥50% stenosis), and 3- vascular evaluation reveals an abnormality in patients with a known cardiac source of embolism. We will also determine the yield of stroke work-up in identifying a major etiology that can potentially lead to a change in treatment. The results of this project may help generate evidence-based and cost-effective evaluation and management strategies. 12008- Agreement between CCS and TOAST Classifications with SiGNThe aim is produce agreement statistics (kappa) comparing CCS with TOAST in SiGN. Analysis will be done for the consortium and stratified by site. Primary analysis will be the 5-item classification. Secondary analysis will be stratified by the presence or absence of diagnostic evaluations.12009- Design of Stroke Genetics Network (SiGN)The goal of this paper is to describe the overall design of SiGN. The specific aims will be:State the goal and objectives of anizational Structure of SiGN. This includes a description of the cores and relationship with the funder. Also will list all members of SiGN and describe the underlying study design, inclusion/exclusion criteria, location and size of each participating site.Phenotyping Effort. This will focus on the use of the CCS system, the central oversight provided by the phenotype committee, the training and central re-adjudication processes.What genotypes are available. This will focus on genotypes acquired within SiGN (via CIDR) and genotypes available from studies who had them prior to SiGNWhat is the primary analysis. This will include a description of the case control GWAS design, including how genotyping was performed, how cases were identified, and the broad analysis plan.12010- The Impact of Neuroimaging on Identification of Etiologic Stroke Subtypes (Project-EMA)Certain topographic features of brain infarcts such location, size, distribution, and age correlate with the underlying stroke etiology. CCS uses such topographic features to support one mechanism over another in the presence of multiple competing mechanisms. In other words, CCS regularizes subtype assignment to a probable known etiologic category rather than undetermined-unclassified category in the presence of multiple competing causes. Because CT and MRI have different sensitivities in identifying topographic infarct characteristics, their absolute yield in terms of shrinking the size of undetermined category is expected to be different. We hypothesize that the use of MRI results in smaller size of undermined category as compared to CT. We will identify cases where the method of brain imaging was CT and compare the proportion of undetermined cases in this population with the proportion in those scanned by MRI in the SiGN dataset. We will also assess the utility of individual imaging-related data entry fields in the CCS (such as lacunar infarcts, internal watershed infarcts, temporally separate infarcts, etc.) in classifying stroke etiology into an evident or probable subtype. The paper may provide insight into the role of brain imaging in evaluation and management of ischemic stroke patients.12011- Study of Fibromuscular DysplasiaSince FMD is a rare disease, we are searching for collaborators with access to samples from FMD patients. We are primarily enrolling FMD patients at several collaborating centers for a planned gene discovery analysis. We are currently searching for collaborators with access to samples to optimize the power of our planned analyses. If enough subjects with FMD exist within SiGN, we propose to conduct a primary analysis of subjects with a diagnosis of FMD and genotypes in SiGN. If enough samples do not exist for a meaningful stand-alone analysis within SiGN, we wish to collaborate with individual cohorts for our studies of FMD. We are interested in obtaining clinical information, angiographic data if possible to verify medial hyperplasia versus other subtypes of FMD, genotype data and/or DNA samples for additional sequencing or genotyping we will conduct as part of our larger study. These samples would ideally be included in the discovery phase of the project which is likely to include whole exome or whole genome sequencing, but the samples may be used in replication analysis depending on timeframes and the available phenotypic data to verify sub types of FMD.Analysis of de-identified data will be conducted in the laboratory of Dr. Santhi Ganesh at the University of Michigan where she has set up the necessary infrastructure for genomic data analysis. She employs in her laboratory one full-time statistical analyst and ~10% FTE of a C++ programmer for computationally intensive procedures, in addition to two postdoctoral research fellows. Her group is experienced in the analysis of GWAS and sequencing data.12012- Total risk factor burden varies in different age groups of ischemic stroke patients: A study of 2505 patients regarding risk factors, age and TOAST subtypesThe prevalence of risk factors for ischemic stroke may vary between different groups of stroke patients. We examine the distribution of individual traditional risk factors as well as the total risk factor burden in different age groups and among the TOAST subtypes in 2505 patients with ischemic stroke from Lund Stroke Register. No genetic data are included in this manuscript. Because some of the patients had CCS classification as part of the SiGN project, this proposal is submitted to the SiGN writing committee after consulting Dr Tatjana Rundek and Dr Steven Kittner at the ISGC meeting in Charlottesville.12013- Genetic Analyses of Lipids in Cerebral Hemorrhage and Small Vessel DiseaseLipid levels are associated with many clinical and radiologic manifestations of cerebrovascular disease, and lipid-lowering therapy has proven beneficial in ischemic stroke prevention. Unfortunately, both observational and clinical trial data suggest this strategy may be associated with increased risk of ICH, but it is not clear whether LDL, HDL, or triglyceride (TG) levels modulate this risk, or how these lipid fractions impact other manifestations of cerebral small vessel disease. Furthermore, it is not clear whether this relationship is causative or associative, nor is the magnitude or direction of effect. Given that statins both decrease LDL and increase HDL levels, and ongoing clinical trials of cholesterol ester transfer protein (CETP) inhibitor drugs offer the promise of substantially increased HDL levels, clarification of the role of lipid fractions in ICH and cerebral small vessel disease has broad implications for the use of lipid lowering strategies for the treatment and prevention of these conditions. We will apply a multidisciplinary approach that leverages expertise in advanced genetic techniques, neuroimaging, and lipid biology to test the hypothesis that genetic variants that influence circulating lipid fractions alter risk of ICH and its related neuroimaging intermediates.We propose to: 1) discover the impact of common genetic variants known to affect lipid levels on the severity of neuroimaging manifestations of cerebral small vessel disease, 2) determine the impact of rare genetic variants in known and novel gene networks with a role in lipid levels on the severity of these same neuroimaging intermediates, and 3) examine the influence of these common and rare genetic variants on MRI measures of white matter integrity, a novel imaging marker of cerebral small vessel disease.12014- Assessment of an Algorithm of Case Report Forms to Describe Stroke SubtypesBackgroundA number of genetic studies have noted genetic associations among stroke sub-types. However, the inability to replicate such associations has led to uncertainty of findings in genetic studies. Two of the main obstacles are the heterogeneity of stroke consisting of several discrete subtypes and small sample sizes. The integration of genetic samples and raw data from large multicenter clinical trials for the development of a large genetic database would significantly increase sample size. We discuss the validation (development) of a systematic method of using data abstracted from multicenter randomized clinical trials that was not designed for genetic and/or phenotypic studies for identification of subtype-specific genetic studies.MethodsAbstracted data were derived from a random sample of 30 patients enrolled in the Secondary Prevention of Small Subcortical Strokes Trial (SPS3). The abstracted data included neuroimaging imaging results, cardiac evaluation (ECG, TTE and TEE) and other laboratory test. Clinical description and past medical history also included. A consensus meeting was held among the SiGN phenotype committee to develop a series of 30 data mapping rules for the creation of a computer algorithm to systemically input abstracted data into the Causative Classification of Stroke system (CCS) for subtyping. Causative Classification of Stroke system is a computerized, evidence-based algorithm that provides both causative and phenotypic stroke subtypes in a rule-based manner. The independent adjudicator then independently reviewed the SPS3 abstracted data for subtyping using the CCS system. Using the 30 data mapping rules created by the SiGN phenotype committee, the adjudicator then subtyped the abstracted data into the CCS system. A number of discrepancies were noted and reviewed with the phenotype committee. An updated series of data mapping rules were created followed by adjudication of abstracted data derived for another random sample of 30 patients enrolled in SPS3.Results:15 out of 30 of the cases had identical subtyping13 – small artery occlusion evidentUndetermined unknown – other cryptogenicCardio-aortic embolism possibleTable one: Comparison of independent adjudication vs. unedited data mapping rules(have an excel )Table 2: Comparison of independent adjudication vs. edited data mapping rules.Conclusion: The development of a data mapping algorithm allows for a systematic way of abstracting data from multicenter randomized trials for the classification of stroke subtypes. 12015- Does sex matter? Sex differences in risk factors and types of strokeThe relative distribution of stroke risk factors, as well as ischemic stroke subtypes, in men compared with women is relatively unknown. While over half of strokes in the U.S. occur in women, and lifetime stroke risk is higher among them, relatively little attention has been paid to gender differences that may be important for stroke risk and treatment. The proposed study will compare the distribution of risk factors for ischemic stroke in men and women overall, as well as by ischemic stroke subtype (assessed by CCS(. In addition, difference in distribution for the subtypes will compared by sex. This paper will contribute to our understanding of the similarities and differences in risk factors for stroke in men and women, as well as whether stroke subtypes substantially differ by sex.12016-Deep Cerebral Phenotypes Meta-AnalysisSmall vessel ischemic stroke and deep intracerebral hemorrhage share numerous clinical characteristics. The attributable risk of hypertension for small vessel ischemic stroke is approximately 70% (compared to 20-40% for other major stroke subtypes) and 60% for deep intracerebral hemorrhage. The location of the strokes are also very similar involving the territories of small perforating vessels in the deep cerebral regions of the brain. Both have double the incidence rate among blacks than whites and have a greater importance of hypertension as a risk factor at younger ages. The odds ratio for hypertension in other subtypes increases with age while the odds ratio for small vessel ischemic stroke and deep ICH is greatest at younger age and decreases with age although still present in older ages. The deep cerebral phenotype proposal was originally discussed at the ISGC meeting in Cincinnati, OH in January, 2010. Since that time, several major GWAS of ICH studies have been completed with location specific data and as the SiGN data is completed, additional data from small vessel ischemic stroke will become available along with risk factor information. The specific aim of the proposal is to determine if specific genetic variants are associated with risk of or protection from deep cerebral phenotypes of small vessel ischemic stroke, deep intracerebral hemorrhage and deep white matter hyperintensity. 12017- Using Publicly Available Controls for GWAS StudiesDue to small effect sizes of associated SNPs, GWAS studies of complex diseases generally require very large sample sizes. An attractive strategy for increasing sample size is to direct genotyping resources towards genotyping cases only and to use publicly available controls if possible. Although potentially efficient and cost-effective, this strategy does entail multiple challenges to ensure comparability of genotyping between cases and controls, particularly when multiple control groups are utilized that have been genotyped on different platforms.The goal of this paper is to describe procedures for utilizing publically available genotypes for large GWAS studies. Using the SiGN Network as an example, we will describe considerations for selection of previously genotyped control datasets, genotype cleaning of cases, and procedures for merging newly and previously genotype data together that ensure comparability of the data. This paper will provide guidelines to the genetics community for the use of previously genotyped controls in GWAS studies. This paper will not present any association results. Rather, its purpose is to describe analytic and data cleaning strategies for using publicly available genotype data, using SiGN as an example.12018- GWAS of plasma levels of the hemostatic proteins TAFI, FSAP and t-PAIn the Sahlgrenska Academy Study on Ischemic Stroke (SAHLSIS), a case control study of ischemic stroke at 18-70 years of age, we have shown increased plasma levels of t-PA antigen as well as two more recently discovered prothrombotic hemostatic factors, TAFI and FSAP, in cases compared to controls. While the basis for inter-individual variation in protein concentration and/or activity for each of these hemostatic proteins are partially explained by known genetic variation, much of this variation still remains unresolved. We have recently performed a GWA study with exome content at the Broad Institute on part of MDC and SAHLSIS, with the remaining samples from SAHLSIS included in the SiGN GWA study. We propose to use the combined GWA data from the Broad Institute from SAHLSIS+MDC and GWA data from SiGN from SAHLSIS to seek novel genetic variants that regulate the respective hemostatic plasma protein level/activity. Preliminary GWAS results for TAFI and FSAP activity levels in SAHLSIS from the genotyping at the Broad Institute revealed gene variants passing the significance threshold located in their respective genes (top P= e-19 to e-51) as well as several interesting candidate loci (P= e-6 to e-8). For t-PA, several candidate loci were identified, with the top hit on chromosome 10 having a P-value of 7.8e-8. We would now like to increase the sample size to gain statistical power to detect possible novel associations with common SNPs and the rare variants. Therefore, we propose to include the samples from SAHLSIS that have been analyzed with regard to protein levels and that now have been genotyped within SiGN (n=444). We are currently also in the process of analyzing 2,300 additional MDC samples for the three proteins levels described above using the same assays as employed in SAHLSIS. With this sample size (n=~3000) and assuming an additive model, the effect size in units of SD per allele that is detectable with 80% power and an alpha of 5% is approximately 0.2 for a minor allele frequency (MAF) of 0.05 and 0.1 for a MAF of 0.30. In addition to identifying possible novel SNPs that reach the genome-wide significance threshold, we will also specifically look for associations for SNPs in (or in LD with) hemostatic genes. Furthermore, we will use pathway analysis to determine if there is an over representation of any biological pathway in the associated genes. Identified loci will then be tested for association to ischemic stroke and subtypes. This will require larger samples and thus specific proposals for the latter project will be presented to the ISGC, Metastroke and/or SiGN.12019- Age-at-onset Informed analysis of ischemic stroke subtypesGenetic association studies have begun to identify the common genetic contribution to ischemic stroke. However, ischemic stroke is a syndrome with considerable heterogeneity. Much of the disease variance is explained by clinical covariates such as age and hypertension. Standard genetic association analysis may not adequately deal with these factors, and therefore different approaches that inform analysis on these important variables is likely to be beneficial. In logistic regression models, including these covariates in general has no beneficial effect [Pirinen et al, Nat Genet 2012]. Recently, covariate-informed designs have been shown to improve power when clinical covariates explain a large proportion of disease risk [Zaitlen et al, PLoS Genet 2012]. In this approach, disease risk is assumed to be a continuous trait called disease liability, and phenotype values are adjusted for clinical covariates such as age-at-onset, wth regression performed on these “posterior liabilities”. In the example of age-at-onset, this has the effect of weighting individuals more if they suffer a stroke at a younger age.We recently implemented an age-at-onset informed approach in genetic data from the WTCCC2 ischaemic stroke study. Using the approach, we identified a novel variant associated with large artery stroke and we also showed that all known association with ischemic stroke subtypes were more significantly associated using the approach. [Traylor et al, 2013 (paper enclosed)] We therefore believe that covariate-informed analyses are likely to identify other novel associations with the disease not identified by standard approaches.12020-Using Available Controls for exome studyCerebrovenous thrombosis (CVT) is an unusual (~1%) cause of stroke. However, patients are often young and current data suggests that there is an underlying genetic aetiology. We have established the Biorepository for the Etiology of Sinovenous Thromboysis (BEAST) to bring together an international collaboration for the collection of highly phenotyped CVT cases. Around 26 investigators from 14 countries have contributed towards this collaboration. We have aimed to undertake an ExomeCore genome analysis on the first 500 cases. 12021-GWAS of all stroke and stroke subtypes (CCS)12022-Age of onset GWAS of stroke (and subtypes?)12023-Sex-specific GWAS of stroke (and subtypes?)12024-Race/ethnic group-specific GWAS of stroke (and subtypes?)12025-Genetic determinants of CCS vs. TOAST-defined stroke subtypes12026- The STARNET-IS (STARNET and Ischemic Stroke) studyClinical studies considering intermediate phenotypes in the form of genomic data isolated from disease relevant tissues have greatly enhanced our understanding of complex traits like atherosclerosis and type2 diabetes primarily by enabling inference of disease-driving gene networks. The architecture and activity of these networks help to better understand the combined effects of many (as opposed to isolated) genetic and environmental risk factors in complex disease biology. In a recent study using meta-analysis including METASTROKE, Dichgans et al. showed that there are common risk variants for coronary artery disease (CAD) and ischemic stroke (IS), particularly for large artery stroke (LAS). We propose to apply a multi-tissue network-based approach to achieve a more comprehensive understanding of the shared biology underlying CAD and IS/LAS.Dr. Bj?rkegren’s Stockholm-Tartu Atherosclerosis Reverse Network Engineering Task (STARNET) study today comprises DNA genotype (>900K SNPs) and RNA samples from >600 well-characterized CAD patients and 100 non-CAD controls obtained during coronary artery bypass grafting and other forms of open-thorax surgery. RNA sequence data (25M read depth) from the atherosclerotic and non-atherosclerotic arterial wall, liver, skeletal muscle and visceral and subcutaneous fat and whole blood of these STARNET subjects have now been generated.Using the STARNET data, we hypothesize that we can expand our understanding of the common genetic/genomic basis of CAD and IS/LAS primarily by using meta analysis GWA datasets of stroke and CAD to study the relative risk-enrichments, including common GWA risk loci, of vascular and metabolic networks identified in STARNET. By providing insights into the disease-network architectures, novel molecular mechanisms and risk factors for IS/LAS can be revealed.Populations to be studied: Cases of all strokes and controls076200Fig 1. Statistical Power to detect eQTL on STARNET cohort00Fig 1. Statistical Power to detect eQTL on STARNET cohortMethod: We will apply an in-house pipeline to quantify eQTLs on stroke GWAS hits, followed by pathway and gene network analysis.Sample size: individual level phenotype (case-control status, age, gender, etc.) and genotype (N=all available sample); or full GWAS results (e.g. pvalue and beta of 1000G imputed SNPs)Power and analysis plan: (1) if we receive individual level genotype and phenotype data, we will conduct 1000G imputation and GWAS on stroke traits; (2) we will then filter the GWAS result by pvalue (e.g. 0.05); (3) focused on the derived SNP and conduct eQTL discovery (e.g. calculating gene expression vs. genotype using EMMAX software); (4) quantify FDR through permutation (5) pathway and gene network analysis to infer the etiological mechanisms of the GWAS SNPs related to IS.12027- Genetic predisposition to Hemorrhagic transformations in ischemic stroke patients with or without rt-PA treatment.Thrombolytic therapy with rt-PA is the only EMA-approved therapy during the acute phase of ischemic stroke. Despite its proven benefits, 2 to 6% of the patients suffer a very damaging symptomatic brain bleeding and 30 to 40% of patients do not recanalize, leading to poor outcome and disability after stroke. The risk of bleeding complications or Hemorrhagic Transformation (HT) is one of the main reasons why treatment is restricted to the first 4.5 hours after symptoms onset, being the total rates of any subtype of HT around 20-23% and 10-15 % for PH (Parenchymal Hematoma, the most aggressive HT) subtype. PH subtype and recanalization failure are associated with the outcome of stroke treated patients. The time from symptoms onset to rt-PA administration is directly proportional to the risk of hemorrhage. As a result, only 6-14% of stroke cases are treated with rt-PA in USA or Europe. Data from Montaner’s group shows that genetic polymorphisms are associated with HT (PH subtype) rt-PA treatment. In fact, the addition of these genetic variants in a clinical prediction algorithm greatly enhanced its predictive value for HT (PH subtype) (AUC clinical-genetic algorithm: 0.720 vs. AUC previous clinical score: 0.603; p = 0.004) (Del Rio-Espinola et al, 2012). This study was performed only with 140 polymorphisms and 1.172 patients. Therefore, we can add more polymorphisms after performing genome wide studies to improve the prediction algorithm, increasing our ability to predict HT to be potentially useful in the clinical practice. Pilot studies with 340 patients treated with rt-PA and with GWAs data showed several new candidates associated with HT and PH subtype. During this year we will have GWAs data of 1.000 subjects treated with rt-PA and HT data available.To decrease the rates of PH hemorrhages will improve the outcome of stroke patients treated with rt-PA any small reduction of the number of disabilities or in the general improvement of the outcome in ischemic stroke will impact dramatically in the health and social public system in numbers of millions of € every year.Knowledge gaps:None powerful prediction tools are available to detect which patients will have a HT after rt-PA.The percentage of ischemic stroke patients treated with rt-PA is very low (between 6-14%) due to HT risk among other causes.Plasma biomarkers as MMP9, Fibronectin, VAP-1, among others, measured before rt-PA administration are very variable and difficult to implement in the clinical practice to improve the management of fibrinolysis. Aims: 1) To find polymorphisms associated with HT and/or PH subtype in patients treated with rt-PA using the GWAs data of SiGN study.2) To analyze whether these polymorphisms are risk factors of spontaneous HT in patients not treated with rt-PA.3) To generate a clinical-genetic algorithm to predict HT and/or PH using the clinical variables and the polymorphisms associated with HT and/or PH. Hypothesis:The risk of HT seems to be genetically influenced, however none GWAs studies with enough sample size have been performed to find these genetic risk factors. To determine genetic biomarkers that could be included in a clinical-genetic prediction model could be an interesting strategy to improve the outcome of patients treated with rt-PA as has been previously described. However, the genetic tests should be performed rapidly to use these polymosphisms in the clinical practice12028- Genetic variations influencing risk of ischemic stroke in the posterior circulation vascular territoryIschemic stroke in the anterior and posterior circulation are assumed to share etiology and pathogenesis to a large extent. However, the anterior and posterior vascular beds differ in several respects: Arteries in the anterior circulation are more richly innervated by sympathetic nerve endings than arteries in the posterior circulation and several studies have shown that vessel wall endothelial function is different in these two territories. Vasodilation and vasogenic edema is more frequent in the posterior circulation and this may be one explanation for Posterior Reversible Encephalopathy Syndrome (PRES) and hypertensive encephalopathy being more prevalent in this territory. It is also known that the posterior circulation territory is more prone to have cerebral infarcts related to migraine. In addition, the posterior circulation is older from a phylogenetic aspect even though in man, the posterior circulation territory develops after the anterior circulation territory from an embryological point of view.There is a knowledge gap regarding possible genetic factors that affect the location of cerebral ischemic insults in the anterior and posterior circulation territories, respectively. We hypothesize that the above mentioned structural and functional differences have an impact on stroke pathogenesis as well as on the evolution of the acute stroke process itself, and that certain genetic variations may influence the risk of developing ischemic stroke specifically in the posterior circulation. We aim to perform a discovery study relating ischemic stroke in the posterior vascular territory defined by MRI to genome wide association data. 12029-Admixture mapping of ischemic stroke in African AmericansThe goal of this project is to estimate the local ancestry across the genome to identify regions associated with risk of ischemic stroke in African Americans. While genome-wide association (GWA) has successfully identified variants associated with ischemic stroke in European ancestry, there have only been limited GWA studies in African American population and are often underpowered to detect genome-wide significance due to the limited number of stroke cases available. We hypothesize that the higher rate of stroke in African Americans is at least partially related to African ancestry (that is, the risk alleles for ischemic stroke may be in linkage disequilibrium with chromosomal regions that are free of European genetic admixture), and that some environmental factors, such as hypertension or smoking, may modify the effects of genetic variants on the development of stroke. In this project, we propose to re-examine the genome-wide SNP array data by conducting innovative analyses that exploits bi-racial ancestry of African American (i.e. admixture mapping) using all SiGN African American cases and additional ~600 cases from non-SiGN studies. We propose to perform the following analysis:Admixture mapping using all African American stroke cases and controls (primary analysis)Admixture mapping of stroke subtypesAdmixture mapping of stroke stratified by various stroke risk factors. 12030-CCS vs. TOAST - Polygenic Score AnalysesAlthough individually significant markers in genome-wide association scans (GWAS) explain limited heritability of complex traits, evidence has been accruing that a considerable proportion of phenotypic variation can be explained by a collection of markers not achieving significance. Thus, while most of the specific genes underlying complex traits have yet to be identified; it is likely that many are represented on current genotyping products. Polygenic score analysis has recently generated much interest for assessing the explanatory power of a collection of markers. Typically a GWAS is conducted on an initial sample, and the markers are ranked by their evidence for association, usually their P-values. The top markers are then assessed for their predictive value in other cohorts.As applied to SiGN: Utilizing TOAST and CCS often results in differing subtype classifications. While the reasons for these differences are complex, potentially a limited number of SNPs could accurately predict subtypes across both systems, thereby implicating true subtype-specific risk variants. Here we propose an evaluation utilizing polygenetic models for CCS and applying to TOAST, and then vice-versa; this to:1) to provide insights into the degree of genetic?overlap between subtypes defined by CCS vs. TOAST, and;2) identify as subset of markers that accurately predict subtypes across both classification schemes. There are many potential analyses that could be performed. Analyses: Cases stratified by each of the 5 major CCS subtypes would undergo GWAS (cases vs. controls) with all SNPs ranked by p-value. Using different p-value thresholds (e.g. p < 0.10, p < 0.05, p < 0.01, p < 0.001, p < 0.0001, etc.) the performance of each set of SNPs would be assessed in predicting TOAST specific subtypes.? Similar analyses would be performed starting with TOAST subtypes and then evaluating predictive correlation with CCS subtype. (GOAL - for these analyses to be performed for the AHA ISC abstract due August 12, 2014) Other Potential Analyses: 1. Cases classified by each of the 5 major CCS subtypes would undergo GWAS (cases vs. controls) with the most significantly associated (e.g. top 1000) SNPs identified by subtype as ranked by p-value. These same subtype-specific SNPs would then be evaluated in all SiGN cases to determine how well these SNPs predict TOAST subtype. Similar analyses would be performed starting with TOAST subtypes and then evaluating correlation with CCS subtype. Overlapping SNPs would be identified. The process could be performed iteratively with a goal to identify a fixed number of highly correlated SNPs. 2. Future Replication: The final lists of subtype-specific SNPs could then be evaluated in another TOAST population (e.g. METASTROKE) to assess subtype-specific correlation.12031-Genetic determinants of MRI confirmed lacunar stroke subtypesGenomewide association studies (GWAS) have begun to identify the genetic variants associated with to ischaemic stroke.PEVuZE5vdGU+PENpdGU+PEF1dGhvcj5Ib2xsaWRheTwvQXV0aG9yPjxZZWFyPjIwMTI8L1llYXI+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ADDIN EN.CITE PEVuZE5vdGU+PENpdGU+PEF1dGhvcj5Ib2xsaWRheTwvQXV0aG9yPjxZZWFyPjIwMTI8L1llYXI+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ADDIN EN.CITE.DATA However to date these have been largely in cardioembolic and large artery stroke and there are no published replicated genetic variants that confer risk of lacunar stroke in Caucasian populations. Although family history data suggest genetic predisposition may be particularly important for lacunar stroke (3), estimates of the heritability of lacunar stroke from GWAS studies have been very low compared to that of the other subtypes.(1,2) Two main factors may contribute to this lack of success. One important such factor is disease heterogeneity. Pathological studies have different vascular lesions in patients presenting with lacunar stroke, with two main pathologies reported, namely focal microatheroma and a diffuse small vessel arteriopathy.(4) The former has been associated with larger single lacunar infarcts and the later with multiple smaller lacunes and leukoaraiosis.(4,5). These two subtypes have been shown to have differing risk factor profiles.(6) These sources point to the existence of pathophysiological subtypes of lacunar stroke, each of which might be presumed to have distinct genetic susceptibility factors. Another factor might be inadequate disease classification. Lacunar infarcts are small and frequently not seen on CT. Diagnostic accuracy is much improved with MRI. Previous workWe have carried out GWAS in 1353 MRI confirmed lacunar stroke and 7397 controls. In this population of MRI confirmed lacunar stroke, in contrast to the WTCCC2 data, using GCTA there was a significant heritability of each of the two subtypes of lacunar stroke; 0.27 (0.10), for the multiple lacunar strokes /leukoaraiosis variant and 0.18 (0.10)) for the isolated lacunar stroke subtype. In a GWAS association analysis we found a different pattern of associations for the two subtypes.The aims of the current application are to:Increase the sample size of patients with MRI defined lacunar stroke, classified into sub-phenotypes. Use the added sample size to identify associations with the MRI confirmed lacunar stroke and its subtypes.12032-Subtype-specific etiology mechanisms for ischemic stroke using genome wide pathway analysisThe molecular mechanism of stroke has been extensively studied usually on single molecule/signaling pathway basis. However, none of the potential therapy target candidate has been successfully translated into clinical application. Unknown mechanism has been suggested to yet be uncovered.Currently, GWAS analysis of worldwide SiGN stroke genotype data has been in steady progress. We will soon have top stroke susceptible loci identified. As stroke is a complex trait, these top few loci would be expected to only account for a very small amount of etiology mechanism while the effect of the rest genome variants would have been ignored. Genome wide pathway analysis can aggregate the otherwise neglected effect of genetic variants which do not normally exhibit genome wide significant p values, and has been successfully used in other complex diseases to uncover new pathological mechanism, such as Alzheimer’s disease, autism, and mood disorders. We propose to use genome wide pathway analysis to(1). investigate and contrast the etiology mechanisms among different subtypes of ischemic stroke; (2). Compare the identified molecular pathways with literature to reconcile existing knowledge on stroke mechanisms and discover novel mechanisms. The results would potentially guide discovery and development of novel molecular targets for stroke prevention, treatment and recovery.Our analysis is planned as below:Summary results of the Meta-GWAS across ethnicities on the overall stroke and stroke subtypes will be obtained from the primary SiGN analysis output for the SiGN primary paper of SiGN;The SNP level association statistics for the region of each gene will be aggregated to generate gene level association statistics; Gene will be assigned to pathways according to publically/commercially annotated pathway database, such as Ingenuity (commercial pathway database) or the Molecular Signatures Database (MSigDB)(public database).The genome-wide gene level association statistics will be subjected to genome wide Gene Set Enrichment Analysis (GSEA). Genes are assigned to pathways according to public Molecular Signatures Database (MSigDB)(). A FDR corrected p-value will be obtained for each pathway. A FDR corrected p-value 0.1 can be an initial cut-off threshold. The enriched pathways for overall stroke and each subtype can thus be obtained.In contrast to above Gene Set Enrichment Analysis (GSEA), which does not use pre-set threshold to filter out SNPs/genes, an alternative approach will also be used to ensure the reliability of the analysis. In this parallel analysis, only the top SNPs (e.g. 1000 SNPs) from each summary Meta-GWAS result will be mapped to genes; Pathway enrichment analysis will then be performed only for these top signals using Ingenuity? Pathway Analysis (IPA?, QIAGEN Redwood City). The identified enriched pathways for overall stroke and each subtype can then be compared with each other to identify overlapping and unique subtype-specific pathways.Known stroke related molecular pathways will be compiled from literatures using PubMed database and related text search tool. These known stroke-related pathway will be compared with our newly identified pathways to prioritize known pathways, and identify novel pathways for further investigation.12033-Early‐Onset Stroke: an Extreme Phenotype to Identify Rare Variants in Ischemic StrokeThe genetic architecture of ischemic stroke likely includes rare‐ or low‐frequency variants with highpenetrance and large effect sizes. Analyses focusing on the exome may increase the likelihood of identifying these rare functional variants providing important insights into disease pathogenesis and implicating potential drug targets. Because there are few well subtyped cases enriched with rare variation to evaluate this premise, our studies will focus on early‐onset stroke, an extreme phenotype thought to be enriched with rare variants, this, in cases that have been carefully subtyped.Specific Aims:1. Perform exome‐specific rare variant analyses on early‐onset ischemic stroke cases and matched controls from the GEOS study to discover variants and genes that contribute to ischemic stroke susceptibility.Stage 1 ‐ Discovery: Exomic genetic data from 840 subjects (all ≤ 49 years of age) with early‐onset stroke and 880 matched stroke‐free controls will be analyzed to identify rare risk variants. All GEOS samples will undergo exome‐chip genotyping. To identify low‐frequency variants and genes contributing to stroke susceptibility, analyses of all ischemic stroke and as stratified by stroke subtype (TOAST/CCS), ethnicity, and gender will beperformed.2. Replicate identified associations in an independent population of young‐onset stroke cases andcontrols.Stage 2 ‐ Replication: Utilizing 1,119 young‐onset (≤55 years) stroke cases from the NINDS SiGN, JHS and FHS and 7,592 controls from the NINDS SiGN, JHS, FHS and ESP we will perform replication studies of the identified variants and genes found to be associated with stroke risk as identified in Specific Aim 1. JHS and FHS cases will undergo TOAST/CCS sub‐typing. SiGN cases have CCS and most have TOAST.3. Determine if variants, genes, or pathways associated with ischemic stroke or its subtypes in youngonset stroke are also associated with ischemic stroke or its subtypes in older‐onset stroke.Stage 3 ‐ Extension to older‐onset stroke: Utilizing 4,795 older‐onset (>55 years) stroke cases from the JHS, FHS and NINDS SiGN and 6,795 controls from the NINDS SiGN, JHS, FHS, and ESP, we will assess the variants and genes associated with early‐onset stroke in Specific Aims 1‐2 for their association with older‐onset stroke.12034-Association between lipid-level associated genetic variants and manifestations of cerebral small vessel diseaseDr. Anderson’s K23, entitled “Genetic Analyses of Lipids in Cerebral Small Vessel Disease” seeks to clarify the role of circulating lipid fractions on the risk of intracerebral hemorrhage, white matter disease, cerebral microbleeds, and lacunar stroke, as manifestations of small vessel disease arising from lipohyalinosis and/or angionecrosis of the cerebral arterioles. Understanding how lipid levels could modify the risk of CSVD could impact how lipid-lowering therapies are implemented in at-risk populations.We have demonstrated an association between decreasing LDL and total cholesterol levels and risk of recurrent deep intracerebral hemorrhage. Research using the Framingham cohort has demonstrated an association between decreased total cholesterol levels and both deep and lobar cerebral microhemorrhages. Investigators for the NOMAS cohort have noted a trending association between decreasing total cholesterol levels and increasing white matter hyperintensities, but only in individuals with the APOE e4 genotype. We propose to use genetic scores informed by well-powered GWAS of lipid-related traits to examine the impact of variants associated with HDL, LDL, TG, and total cholesterol levels on the risk of small vessel stroke, number and location of cerebral microhemorrhages, and severity of white matter hyperintensities on MRI. Using genetic variation to stratify lipid exposures has several advantages. First, there is implied causality of associations between lipid level-associated variants and CSVD phenotypes. Second, genetic variants are not impacted by diet or statin exposure. Finally, lipid level-associated variants provide information on lifetime lipid exposures, rather than at specific time points when lipid profiles are measured.The results of this analysis will provide new information about the role of lipids in the pathogenesis and progression of cerebral small vessel disease, will provide information on what phenotypes might be particularly susceptible to risk imparted by lipid levels, and will help identify clinical markers of CSVD severity that could be used to stratify CSVD risk for future trials aimed at using individualized strategies to set lipid goals to minimize risk of both large and small vessel disease. 12035-Shared Heritability of Cerebral Small Vessel Disease Phenotypes Using Genome-wide Summary StatisticsIn anticipation of coming opportunities for whole genome sequencing (WGS) in complex diseases, the ISGC will need to determine how best to select cases for future WGS studies. One hypothesis attractive to the stroke community is that cerebral small vessel disease, which manifests through a variety of phenotypes, has a common biological underpinning. If this hypothesis were to be true, then combining samples for future rare variant genetic association studies such as WGS would substantially increase power for variant discovery. As a first step toward determining whether this hypothesis may be correct, we propose to measure the degree to which multiple manifestations of cerebral small vessel disease share a common genetic contribution.It is now possible to determine the degree of genetic overlap between traits by leveraging atechnique called LD score regression, which allows accurate estimation of polygenicity usingsummary GWAS statistics (Am J Hum Genet. 2014 Nov 6;95(5):535-52 and ). We will utilize this approach in this project,testing shared genetic liability among clinical and radiographic manifestations of cerebral smallvessel disease. A strength of this approach is that it employs summary statistics and can inferpopulation structure as well as duplicate individuals using backward inference of regional linkage disequilibrium. We are requesting GWAS summary statistics using 1000 Genomes imputation from METASTROKE and SiGN for all stroke subtypes by TOAST and/or CCS criteria, and the ISGC ICH GWAS for deep and lobar ICH. These summary datasets will be used to determine univariate and multivariate heritabilities, as well as correlation matrices for shared heritabilities among each pair of traits as well as across all analyzed traits. Our overall goal is to determine whether pooled analysis (either through existing GWAS data or future sequencing efforts) of individuals exhibiting different manifestations of cerebral small vessel disease is scientifically justified, leveraging larger sample sizes than could be accomplished in case-control studies of individual traits. All analyses will be carried out at Massachusetts General Hospital using the Broad Institute computational resources. Because our project examines genetic overlap among multiple cohorts in aggregate, no formal replication cohort is required. An identical proposal has been submitted and approved by the METASTROKE steering committee.12036- Copy Number Variants and Risk of Ischemic Stroke: a Pilot StudySiGN has successfully performed a GWAS including ~14,000 individuals with Ischemic stroke and ~30,000 control subjects. Among these, all stroke cases and a subset of controls were genotyped using the 5 million Illumina chip by CIDR. By design the 5 million Illumina chip focuses on SNPs and contains few other genetic variations, with only about 100 indels included for example. While copy number variants (CNVs) were poorly represented, methods currently exist that can utilize the existing SiGN data to evaluate for CNVs. We hypothesize that there are large CNVs that contribute to stroke risk and that this may be detected by analyzing signal intensity and independent variance of data from the 5 million Illumina chip results in SiGN. CNV analyses of this whole-genome data will permit us to: 1) study?rare genetic variation; 2) to?detect?unforeseen?variants, ?and; 3) to study large variants, possibly affecting several genes (large CNVs may cover >10 genes).Samples will be analyzed with pennCNV to identify CNV. CNVs will be prioritized with regard to rarity (<0.001 or <0.01 in own and public control samples) and functionality (“genic” or “exonic” CNVs affecting coding sequences). Following CNV detection in the cases and controls, association analyses will be performed evaluating all stroke and stratified by subtype (TOAST and CCS) as the number of study subjects in each strata will permit. In addition to age, ethnicity, and gender, additional covariates will include standard vascular risk factors. Pending these results, a larger study using intensity data from other SiGN centers will be proposed allowing other interested groups to participate. Power analysis: The proposed pilot?includes the study of?very rare (or even unique) genetic variants, power calculations?would follow guidelines for gene-based association studies, as used in exome studies of rare variation (SKAT-O,?etc.). We will also perform functional enrichment analysis of all CNV findings to detect association with predefines gene sets (Pathways, GeneOntology biological processes). Given that this pilot’s primary goal is to test the feasibility of successfully detecting CNVs in the SiGN data, we do not present/perform detailed power calculations.12037- Analysis of platelet genetic risk variants in relation to risk of stroke in HispanicsThrough HCHS/SOL, we have performed a GWAS analysis of platelet count in ~13,000 Hispanic/Latinos and identified a half dozen novel variants (either new loci or new variants at known platelet loci) not previously reported in prior European GWAS. Several of these new platelet count variants appear to be specific to Native American ancestral populations and therefore can only be evaluated in Hispanic or Native American samples. Since platelets are related to thrombosis, and SiGN has 942 Hispanic/Latino stroke cases, we are interested in assessing whether these new platelet-associated variants might be related to stroke risk in SiGN Hispanics. We propose assessing each platelet variant individually with respect to stroke risk as well as combining them in a platelet “genetic risk score”.12038- Identification of novel genes and variants associated with stroke phenotypes by using CSF and plasma analyte levels as quantitative traitsThere is a clear need to develop and apply additional approaches if we are to solve the complete genetic architecture of stroke. One alternative approach is to use intermediate quantitative traits as phenotypes for genetic studies. Our previous research has successfully used protein levels to identify novel variants and genes associated with Alzheimer’s disease ADDIN PAPERS2_CITATIONS <citation><uuid>C681AA3F-85D1-4A80-8011-B83DC6756737</uuid><priority>8</priority><publications><publication><uuid>5F8E5C4F-ED2C-4ADF-BC13-E3E5C0CDD086</uuid><volume>6</volume><accepted_date>99201007291200000000222000</accepted_date><doi>10.1371/journal.pgen.1001101</doi><startpage>e1001101</startpage><publication_date>99201009001200000000220000</publication_date><url> associated with cerebrospinal fluid phospho-tau levels influence rate of decline in Alzheimer's disease.</title><location><html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"/><title>Sorry...</title><style> body { font-family: verdana, arial, sans-serif; background-color: #fff; color: #000; }</style></head><body><div><table><tr><td><b><font face=times color=#0039b6 size=10>G</font><font face=times color=#c41200 size=10>o</font><font face=times color=#f3c518 size=10>o</font><font face=times color=#0039b6 size=10>g</font><font face=times color=#30a72f size=10>l</font><font face=times color=#c41200 size=10>e</font></b></td><td style="text-align: left; vertical-align: bottom; padding-bottom: 15px; width: 50%"><div style="border-bottom: 1px solid #dfdfdf;">Sorry...</div></td></tr></table></div><div style="margin-left: 4em;"><h1>We're sorry...</h1><p>... but your computer or network may be sending automated queries. To protect our users, we can't process your request right now.</p></div><div style="margin-left: 4em;">See <a href="">Google Help</a> for more information.<br/><br/></div><div style="text-align: center; border-top: 1px solid #dfdfdf;">&copy; 2013 Google - <a href="">Google Home</a></div></body></html></location><submission_date>99201004221200000000222000</submission_date><number>9</number><institution>Department of Psychiatry, Washington University School of Medicine, St. Louis, Missouri, USA. cruchagc@psychiatry.wustl.edu</institution><subtype>400</subtype><bundle><publication><title>PLoS genetics</title><type>-100</type><subtype>-100</subtype><uuid>C9001744-7CBC-4328-B9D5-4235410AB71B</uuid></publication></bundle><authors><author><firstName>Carlos</firstName><lastName>Cruchaga</lastName></author><author><firstName>John</firstName><middleNames>S K</middleNames><lastName>Kauwe</lastName></author><author><firstName>Kevin</firstName><lastName>Mayo</lastName></author><author><firstName>Noah</firstName><lastName>Spiegel</lastName></author><author><firstName>Sarah</firstName><lastName>Bertelsen</lastName></author><author><firstName>Petra</firstName><lastName>Nowotny</lastName></author><author><firstName>Aarti</firstName><middleNames>R</middleNames><lastName>Shah</lastName></author><author><firstName>Richard</firstName><lastName>Abraham</lastName></author><author><firstName>Paul</firstName><lastName>Hollingworth</lastName></author><author><firstName>Denise</firstName><lastName>Harold</lastName></author><author><firstName>Michael</firstName><middleNames>M</middleNames><lastName>Owen</lastName></author><author><firstName>Julie</firstName><lastName>Williams</lastName></author><author><firstName>Simon</firstName><lastName>Lovestone</lastName></author><author><firstName>Elaine</firstName><middleNames>R</middleNames><lastName>Peskind</lastName></author><author><firstName>Ge</firstName><lastName>Li</lastName></author><author><firstName>James</firstName><middleNames>B</middleNames><lastName>Leverenz</lastName></author><author><firstName>Douglas</firstName><lastName>Galasko</lastName></author><author><lastName>Alzheimer's Disease Neuroimaging Initiative</lastName></author><author><firstName>John</firstName><middleNames>C</middleNames><lastName>Morris</lastName></author><author><firstName>Anne</firstName><middleNames>M</middleNames><lastName>Fagan</lastName></author><author><firstName>David</firstName><middleNames>M</middleNames><lastName>Holtzman</lastName></author><author><firstName>Alison</firstName><middleNames>M</middleNames><lastName>Goate</lastName></author></authors></publication><publication><uuid>DEA88C2C-4FB4-41C4-BD21-6C78FA25B0B4</uuid><volume>21</volume><doi>10.1093/hmg/dds296</doi><startpage>4558</startpage><publication_date>99201210151200000000222000</publication_date><url> fluid APOE levels: an endophenotype for genetic studies for Alzheimer's disease.</title><location><html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"/><title>Sorry...</title><style> body { font-family: verdana, arial, sans-serif; background-color: #fff; color: #000; }</style></head><body><div><table><tr><td><b><font face=times color=#0039b6 size=10>G</font><font face=times color=#c41200 size=10>o</font><font face=times color=#f3c518 size=10>o</font><font face=times color=#0039b6 size=10>g</font><font face=times color=#30a72f size=10>l</font><font face=times color=#c41200 size=10>e</font></b></td><td style="text-align: left; vertical-align: bottom; padding-bottom: 15px; width: 50%"><div style="border-bottom: 1px solid #dfdfdf;">Sorry...</div></td></tr></table></div><div style="margin-left: 4em;"><h1>We're sorry...</h1><p>... but your computer or network may be sending automated queries. To protect our users, we can't process your request right now.</p></div><div style="margin-left: 4em;">See <a href="">Google Help</a> for more information.<br/><br/></div><div style="text-align: center; border-top: 1px solid #dfdfdf;">&copy; 2013 Google - <a href="">Google Home</a></div></body></html></location><institution>Department of Psychiatry, Washington University School of Medicine, St Louis, MO 63110, USA.</institution><number>20</number><subtype>400</subtype><endpage>4571</endpage><bundle><publication><title>Human Molecular Genetics</title><type>-100</type><subtype>-100</subtype><uuid>B3E00959-F0D5-4CDC-91AC-156F5EB2FF94</uuid></publication></bundle><authors><author><firstName>Carlos</firstName><lastName>Cruchaga</lastName></author><author><firstName>John</firstName><middleNames>S K</middleNames><lastName>Kauwe</lastName></author><author><firstName>Petra</firstName><lastName>Nowotny</lastName></author><author><firstName>Kelly</firstName><lastName>Bales</lastName></author><author><firstName>Eve</firstName><middleNames>H</middleNames><lastName>Pickering</lastName></author><author><firstName>Kevin</firstName><lastName>Mayo</lastName></author><author><firstName>Sarah</firstName><lastName>Bertelsen</lastName></author><author><firstName>Anthony</firstName><lastName>Hinrichs</lastName></author><author><lastName>Alzheimer's Disease Neuroimaging Initiative</lastName></author><author><firstName>Anne</firstName><middleNames>M</middleNames><lastName>Fagan</lastName></author><author><firstName>David</firstName><middleNames>M</middleNames><lastName>Holtzman</lastName></author><author><firstName>John</firstName><middleNames>C</middleNames><lastName>Morris</lastName></author><author><firstName>Alison</firstName><middleNames>M</middleNames><lastName>Goate</lastName></author></authors></publication></publications><cites></cites></citation>. The benefit of using endophenotypes such as these is that they have greater statistical power because they are more closely linked to the genetic regulation.We have CSF and plasma levels for more than 180 analytes (Myriad RBM Discovery MAP; ) measured in 800 individuals in whom we also have access to corresponding genetic information (GWAS, exome-chip and sequencing data). Some of these analytes are known to be important for stroke: baseline matrix metalloproteinase-9 (MMP-9) plasma levels were found to be an independent predictor of hemorrhagic transformation after treatment with tPA—with an odds ratio (OR) of 9.62. Plasma C-reactive protein (CRP) levels proved to be a significant and independent predictor of mortality following tPA use and IL-6 levels predicted neurological deterioration after stroke. We propose to analyze whether the most significant genetic associations for all the RBM analytes also show an association with stroke risk or vice-versa. These studies may help identify potential biomarkers for stroke in general or specific subtypes of stroke; or identify potential mechanisms involved in stroke risk.12039- MEGASTROKEWe will investigate the association of all variants with ischemic stroke and subtypes not restricted to variants reaching genome-wide significance or suggestive significance in previous individual studies. To this end we want to combine all available summary statistics to improve the power to detect an association with ischemic stroke and its subtypes LVD, CE and SVD. Studies that have already agreed to participate in this effort are the cross-sectional METASTROKE study (10,307 cases / 17,326 controls) and the prospective CHARGE IS study (3,028 incident IS cases). Other studies identified by us as being able to contribute to this effort and who have in part already agreed to participate into this proposal include: SIFAP, CADISP, GLASGOW, COMPASS, deCODE, NOMAS, LUND, MDC, VHIR-FMT-BARCELONA, HVH-1, RACE, SAHLSIS, INTERSTROKE, UTRECHT and JHS. In total, approx. 30,000 IS cases should be available for analysis.All available summary statistics will be meta-analyzed using a fixed effects inverse-variance model. HRs from prospective cohorts and ORs from cross-sectional cohorts will be meta-analyzed together. Stroke subtypes will be integrated with TOAST subtyping, if available. We are aiming to include all available samples into our discovery analysis, regardless of imputation panel to maximize our power to detect an association.12040- Meta-Analysis of GWAS of intracranial saccular aneurysms and large artery atherosclerosis ischemic stroke Intracranial saccular aneurysm and large artery atherosclerosis ischemic stroke share with one anotherthe vascular territory involvement. For both large artery atherosclerosis ischemic stroke and intracranialsaccular aneurysms, disease of the wall and within the wall with respect to atherosclerosis, variation inelastin and other structural wall components are observed and a strong influence of cigarette smokingand hypertension exists. The two vascular subtypes already share two common genetic risk loci: the 9p21and HDAC9 locus.Heretofore, stroke has been classified by major clinical features such as hemorrhage and ischemia and the territory of involvement which is important in discovering genetic risk factors specific to theirdevelopment. Here we propose to evaluate the potential of a large artery risk factor which predisposesindividuals to a potential spectrum of secondary conditions that lead to the phenotypic variation ofsaccular intracranial aneurysms and large artery atherosclerosis ischemic stroke.12041-Sex differences in the causal relationship between diabetes and stroke It has long been appreciated that diabetes confers a greater coronary hazard among women than in men (Peters et al. Diabetologia 2014; Huxley et al. BMJ 2006) and more recently stroke (Peters et al, Lancet 2014). The mechanisms that underlie these sex differences in the associations between diabetes and cardiovascular disease risk are largely unknown. We are therefore planning to conduct a Mendelian randomization analysis using summary level data to examine whether the causal relationship between adiposity traits, metabolic traits, diabetes and CHD, and stroke is different between men and women. In order to conduct these analyses, we request from NINDS SiGN a list of SNPs, based on DIAGRAM (Morris et al, 2012), MAGIC (Manning et al, 2012) and GIANT (Locke et al, 2015), with their betas and standard error separately for men and women.12042-Variants in myeloperoxidase genes influences risk of cerebral small vessel diseaseGrowing evidence supports the role of inflammation in cerebrovascular disease, particularly via its role in atherosclerotic initiation and progression, and as a critical mediator of brain injury response. Several other known clinical risk factors for stroke, including diabetes, obesity, hypertension and thromboembolism are associated with inflammatory profiles. Myeloperoxidase (MPO) is an enzyme of the innate immune system, released from lysomes of activated neutrophils and monocytes with pleiotropic roles in promoting oxidative stress, inflammation, and inducing endothelial dysfunction. Elevations in MPO levels have been associated with clinical and imaging magnifications of cerebral small vessel disease (increased WMHV and CMB number), and correlated with increased risk of stroke in Fabry disease. However, challenge remains in establishing whether inflammatory changes are causal in pathophysiology of disease, or are instead a consequence of ongoing injury as circulating biomarkers are poor proxy measures of the inflammatory response, and subject to considerable variance. Preliminary analysis has demonstrated an association between genetic determinants of elevated MPO levels (14 independent loci) and increased ICH risk. Furthermore, when examining risk of ischemic stroke, our severely underpowered preliminary analysis using GASROS casesd and controls suggests that the magnitude and direction of effect are consistent with what we have seen in ICH. 12043- The contribution of common variants in COL4A1, COL4A2, NOTCH3, HTRA1, TREX1 and CECR1 genes to common cerebrovascular phenotypes- Meta-analyses of existing GWAS data to determine whether common SNPs in COL4A1, COL4A2, NOTCH3, HTRA1,TREX1 and CECR1 genes are associated with a range of sporadic forms of cerebrovascular disease: ischaemic stroke and subtypes (TOAST/CCS), intracerebral haemorrhage (ICH) and subtypes (lobar and non-lobar), white matter hyperintensities (WMH), and brain microbleeds (BMBs). We hypothesize that the associations will be specific to – or at least strongest with – cerebral small vessel disease phenotypes (lacunar ischaemic stroke, deep ICH, deep BMBs and WMH). This application builds on our previous work which successfully used the approach to study the COL4A1 and COL4A2 genes and demonstrated their association with cerebral small vessel disease. Background: Cerebral small vessel disease (SVD) affects the small arteries, arterioles, venules and capillaries of the brain, and leads mainly to lesions in subcortical brain structures – ischaemic lacunar strokes, deep intracerebral haemorrhages (ICH), white matter hyperintensities (WMH) and deep brain microbleeds (BMBs). Currently mutations in at least 6 genes (COL4A1, COL4A2, NOTCH3, HTRA1,TREX1,CECR1) are known to cause rare familial forms of SVD but the genetic risk factors of common sporadic forms remain largely unknown. Genes causing rare familial forms of cerebral SVD may also contain variants conferring risk for sporadic cerebral SVD. Our recent study has shown a significant association between common SNPs in the COL4A2 gene and sporadic deep ICH, and a similar trend for association with other SVD phenotypes (lacunar ischaemic stroke and WMHs).We hypothesize that variants in NOTCH3, HTRA1, TREX1 and CECR1 may also be associated with sporadic SVD.Population, methods, sample size, power, analysis plan: We propose to undertake meta-analyses of existing, cohort level, summary GWAS data of the NINDS SiGN ischaemic stroke case-control datasets in populations of European origin (excluding cohorts included in the Metastroke dataset, since data for these has already been included). We will investigate associations of common SNPs (minor allele frequency ≥1%) in COL4A1, COL4A2, NOTCH3, HTRA1, TREX1 and CECR1 genomic regions with all ischaemic stroke and ischaemic stroke subtypes. We will perform genetic association meta-analyses using a fixed effects inverse-variance weighted model in METAL meta-analyses software. We have already performed preliminary analyses for ischaemic stroke and subtypes (using the Metastroke dataset), and for ICH and subtypes. We are in the process of starting to analyse the WMH data. Including the NINDS SiGN cohorts would increase our ischaemic stroke sample size substantially. The current stage of the project is limited to analysing populations of European origin, however in the later stages, we are looking to extend it to perform trans-ethnic meta-analyses.12044- Genetic predisposition to depression and risk of stroke - Qibin Qi, SmollerLarge epidemiological studies have demonstrated that depression is associated with increased risk of stroke (1). However, it remains unclear whether genetic variants associated with depression are related to stroke. Information on the association between depression-related genetic variants and stroke might provide some information to better understand the underlying genetic factors relating depression and stroke and might help clarify the causality. Because genetic variants are randomly assigned and generally uncorrelated with environmental factors, the observed genetic associations should be unaffected by confounding factors and also free of reverse causation. Thus, in this proposed manuscript, we aim to examine whether genetic predisposition to depression is associated with increased risk of stroke overall and by subtype and whether there is a genetic correlation between depression and stroke. Given the relatively small variation in depression explained by single genetic variants, we propose to use a collection of multiple genetic variants in a polygenic score (PGS) to represent an overall genetic predisposition to depression. Polygenic score analysis has recently generated much interest for assessing the explanatory power of a collection of markers (2). In this proposed analysis we will examine whether a polygenic score based on the largest available genomewide association analyses of major depressive disorder (MDD) from the Cross-Disorders Study is associated with stroke risk (3). The score weights are publicly available and we willl use those; they can be downloaded. We will first check the distribution of the polygenic scores across the 22 cohorts of SIGN. We will do logistic regressions with the polygenic score treated as a continuous variable, and in separate analyses, treated as quartiles. Both individual level data pooled across the 22 cohorts of SiGN and meta-analyses will be done to assess the relationship of PGS to ischemic stroke and its subtypes. We will also use LD score regression to get an estimate of genetic correlation between depression and stroke to test whether they share common genetic components. Hypotheses to be tested: Hypothesis 1: Higher polygenic scores will be associated with higher risk of overall ischemic stroke . Hypothesis 2: There will be variation in the relationships of predisposition to depression and subtype of stroke. For example it may be that cardioembolic- strokes show more genetic relationship with depression than lacunar strokes. Hypothesis 3: There is a genetic correlation between depression and stroke.12045- The influence of β1-adrenergic receptor polymorphisms on cardiovascular outcomes and antihypertensive treatment in the Secondary Prevention of Small Subcortical Strokes (SPS3) trial The functional Ser49Gly variant (rs1801252) in ADRB1 has previously been associated with various cardiovascular phenotypes; however, few studies have examined its association specifically with stroke and stroke subtypes. In part, as this variant is not typed on any commercial GWAS/SNP chips, as it performs very poorly. A prior study suggested that Gly49 allele carriers may be at a greater risk for ischemic stroke (Kumar et al. 2014; PMID:25510377). Additionally, within SPS3-GENES we have genotyped this SNP using a TaqMan assay and we have observed both a main effect association and a pharmacogenetic association between Gly49 allele carriers and risk for adverse cardiovascular outcomes driven primarily by ischemic stroke.The proposed study aims to replicate the main effect association between Ser49Gly and stroke that was discovered in SPS3-GENES and investigate this association in stroke subtypes. We hypothesize that Gly49 allele carriers will be at an increased risk for stroke, particularly ischemic stroke. Proposed analyses: Analysis in SiGN on rs1801252 and its association with overall stroke and stroke subtypes, stratified on HTN (hypertensive) status.Model: Stroke = rs1801252 + age + gender + race/ancestry + history of HTN + history of diabetes + history of CAD + smoking status(Covariates may be dropped depending on number of missing data)Analysis run in the Overall SiGN database (both overall stroke and stroke subtypes), and then stratified on HTN status (yes/no) (again for overall stroke, and stroke subtypes).Analysis to exclude SPS3 individuals, as that is where our discovery is.As the replication of the pharmacogenomics findings within SPS3-GENES fall outside of the scope of SiGN, we will be following up with SiGN PIs via email to identify any groups who have collected anti-hypertensive drug information and would be interested in participating in a replication effort for the pharmacogenomic signal.12046- Evaluation of Urokinase Locus and Ischemic StrokeIdentification of biomarkers mediating coronary artery disease (CAD) risk such as blood cholesterol and blood pressure has led to tremendous advances in prevention and treatment. While numerous biomarkers have been associated with CAD in epidemiological studies, even the strongest associations can be biased by reverse causation and confounding, limiting their clinical utility. Mendelian randomization (MR) can mitigate these limitations and thus help identify blood mediators of CAD. We therefore have sought to identify blood mediators of CAD and related endpoints through a comprehensive analysis of 237 blood biomarkers using MR. Through our screen, we identified urokinase as a novel mediator of CAD and this association is consistent with the biological role of urokinase as an inhibitor of thrombosis.No individual-level data are requested; only summary associations are needed. We request that all subjects be included in the analyses. No individual-level data is required. We request that the analyses be adjusted for age, sex and ethnicity. No individual-level genotypic data is needed. We request summary level data for associations between rs2227552 and surrounding SNPs (+/- 300 KB) with (1) ischemic stroke sub-type and (2) all stroke, adjusted for age, sex and ethnicity. For each SNP, we request the following columns: MAF, N, beta, SD and effect allele.12047- Phenome-wide analysis of age related genome-wide polygenic scores- Not approvedRecent studies suggest shared genetic effects between cognitive measures and adverse age-related outcomes such as stroke, Alzheimer’s disease or coronary artery disease ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"14t2o6ipg5","properties":{"formattedCitation":"(1)","plainCitation":"(1)"},"citationItems":[{"id":492,"uris":[""],"uri":[""],"itemData":{"id":492,"type":"article-journal","title":"Shared genetic aetiology between cognitive functions and physical and mental health in UK Biobank (N= 112 151) and 24 GWAS consortia.","container-title":"bioRxiv","page":"031120","source":"Google Scholar","author":[{"family":"Hagenaars","given":"Saskia P."},{"family":"Harris","given":"Sarah E."},{"family":"Davies","given":"Gail"},{"family":"Hill","given":"William David"},{"family":"Liewald","given":"David CM"},{"family":"Ritchie","given":"Stuart J."},{"family":"Marioni","given":"Riccardo E."},{"family":"Fawns-Ritchie","given":"Chloe"},{"family":"Cullen","given":"Breda"},{"family":"Malik","given":"Rainer"},{"literal":"others"}],"issued":{"date-parts":[["2015"]]}}}],"schema":""} (1). Nevertheless, it is unclear whether this can also be found for non-cognitive measures such as personality, motivation or behavior problems. By applying a phenome-wide approach, we aim to identify predictors of age related outcomes. We will generate a variety of polygenic scores based on GWAS for age-related phenotypes and relate these to a large number of cognitive as well as non-cognitive variables. We predict that genetic correlations are present for the associations between polygenic scores for age-related traits and cognitive, as well as non-cognitive measures. 12048- A network approach to identifying new candidates for stroke and stroke subtypesIdentification of the genetic contributors to stroke, stroke subtypes, and stroke risk factors is difficult because of the complex ideology of the disease(s). Advanced computational techniques that can assess the impact of many genes together rather than the traditional “one gene one phenotype” approach are lacking in the field of cerebrovascular disease but may be useful to identify regulated in athero tissue (p = 3.8x10-7). In fact, genes with high module connectivity tend to be highly associated with atherosclerosis overall (R2=0.74, p = 1x10-100), and the top modules are also enriched for genes known to be important in cell adhesion, inflammation, as well as known stroke candidates. These data not only lend credence to the validity of our analysis but also indicate that the new genes that have been identified may be good candidates for cerebrovascular disease. Our analyses show a powerful new approach to the identification of new clinically relevant candidates for atherosclerosis, stroke, and stroke subtypes.We hypothesize that the “Hub” (ie most connected) genes in each of these athero associated “modules” are found will be excellent candidate genes for investigation.Our aim is to perform genetic analyses in the SiGN data set for both overall ischemic stroke as well as stroke subtypes using the information gained from our preliminary analyses. We would like to use summary data from SiGN to investigate these loci under the hypothesis that they will be significantly associated with stroke and stroke subtypes when held to a lower p-value correction standard.12049- Identification of variants of COLlateral GENES linked to variation in collateral grade in stroke (COLGENES).Until recently, mechanisms responsible for variation in collateral status were unknown. Work in animals has found that aging and other cardiovascular risk factors cause a decline in collateral number and a smaller lumen diameter in those that remain (collateral rarefaction), findings that have derived support in humans with aging, metabolic syndrome or hyperuricemia. While such rarefaction may account for a small amount of collateral variability, recent studies suggest genetic background may be the primary determinant. Studies in 25 strains of young healthy young adult mice have shown that collateral extent varies by ~50 fold and that infarct volume after permanent MCAO evidences a 30-fold variation that closely follows these strains. Remarkably, a single polymorphic novel gene, Rabep2, has just been identified to be responsible for 80% of this variation. Mice and humans are 92% similar genetically. Moreover, the angiogenic pathway that Rabep2 drives to control formation of collaterals during embryogenesis (ie, collaterogenesis) is likely well-conserved among vertebrates. Thus, we predict it is also likely that polymorphisms at human RABEP2 and/or other “collateral genes” are a major determinant of variation in collaterals in humans (eg, 3 additional loci (QTL) have been identified on different chromosomes that account for all or much of the remaining variation in collaterals in mice, and knockout and transgenic studies have identified involvement of the following known angiogenic genes in the collaterogenesis pathway: Vegfa, Flk1, Dll4, Notch, ADAM10, ADAM12, Gja4, Clic4. Since collateral score is obtained at certain CSCs using CT, MR or angiographic imaging protocol during AIS triage, this collateral phenotype can be combined with genotype and exome sequencing data, together with stroke evolution, functional outcome and clinical and demographic data, to study the genetic basis underlying variation in cerebral collaterals in humans. We are thus in a unique position to test, for the first time, association between collateral status and polymorphisms at RABEP2 and other candidate collaterogenic genes (genome-wide studies employing the Collaborative Cross and RNAseq, soon underway in mice, will to identify the remaining high-priority candidate genes controlling collateral variation in the mouse species). This effort has recently begun in a prospective study of AIS patients, GENEtic Determinants of Collateral Status in Stroke (GENEDCSS). COLGENES will be a companion study to GENEDCSS that will examine the same question using existing genetic, collateral score, outcome and demographics obtained from previous AIS patients available within the MRI-GENIE database and other databases within the SIGN/ISGC. Combining the findings of both studies offers a powerful way to identify polymorphisms linked to the wide variability in collaterals in humans.Validation of RABEP2 and/or a related gene(s) linked to variability in collateral status in acute stroke patients will help to develop a diagnostic biomarker test capable of distinguishing subjects with poor vs. good collaterals when they are healthy before stroke onset. These results could also inform future research aimed at identifying target genes or proteins that prevent collateral loss in individuals without stroke but with cardiovascular risk factor presence. Robust identification of genetic and modifiable determinants of collaterals will pave the way for the development of new therapies aimed at preventively enhancing collaterals before stroke or augmenting collateral circulation in patients with stroke. 12050- Association of genetic variation in RABEP2 and related genes associated with collateral status with risk of incident stroke, stroke subtypes, and outcomeThis will be a look up. We plan to examine the association of RABEP2 variants with risk of stroke, severity of stroke, and degree of disability after stroke in the SiGN consortium. The prospective study GENEDCSS is looking at this question using imaging-confirmed collateral scoring in conjunction with genotyping. The current proposal is aiming to increase the power albeit at the expense of the loss of imaging data. We wish to look for an association in the SiGN GWAS of Stroke. We expect that the SiGN sample will be adequately powered to detect a large association. We will alos conduct parallel investigation in METASTROKE and GISCOME12051 Mendelian randomization study of the role of diabetes, body mass index, glycemia, and insulinemia in the development of ischemic stroke and its subtypesA lookup of association with all ischemic strokes and the main subtypes (large-artery stroke, small-vessel disease, and cardioembolic stroke) for 49 diabetes-associated SNPs, 36 fasting glucose-associated SNPs, 37 insulin-related SNPs, and about 100 BMI-associated SNPs (77 and 32 BMI-associated SNPs from two prior GWAS). The lookup will be performed by Matthew Traylor. Subsequent analysis will be done by Susanna Larsson.The increasing prevalence of type 2 diabetes (T2D) and obesity is a global problem. Adiposity, often measured as body mass index (BMI), and T2D have been associated with increased risk of ischemic stroke in a number of observational studies. Furthermore, observational studies have reported positive associations of fasting glucose and fasting insulin concentrations with risk of stroke. However, T2D, BMI, elevated fasting glucose and insulinemia are highly correlated with each other. BMI also associates with hypertension, the major risk factor for stroke. Hence, the causal relationships of T2D, BMI, glycemia, and insulinemia with risk of stroke remain unclear.The aim of this study is to implement a Mendelian randomization approach to investigate the role of T2D, obesity, glycemia and insulinemia in the pathogenesis of ischemic stroke and its subtypes and to determine which associations are causal. Three previous studies have examined the causality of the association between BMI (using one SNP or genetic scores comprising 14 or 32 SNPs) and risk of all strokes or ischemic stroke, with inconsistent results. None of those studies reported results on ischemic stroke subtypes. To our knowledge, no previous study has assessed the potential causal associations of T2D, glycemia, and insulinemia with ischemic stroke. 12052 Causal role of blood pressure and lipid traits in the development of ischemic stroke: a Mendelian randomization studyCommon cardio-metabolic risk factors as obesity, hypertension, diabetes and dyslipidemia are known to independently predict cardiovascular disease including stroke. However, these factors are highly correlated and some of these associations may not be truly causal. Meta-analyses of blood pressure and LDL cholesterol lowering trials have consistently shown risk reduction of stroke. However, there is insufficient evidence from HDL cholesterol or triglyceride targeted trials that HDL and triglycerides play a causal role in the development of stroke. Genetic markers are randomly distributed at conception and thus can be used to overcome biases as confounding and reverse causation in observational studies.Previous Mendelian randomization studies have shown no causal association between HDL cholesterol and coronary heart disease. However, to our knowledge no such studies were performed with ischemic stroke to date. We aim to conduct a Mendelian randomization study to understand causal role of blood pressure, LDL cholesterol, HDL cholesterol and triglycerides in stroke. First, we will create trait-specific genetic risk scores for systolic blood pressure, LDL cholesterol, HDL cholesterol, and triglycerides from already known variants and use them as instrumental variables to obtain their causal estimates in relation to stroke in the prospective Malm? Diet and Cancer Study (N=30,447). Second, we aim to conduct a multivariable Mendelian randomization using the β coefficients of 97 BMI-, 49 waist hip ratio-, 29 blood pressure-, 185 lipid-, 36 fasting plasma glucose-, 26 fasting plasma insulin-, and 12 adiponectin SNPs obtained from publicly available data releases of the latest GWAS meta-analyses to obtain causal estimates for blood pressure, LDL cholesterol, HDL cholesterol and triglycerides correcting for pleiotropic associations of genetic variants with other cardio-metabolic traits. To conduct such analyses, we also need the β coefficients and standard errors of the same SNPs from the SiGN stroke GWAS meta-analyses for ischemic stroke and ischemic stroke subtypes.12053 CCS-level of certainty analysisHypothesesThe currently identified “hits” for LAA will be Restricting the LAA phenotype to extra- or intra-cranial disease will allow discovery of new sub-subtype specific genetic loci Extra- and intra-cranial LAA will differ from each other in their genetic architecture.Association testingGWAS will be run in the ancestry-specific stratum (using PLINK or SNPTEST)Association testing will test the association of disease status (case/control) and genotype, correcting for the top ten principal components and sex.These covariates are identical to the covariates included in the main SiGN analyses. Meta-analysisStrata GWAS will be QC’dCheck of genomic inflation factor (lambda)Check of lambda by minor allele frequency, by imputation qualityRemoval of SNPs that show excess inflationThis QC is identical to what was done in SiGNConcordance of SNP frequency with publicly-available SNP informationSummary-level data will be combined in an inverse-variance weighted fixed effects meta-analysis (identical to what was done in SiGN) using MANTEL.12054-CCS LAA extra- or intra-cranialHypothesesThe probability of discovering subtype specific genetic loci increases as the level of certainty in attributing a subtype to the cause of stroke increases (evident>probable>possible)Minor CE and major CE differ from each other in their genetic architecture.Minor CE and cryptogenic categories are genetically similarAssociation testingGWAS will be run in the ancestry-specific stratum (using PLINK or SNPTEST)Association testing will test the association of disease status (case/control) and genotype, correcting for the top ten principal components and sex.These covariates are identical to the covariates included in the main SiGN analyses. Meta-analysisStrata GWAS will be QC’dCheck of genomic inflation factor (lambda)Check of lambda by minor allele frequency, by imputation qualityRemoval of SNPs that show excess inflationThis QC is identical to what was done in SiGNConcordance of SNP frequency with publicly-available SNP informationSummary-level data will be combined in an inverse-variance weighted fixed effects meta-analysis (identical to what was done in SiGN) using MANTEL12055 Genetic architecture of basilar artery diameter and basilar dolichoectasiaIntracranial arterial dolichoectasia (IADE), the presence of elongated and dilated intracranial arteries, is found in up to 12% of stroke patients. IADE most commonly affects the basilar artery. Increased basilar artery diameter is associated with recurrent stroke and mortality. The cause of IADE is unknown. Previous studies have reported associations between IADE with older age and the presence of vascular risk factors, especially hypertension, smoking, or a history of stroke or myocardial infarction (MI). IADE is more common in patients with lacunar stroke and white matter disease. However, in 20% no atherosclerotic risk factors are found. In a minority of patients, IADE is seen in conjunction with metabolic and polycystic kidney disorders. Using data from the ISGCNeuroimaging repository we will analyze brain MRI to determine basilar artery diameters andestablish the presence of dolichoectasia. We will perform a GWAS to identify SNPs associated with dolichoectasia and basilar artery diameter.We further hypothesize that SNPS associated with small vessel disease and white matter hyperintensity are also associated with basilar artery diameter and dolichoectasia. We also hypothesize that common variants in genes associated with metabolic disease (Fabry, Pompe, polycystic kidney disease) increase the risk of dolichoectasia. 12056 Common gene variants affecting subcortical volume and stroke riskPending approval12057 Targeted Pathway Analysis of Prothrombic Genes in Early-Onset StrokeTo be revised and resubmitted12058 Atrial fibrillation genetic risk and ischemic stroke subtypesNearly 7 million Americans have had a stroke and 800,000 more will have one this year.Unfortunately, despite the best current treatments about one fourth of individuals that have had a stroke will have a recurrence. Atrial fibrillation (AF) is a leading cause of stroke, and when AF is recognized, strokes can be effectively prevented with anticoagulation. However,AF can be intermittent and asymptomatic, making detection challenging. Therefore, the relative contribution of AF to various stroke subtypes is unclear. Indeed, as many as one-third of stroke patients are said to have “cryptogenic” stroke when a standard clinical evaluation is unremarkable and it is assumed that occult AF causes many of these cryptogenic strokes.In recent years, a widespread heritable component underlying AF has been recognized and we and others have identified genetic susceptibility loci for AF. We have observed that AF genetic risk associates with a substantial AF risk gradient.19 In unpublished data, we have also observed that AF genetic risk associates with ischemic stroke even in individuals without known AF, suggesting that AF genetic risk may be a marker for occult AF. There is therefore a critical need to determine the extent to which AF genetic risk associates with ischemic stroke subtypes. Assessing such relations will both facilitate a better understanding of the underlying causes of stroke and address the extent to which AF genetic risk might be used as a tool to help identify stroke survivors at greatest AF risk.Our long-term goal is to reduce stroke-related morbidity from AF. The objective of this proposal is to utilize AF genetic risk to address the contributions of AF to stroke risk. We hypothesize that AF genetic risk 1) associates with AF in a sample of stroke cases and controls, 2) is enriched in ischemic stroke subtypes as compared to controls, and 3) is most enriched in cardioembolic and cryptogenic stroke subtypes. We seek to leverage the detailed phenotypic data in SiGN to specifically:Aim 1: Validate whether AF genetic risk associates with AF in stroke cases and controls within SiGN, using individuals carrying the diagnosis of AF with large artery atherosclerosis or small artery occlusion stroke subtypes as compared to controls, stratified by definite, probable, and possible CCS phenotypes for each of the major stroke etiopathologic classifications.Aim 3: Quantify the degree to which AF genetic risk is enriched in definite, probable, and possible cardioembolic stroke by CCS as compared with cryptogenic or undetermined stroke classifications, and determine whether AF genetic risk identifies a subpopulation of cryptogenic stroke subjects with a “CE-like” AF-risk pattern.Approach:Aim 1: Validate association between AF genetic risk and AF. In published19 and unpublishedwork we have validated associations between polygenic AF risk and AF. In this aim we will seekto validate whether AF genetic risk associates with AF within SiGN utilizing features1919ascertained within the context of the Causative Classification System for Ischemic Stroke(CCS).Polygenic AF risk: We will first calculate a polygenic AF risk score for each individual. In brief, AFgenetic risk will be comprised of the linear combination of weighted allele dosages for SNPsassociated with AF at P<1x10-5 in a recent 1KG meta-analysis from the AFGen Consortium(unpublished). Nonredundant SNPs will be selected using standard pruning techniques. SNPweights will be comprised of the log relative risk from the AFGen meta-analysis. We anticipateincluding SNPs that are either directly genotyped or imputed with a quality r2≥0.7. We willinclude cohorts with ≥90% of SNPs in the score meeting this quality metric.AF in SiGN: We will define AF among individuals within SiGN by the presence of at least one ofthe following CCS variables: 4.b.i. (left atrial thrombus), 4.b.iii. (AF), 4.b.iv. (paroxysmal AF),4.b.v. (sick sinus syndrome), 4.b.vi. (atrial flutter).We will then examine relations between AF genetic risk and AF in individuals in whom AF hasbeen ascertained as present or absent in SiGN using multivariable logistic regression:AF/control ~ AF genetic risk + covariatesWe will fit models adjusted for: 1) age, sex, and cohort-specific ancestry (e.g., principalcomponents), and 2) an extended set of covariates if available (in addition to model 1covariates, hypertension, diabetes, height, weight, smoking status, heart failure history,myocardial infarction history).Aim 2. Examine whether AF genetic risk is enriched in cardioembolic stroke subtypes. We will examine whether AF genetic risk associates with ischemic stroke and stroke subtypes in casecontrol [analyses using multivariable logistic regression using the same adjustments outlined inAim 1. We specifically will examine the following comparisons:all ischemic/control ~ AF genetic risk + covariatescardioembolic/control ~ AF genetic risk + covariateslarge artery atherosclerosis/control ~ AF genetic risk + covariatessmall artery occlusion/control ~ AF genetic risk + covariatesAim 3: Quantify the degree of AF genetic risk enrichment among the cryptogenic andundetermined stroke subtypes. Since AF genetic risk may be enriched in a subpopulation ofcryptogenic stroke with occult AF, we seek to contrast the relative degree of AF genetic riskenrichment in cryptogenic/undetermined and cardioembolic stroke. We will first compareassociations between AF genetic risk and cardioembolic or cryptogenic/undetermined strokerelative to a referent stroke group using multivariable logistic regression using the sameadjustments outlined in Aim 1. We will define the referent stroke group as small vessel stroke.We will then assess the relations between AF genetic risk and the log-odds of undeterminedrelative to cardioembolic stroke.Crypto/undetermined vs. small vessel ~ AF genetic risk + covariatesCE vs. SV ~ AF genetic risk + covariatesCrypto/undetermined vs. CE ~ AF genetic risk + covariates12059 Assessing genome-wide associations with ischemic stroke using alternative definitions of stroke subtypes: The SIGN StudyStroke subtype classification, applying standard criteria to clinical data, can reduce the heterogeneity of ischemic stroke for genetics studies. To date, most of the replicated genetic loci for ischemic stroke appear subtype-specific. Increasing homogeneity of phenotype comes with a trade-off in sample size, which influences the potential for successful identification of new loci. We estimated genetic associations using the union and intersection of two widely used stroke subtyping systems to assess the influence of sample size and homogeneity on test statistics. 12060 Lesion Location in Large Artery Atherosclerosis: Do genetic risk factors and underlying genetic architecture differ in intracranial and extracranial atherosclerotic stroke? Stroke due to large artery atherosclerosis (LAA) may be sub-classified as extra-cranial, intra-cranial, or occurring in both locations. The NINDS Stroke Genetics Network (SiGN) recently identified TSPAN2 as a third specific locus, in addition to HDAC9 and 9p21, associated with stroke due to LAA. Other loci including 6p21, ABO, and 12q24 may be associated with LAA, but results are inconsistent. We hypothesize that (1) extra- and intra-cranial LAA differ by genetic architecture, (2) previously identified loci for LAA stroke are phenotype specific by intra- or extra-cranial subgroup, and (3) restricting stroke phenotype to extra- or intra-cranial LAA will promote discovery of new genetic associations. We are conducting a meta-analysis using case-control stratum from the discovery phase of the original SiGN Genome Wide Association Studies (GWAS). We derived phenotypic data for the GWAS using the Causative Classification System (CCS) with three levels of certainty. Four phenotypic groups are being examined: CCS evident and probable LAA with any extracranial (n=1482), any intracranial (n=894), only extracranial (n=1244), or only intracranial (n=656) stenosis. The GWAS is being run in the ancestry-specific stratum, which individually underwent quality control, prephasing, and imputation. The ancestries used in the study include ten European, two African, and one Hispanic cohort with a thousand genomes within each stratum. Association testing will analyze the association of each LAA phenotype and genotype, correcting for the top ten principal components and sex, which are identical covariates as those included in the main SiGN analyses. The quality control of the strata GWAS will also be identical to the previous SiGN analyses. Summary-level data will be combined in an inverse-variance weighted fixed effects meta-analysis using MANTEL. The analysis is in progress and results of the association between LAA phenotype and underlying genetic architecture are forthcoming.12061 A sex-specific genome-wide association study of ischemic stroke in the Stroke Genetics Network (SiGN) Stroke is the second leading cause of death and a major cause of severe disability worldwide. Experimental, epidemiological, and clinical studies have consistently documented differences between men and women in stroke incidence, prevalence, outcomes, and risk factors. Although stroke heritability has been reported to differ by sex, the genetic variants that may underlie these sex disparities are unknown.We performed a genome-wide association (GWA) analysis stratified on sex in 12,577 ischemic stroke cases (6811 men, 5766 women) and 25,643 stroke-free controls (12,179 men, 13,464 women ) from the Stroke Genetics Network (SiGN). Testing for heterogeneity of effects between the sexes was conducted using a 1-df Chi-square test.We identified a region on chromosome 19q13.2 harboring common SNPs (minor allele frequency (MAF)=0.11 to 0.25) associated with all ischemic stroke in men (lead SNP P=7.4x10-8; odds ratio (OR)=0.81) but not in women (P=0.38; OR=1.03). A common variant (MAF=0.19) on chromosome 10q22.2 showed evidence of heterogeneity by sex (P_het=7.7x10-8) with sex-specific associations in opposite direction (OR_men=0.86, P=4.4x10-6; OR_women=1.10, P=2.8x10-3). Subtype-specific analyses also identified several sex-specific associations.In this sex-stratified GWA analysis, we identified several loci with sex-specific effects on ischemic stroke and its subtypes. The majority of these loci were in intergenic regions of the genome, suggesting that sex-specific effects on stroke may occur via mechanisms acting on gene regulation. Replications and extension of these findings is planned.12062 Public release of existing genome-wide association study results for the paper “Loci associated with ischaemic stroke and its subtypes (SiGN): a genome-wide association study.” Lancet Neurol 2016; 15: 174–84.12063 Addition of GWAS results for ischemic stroke to the Platform for Accelerated Genetic Discovery in Cerebrovascular Diseases12064 Bayesian Fine-Mapping of MEGASTROKE loci using Multinomial Logistic Regression12065 Assessment of a simple phenotype approach to ischemic stroke subtyping for genetic studiesNearly all replicated ischemic stroke loci are subtype-specific. Stroke subtyping is a labor intensive and expensive to implement in large population studies, particularly within large electronic medical record systems. Existing stroke subtyping systems (CCS and TOAST) were designed for identifying clinical etiology and prognosis. It has never been tested whether a simpler approach based on the presence or absence of a single phenotypic feature would be equally informative for genetic studies. Because of the documentation of the individual items in the CCS within SiGN, it is possible to compare, within the same population, the strength of association for known genetic signals between the CCS and phenotyping based on a single item within the CCS. If a similar strength of association for known genetic signals is obtained with a simpler approach amenable to automated data extraction, then this would increase the feasibility of doing genetic studies of ischemic stroke subtypes in very large electronic medical record systems or very large cohort studies. Further validation studies would still be needed because the high quality information in the CCS is not directly available in electronic medical ecords.12066 COMPASS 1000 Genome Look-up of StrokeStroke is the fifth leading cause of death and number one cause of long term disability. Genetic and environmental factors are key factors for stroke risk. African Americans in particular, have increased stroke incidence, stroke mortality, and suffer strokes at younger ages as compared to their Caucasian peers. Most genome wide association studies (GWAS) of stroke have been conducted in populations of European descent. Using data obtained from the newly-formed Consortium of Minority Population genome-wide Association Studies of Stroke (COMPASS), we conducted the first discovery GWAS meta-analysis of stroke in African-Americans and have subsequently performed a meta-analysis of stroke using genetic data imputed from the 1000 Genomes reference panel. Our group, COMPASS, aims to identify genetic variants associated with stroke (total and ischemic) in African Americans using genome-wide genetic data. We hypothesize that imputation and association analyses using 1000 Genomes imputed data will identify novel locus associated with stroke (and stroke subtypes) in African Americans, as well as replicate prior associations reported in the literature, hopefully providing greater resolution across previously associated loci. Due to the absence of a large African American stroke population to serve as a replication group, we are requesting a look-up of approximately 38 loci identified in our African American meta-analysis using 1000 Genomes imputed data12067Integrative genomics assessment of stroke risk by gender and ethnicity in multiple cohorts.Taking advantage of high statistical power and rich data in both genomics and phenotypes from large studies in diverse populations, we will first investigate the biological pathways and gene networks that are perturbed by genetic variation affecting stroke risk using the NINDS Stroke Genetics Network (SiGN). We will then confirm novel signals using data from the Genetics of Early Onset Stroke (GEOS) Study, the Framingham Heart Study (FHS), the Jackson Heart Study (JHS), and the Women’s Health Initiative (WHI). To further elucidate how the identified pathways mediate their effects, analyses including protective (physical activity) and risk (smoking) factors will be performed stratified by sex and ethnicity. We plan to apply innovative computational methods to the complex genomic data to model gene-networks for stroke risk and determine the specific pathways perturbed by genetic factors in stroke pathogenesis. Finally, we will use innovative causal mediation analysis to quantify each specific pathway and key driver’s contribution to the complex interrelationships among genetics, physical activity, smoking status, and the development of stroke. The following specific aims will be achieved: Aim 1. To identify and replicate biological pathways and gene networks affecting the risk of ischemic stroke separately by African-Americans (AA) and European-Americans (EA) ancestry and by sex. 1.a. We will construct gene regulatory networks (Weighted Gene Coexpression Network Analysis [WGCNA], Bayesian networks and machine learning Graphical Models. 1.b. Using a high-throughput analytical pipeline recently developed by members of our team, data-driven gene-networks, along with knowledge-based biological pathways, will be integrated with GWAS and functional genomics to identify pathways associated with stroke risk in different populations characterized by ethnicity, age-of-onset, sex, and subtypes of stroke.Aim 2. To explore whether and to what extent the stroke-related genetic variations exert their effects on stroke risk through regulating the levels of physical activity and smoking status. Physical activity and smoking are major modifiable factors for stroke. We propose to decompose the effects due to genetic variations identified in Aim1 on stroke into direct effects vs. indirect effects that pass through physical activity and/or smoking.12068 Common gene variants affecting subcortical volume and stroke risk Not approvedA persistent challenge in stroke neurorehabilitation research is the vast heterogeneity of the post-stroke population. Previous studies have suggested inconsistent relationships between post-stroke neuroanatomy and motor recovery based on smaller samples. However, large, diverse datasets can provide the statistical power to robustly form and evaluate these hypotheses. The ENIGMA Stroke Recovery Working Group uses innovative meta-analytic approaches to generate large datasets (goal n>3000) of neuroimaging and behavioral data, collected across multiple study sites (). We ran a preliminary analysis relating post-stroke neuroanatomy with upper limb motor impairment. Based on previous work, we hypothesized that regions of the basal ganglia should most strongly relate to motor impairment. Structural T1-weighted MRIs from over 1400 patients across 14 sites have been committed. Data from 251 stroke patients across 8 research samples were included in the preliminary analysis. ENIGMA protocols were used to extract subcortical measures and perform quality control checks. Regression analyses examined subcortical volume as a predictor of motor impairment; additional covariates included age, sex, time since stroke, hemisphere affected, and total intracranial volume. Focal effects of each lesion on the brain volumes were manually marked and included in the model (volume=0); 10,000 permutations were used to obtain a non-parametric estimate of the statistical significance. In line with our hypotheses, we found a number of significant associations between subcortical volume and motor impairment, particularly in the basal ganglia and lateral ventricles. The strongest result was found in the putamen, with putamen volume positively relating to motor function (p=0.0098, =1065.59). Individually analyzing the three largest sites resulted in only weak and inconsistent results across sites.Interestingly, a recent ENIGMA2 GWAS (Hibar et al., 2015, Nature) examining genetic effects on subcortical volumes also found the strongest effects for the putamen, where a novel intergenic locus with replicable influence on volume (rs945270;?P?= 1.08 × 10?33; 0.52% variance explained) showed evidence of altering the expression of the?KTN1?gene in both brain and blood tissue. Variants influencing putamen volume clustered near developmental genes that regulate apoptosis, axon guidance and vesicle transport.?Given the overlap of the putamen as both the strongest predictor of motor impairment after stroke and the strongest effects of genetic variation in the ENIGMA-2 GWAS, we are now requesting access to the SiGN Stroke GWAS data in order to look into the genetic correlation of the stroke GWASs with our ENIGMA-2 GWAS. We hypothesize that there may be shared genetic variants that both affect putamen volume and are relate to stroke incidence or recovery. Our primary objective is to run a genetic overlap test with LD score – this would require the complete?GWAS?summary statistics (SNP (rsID), alleles (major/minor or effect/non-effect), allele frequency (major/minor or effect/non-effect), effect size/odds ratio (with standard error), p-value, sample size).?If it is not possible to receive the complete summary statistics with all SNPs, we could instead send our ENIGMA 2?GWAS?data for SiGN to run, if they have the resources to do so. If manpower is needed, one of our team members could run the analysis on SiGN’s servers.Alternatively,?if a genetic overlap test is not possible using LD score regression, we could also potentially estimate genetic overlap using polygenic risk scores (PRS) calculated using the top 1% of SNPs in the?GWAS.12069 Genome-wide association study for atrial fibrillation (AF)Aim: Detect novel common genetic variation associated with AFAtrial fibrillation (AF) is the most common heart rhythm disorder, affecting more than 33 million people worldwide. Genome-wide association studies (GWAS) within the AFGen consortium have identified 24 genetic loci associated with AF in published and unpublished work1-4. We aim to analyze a larger number of subjects than in previous GWA studies for atrial fibrillation. This approach will increase power to find common genetic variants associated with AF. The findings could enhance the understanding of underlying biological mechanisms and the genetic basis of this polygenic disease.The primary analyses will use logistic regression, adjusting for age at DNA draw and sex. Each SNP will be modeled using an additive effect. If necessary, the model will be adjusted for cohort/study site, use generalized linear mixed models to account for correlated family data and use principal components for population stratification.We propose to meta-analyze the GWAS results from the SiGN consortium with summary-level results from the AFGen consortium, the Broad AF study and the UK Biobank. Our main analysis would include subjects of all ancestries.Additional analyses: We will perform sub-analyses for each ancestry (eg European, African, Asian) to detect ancestry specific associations. Analysis logistics:We anticipate that all analyses would be conducted by SiGN investigators (either cohort-specific analysts or centrally via a SiGN analyst). Depending on the approach, the results would be meta-analyzed across cohorts (if performed on a cohort-level basis) or pooled with adjustment for each cohort (if performed centrally in SiGN).12070 Do genetic risk scores improve the predictive power of known risk factors for cardiovascular disease? A precision to medicine approachNot approved, entire data set not being released at this time. 12071 DNA to disease: unfolding the human genome to identify disease-causing variation in humans Genome-wide association studies have been hugely successful in identify regions of the genome containing common genetic markers that associate to common disease. However, translating these complex loci – often non-coding, containing hundreds of correlated variants and tens or hundreds of genes – remains a primary challenge in the study of common disease. I will develop a method that uses three-dimensional genome folding to identify sets of disease-associated variants that perturb gene regulation. Within these variant sets, I will implement methods to identify disease-relevant cell types and likely causal variants. My approach will be merged into a (publicly-available) seamless pipeline that can be employed at genome-wide scale in any multifactorial disease of interest. Specifically, I will use Hi-C data to identify SNPs residing in interacting regions of the genome that, together, confer higher risk of disease than either SNP alone. I plan to apply this approach in known stroke loci, so as to investigate potential biological underpinnings of known stroke signals. I will then integrate with other omics data – such as tissue and cell types, and gene expression – to further elucidate potential mechanisms of disease. ................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.
Related searches
- university of maryland jobs baltimore
- university of maryland nursing jobs
- university of maryland hospital jobs
- university of maryland medical system careers
- university of maryland jobs
- university of maryland college park jobs
- university of maryland baltimore jobs
- university of maryland hospital baltimore
- university of maryland baltimore careers
- university of maryland jobs openings
- university of maryland hospital careers
- university of maryland baltimore county jobs