Margaret A. Pericak-Vance, PhD - US grants

Gusella, et al. have localized the HD gene to chromosome 4 (CH4) in several large multi-generational pedigrees using two linked restriction fragment polymorphisms (RFLPs). We propose to define other RFLPs linked to the HD gene as well as to investigate the possibility of genetic heterogeneity in HD by the sampling of local HD families which we have ascertained. We are constructing a sorted CH4 library for the efficient generation of RFLPs on this chromosome. The generation of CH4 RFLPs will also allow us to pursue two additional lines of research: localization of the gene(s) for peripheral neurofibromatosis (NF) and the further localization of the dentinogenesis imperfecta gene (DGI1). In different family studies NF has been shown to be linked to myotonic dystrophy (DM) on chromosome 19 (CH19) and provisionally to the group specific component (GC) on CH4. DGI1 is linked to GC. We have ascertained and will sample local NF families for the study. The DGI1 families will be obtained through Dr. P. Michael Conneally. The NF families will also be screened with our CH19 RFLPs generated as part of the large study on DM. This will allow us to confirm linkage of NF to either CH19 and/or CH4 as well as to investigate the question of genetic heterogeneity in this disease. Once a linked RFLP is found we will continue to screen our libraries in order to generate additional linkage markers. Closely linked RFLPs will be useful in both carrier detection and prenatal diagnosis. Linkage markers on either side of the disorder can also improve the accuracy of genotypic determinations. Finally, we propose to construct a general linkage map of CH4 and CH19 using our RFLPs in conjunction with CH4 and CH19 genotypic markers. We will use our DGI1, NF, and HD families as well as other large multigenerational source pedigrees (i.e. the Venezuelan pedigree, a family with familial hypercholesterolemia (FCH), and our own DM pedigrees) for the general mapping studies. Multilocus linkage analysis will also be imployed. Thus, these experiments will be instrumental in both helping to further establish the chomosomal location of the HD, NF, and DGI1 genes as well as in providing an available linkage map of CH4 and CH19, enabling investigators to link other genetic disorders to this chromosome.

This Program Project involves clinical neurology, genetic linkage analyses and molecular genetic techniques and strategies to study genetic neurological disorders that are at different stages of genetic research. The Project involves six projects and one Core. Project 1 describes diagnostic evaluations, DNA collection, and general linkage analyses for several diseases for which the genetic locus is unknown including non-Duffy-linked Charcot-Marie-Tooth Disease Type 1, vestibular periodic ataxia, autosomal dominant limb girdle muscular dystrophy, oculopharyngeal muscular dystrophy and facio humeral muscular dystrophy. Project 2 proposes continued clinical and linkage studies in tuberous sclerosis, a phakomatous disease for which a tentative locus has been suggested but not proven. Project 3 proposes fine chromosome mapping for two phakomatous diseases, neurofibromatosis and von Hippel-Lindau disease, for which the regional location of the chromosome loci have been established. Project 4 proposes to continue the family development and general linkage analysis for inherited spinal muscular atrophies and other inherited motor neuron diseases, and proposes subtraction hybridization strategies to provide candidate genes. Project 5 describes detailed gene analyses, including characterization of the deletion mutations in Duchenne muscular dystrophy as well as a genetic analysis of the site of mutations, and to study patterns of expression of the DMD gene product in DMD carriers and in a new canine model with deletion of the dystrophin gene. Project 6 proposes to use molecular genetic strategies to define putative antigens in autoimmune myasthenia gravis patients that directly correlate with the pathogenesis and severity of disease, as well as proposing to initiate family genetic studies to determine possible genetic factors that influence the expression of the disease. Core Unit A provides the overall data management and contains the DNA banking facility for the various projects.

Chromosome 19 (CH19) comprises approximately 2.1% of the human genome. Several diseases (e.g. , myotonic muscular dystrophy, malignant hyperthermia) have been localized to this area and the regions around these disorders have been well characterized and mapped. The accumulation of both physical and genetic markers over the entire length of CH19 has been relatively slow. Considerable effort is now being made by several groups to generate additional CH19 markers. As these new markers become available, the ability to construct complete and comprehensive maps of CH19 is possible. The purpose of this workshop is to compile and consolidate the available genetic and physical mapping data from the various laboratories. These individual maps will then be used as a basis to produce a consensus map for the chromosome. Through the workshop efforts, we will be able to identify problem areas that need additional data as well as areas of overlap. In addition, we will identify inconsistencies in the maps. The conference, to be held in Charleston, South Carolina, will be formatted such that each group represented will give a short presentation of their data. All aspects of mapping will be covered including both physical mapping and genetic mapping studies. The individual presentations will be followed by workshop sessions whose emphasis will be to generate a consensus CH19 map and to delineate areas for future concentration. The data will be compiled for the CH19 committee (Drs. Hans-Hilger Ropers and Pericak-Vance) for HGM10.5 and 11 for incorporation into the committee report.

information systems; linkage mapping; biomedical facility; neurogenetics; artificial chromosomes; family genetics; degenerative motor system disease; genetic markers; genetic library; computer assisted sequence analysis; statistical service /center; subtraction hybridization;

Genetic linkage analysis has been an extraordinary tool for identifying disease genes. Over the past ten years, linkage analysis has played a key role in the cloning of several disease. Shortly through the efforts of the Human Genome Initiative, maps of highly polymorphic markers with spacing of 5-10 cM will be available for all chromosomes. These accurate and well defined maps will allow for the efficient screening of linkage of most single gene disorders. In addition, they will provide the foundation for the genetic dissection of diseases with complex patterns of inheritance. The ability to tease out both major gene and moderate effects and their contributions to the etiology of the disorder in question could serve as the first step in combatting the more common human afflictions. The goal of the present proposal is to continue our genetic mapping studies of neurogenetic disorders. Using our past experiences, resources and successes as a guideline we will localize and identify major gene affects in a number of diseases. We will employ a systematic screening approach that combines the use of well-mapped highly polymorphic marker systems with expert clinical and state-of-the art genetic epidemiological tools. We will begin out studies by searching for the chromosomal localization in a number of disorders such as nonchromosome 1 linked CMT2, oculopharangeal muscular dystrophy, and non- chromosome 4 linked facio-scapular humeral muscular dystrophy (FSHD) and unlinked autosomal dominant limb girdle muscular dystrophy (LGM1). We will examine and characterize the extent of heterogeneity in LGM1 and FSHD. Concurrent with the screening of above disorders, we will continue to evaluate and collect DNA on new disorders such as Native American Myopathy and Haw River Syndrome. We will also develop during the course of the grant several complex disease data sets (i.e. neural tube defects and familial narcolepsy). These diseases will be then be incorporated into the genomic screen in subsequent years. Once linkage is established for a particular disorder, we will proceed with our fine mapping objectives in preparation for the application of molecular biological tools to identify the basic defect. Finally, we will investigate both theoretical and applied statistical approaches to the use of linage disequilibrium in mapping diseases. A successful linkage in any of these disorder will represent the first step towards the long range goal of treatment of these affections.

We request support for the 1995 annual single chromosome workshop on human chromosome 9. The purpose of this meeting is to: 1) provide a forum for all major groups mapping chromosome 9 to identify and exchange mapping information; probes (including genes, STSs, cosmids YACs); somatic cell hybrid lines; chromosome 9-specific libraries expressed sequences; mapping strategies and approaches both in use and planned; database resources and software management tools (programs); 2) create a consensus genetic and physical map of the chromosome; 3) stimulate contact and exchange between human chromosome 9 mappers and those mapping syntenic regions of the mouse and other genomes. The first international chromosome 9 workshop was held in Cambridge, England (March 22-24, 1992), was attended by 70 participants from nine countries, and served to bring together all major elements of the chromosome 9 mapping community, providing significant coordination for future mapping efforts. This was followed by 2 additional chromosome workshops, one at Chatham Bars, MA (1993) and this years workshop in Cambridge, England. Published reports have followed each subsequent workshop citing expotential growth in terms of available resources and genetic and physical mapping data. We plan a similar scientific agenda at the ext workshop updated to include new methodologies and data, with subgroups chaired by section leaders to work on several different issues: (in detail of strategies and integration of different types of maps, chromosome 9 resources, and comparative mapping in human and mouse with special reference to human chromosome 9). The chromosome 9 mapping community remains a highly motivated group of an international scope committed to the further development and growth of chromosome 9. The chromosome workshops format have provided the venue to allow for the rapid and efficient exchange and collation of information and data.

Autisic Disorder (AD) is a severe neuro-developmental disorder characterized by marked social and language abnormalities and by sterotypic repetitive behavior. The etiology of AD is unknown, but it is believed to be strongly influenced by genetic factors. The goal of this new project is to identify all genes with major to moderate effectso n the risk of autistic disorder. The investigators propose to achieve this goal through six specific aims. 1) To ascertain and sample 200 multiplex AD families as well as 200- 250 sporadic AD patients and their parents between the two study centers at Duke University and the University of South Carolina. Strict diagnostic criteria will be used based on validated diagnostic instruments. 2) Perform a 10 cM genomic screen using a multi-tiered approach for replication and follow up of interesting regions and state-of-the-art statistical genetic analysis. Both model-based and model -ree methods of anlaysis will be used. Initial screening will be based on 100 families in order to identify regions warranting follow-up. 3) Follow-up regions of interest identified by the initial genomic screen. Criteria for follow-up will include a p-value less than about 0.01 for model free analysis and/or its equivalent of a lod score greater than 1. Follow-up will include genotyping of the second tier of 100 multiplex familes for the original marker and two new flanking markers in the entire data set. 4) Genotype very interesting regions that reach a p-value level <0.0001 and/or lod score of >3.00 with all know polymorphic markers in the region to permit both linkage analysis and family-based association studies to detect any linkage disequilibrium. 5) Examine potential candidate genes. If the screen and follow-up identify one or more small regions, genes in those regions will be examined. Candidate genes or linkages identified or suggested by other researchers also will be followed up. Analyses will include linkage and association/disequilibrium analyses. 6) Characterize Japanese and Finnish families provided by collaborators. These data sets have the advantage of being relatively homogeneous and can be used to confirm the generality of a major effect as well as in fine mapping.

This proposal is a continuation of our grant to delineate the genetics of Alzheimer disease [AD], the most common form of dementia after age 40. Within the past five years, four AD genes have been described. APP, PS-I and II are autosomal dominant causative loci in early (<60) familial AD but represent <2% of all AD cases. The vast majority of cases are late- onset familial or sporadic AD. Through the present grant, the laboratories of Drs. Pericak-Vance and Haines were the first to describe the increased risk related to the APOE-4 allele and the protective effect of the APOE-2 allele. By all estimations, APOE accounts for approximately 50% of the total predicted genetic effect and is the single most important biological risk factor yet identified in AD. Although this finding has dramatically changed the focus of AD research, another 50% of genetic etiology of AD remains unexplained. In addition to delineating the APOE effect, we have completed a genome screen using our initial set of extended late-onset pedigrees and have identified several regions deserving more scrutiny. We have now expanded our data set from 52 to 200+ multiplex late-onset AD families (250+ sampled affected sibpairs) and have 500+ sporadic AD patients and 300+ spouse controls. We also have identified the genetically isolated Indiana Amish population who, despite having a lower prevalence of AD and a decreased frequency of APOE-4, maintain familial aggregation. The above resources permit a more detailed and sophisticated genome screening and analyses to identify all major and moderate genetic effects in AD, something not possible even three years ago. We will use newly described sibpair and affected- relative-pair analyses in conjunction with lod score analysis to make greatest use of the screening data. We will also examine gene/gene interactions between the regions we (or others) conform as harboring AD genes. The ultimate goal of our proposal is the identification of all major loci involved in AD, the first step in combating this devastating neurodegenerative disease.

The Family Ascertainment, Linkage Analysis, and Informatics Core provides a comprehensive framework for clinical and statistical resources necessary to identify genes which predispose to human disease. These functions are highly interdependent and critical to the success of linkage studies. Central to both the family ascertainment and statistical components is the PEDIGENE database.. PEDIGENE is a relational database that integrates family history, clinical, and genotypic marker results together with DNA banking and genomics that integrates family history, clinical, and genotypic marker results together with DNA banking and genomics functions from Core B. This flexible, highly secure genetic database system continues to be instrumental in the rapid and accurate assimilation of and access to all types of genetic data. This core will serve as the umbrella for coordinating and performing all linkage studies in projects 1 and 3, from initial linkage through characterization of heterogeneity through fine mapping, in Mendelial diseases. These diseases include Charcot-Marie- Tooth disease type 2, familial spastic paraparesis, the autosomal dominant limb-girdle muscular dystrophies, facioscapulohumeral muscular dystrophy, and the Lumbee myopathy. The Core also provides consulting support for complex trait analysis such as in project 2, including non-parametric linkage analysis (siblink) and TDT. In addition, this core provides seed support for several projects under development including studies of neural tube defects and Chiari type 1 malformation. Ultimately, these projects will be developed to a point to ensure independent funding, thereby maximizing the impact of the availability of these critical core resources.

Description (from applicant's abstract): Parkinson's disease (PD) is a neurodegenerative disorder affecting over 1,000,000 people in the United States. Onset symptoms are variable and generally occur late in life. The mode of inheritance for the vast majority of familial PD cases is unknown. A number of candidate genes have been suggesting including CYP2D, APOE, and interleukin 2 using case/control studies but these loci have not been consistently replicated, emphasizing the problem with case/control design. A very small percentage of PD cases have autosomal dominant inheritance. Recently mutations in the alpha-synuclein gene have been identified as the cause of a very rare form of autosomal dominant early-onset PD but the genes underlying the common late-onset PD remain unknown. The investigators have ascertained and sampled 215 independent discordant sibships with familial PD and are continuing to collect an additional 200 familial PD-discordant sibships for testing of positive results from the first dataset. In addition, they will collect 400 sets of discordant sibpairs with sporadic PD. These families will form the basis for association testing using sib disequilibrium methods, which will circumvent some of the problems with the case/control approach. The investigators will test for association candidate genes identified in the literature, in their own linkage screens, and through the SAGE analysis in Project II of this proposal in order to identify genes underlying the complex etiology of PD. Risk factor information will be available for gene/ environmental investigations through Project III of this proposal once susceptibility genes are identified. In addition, methods will be developed to adjust for uncertainty in status of unaffected siblings in the discordant sibship tests, and to adjust for multiple tests at tightly linked markers. These studies will help define the underlying etiology of this complex disease.

As the US population ages and the number of adults living to age 80 increases, the prevalence of diseases more common in late life (e.g., Alzheimer disease, Osteoarthritis, Parkinson disease) is also expected to increase. Many lines of evidence suggest that these common diseases have a multi-factorial, or complex, etiology comprised of both genetic and environmental factors. In the past decade, molecular and statistical advances in the study of genetic mechanisms have facilitated dramatic gains in the studies of the etiology of these complex diseases. For example, the availability of huge numbers (greater than 11,000) of microsatellite markers and their well-ordered genome maps has made rapid genomic screening for susceptibility loci possible. Concurrent with these laboratory based landmarks has been the development of new and improved statistical tools of genetic epidemiology designed specifically to address the problems in complex disease analysis. The Genetics of Late Life Disease Core provides a link to the Duke Center for Human Genetics and will provide laboratory, statistical, and bio-informatics support for the OAIC projects studying genetic mechanisms in late-life disease. The core will provide genetic epidemiologic consultation, support for ascertainment of subjects, laboratory and technical expertise for DNA banking and establishment of cell lines, and facilitate genetic mapping studies on complex diseases of late life described in the projects. These efforts include, but are not limited to, genome screening for gene localization, fine mapping of linked regions, and mutation screening. The core also provides a comprehensive structure for the analytic and bio-informatics resources necessary to investigate genetic susceptibility for human disease. Central to this mission is the PEDIGENE database. PEDIGENE is a relational database that seamlessly integrates family history, clinical data, DNA banking, and genotypic marker results. This flexible, highly secure genetic database system continues to be instrumental in the rapid and accurate assimilation of and access to all types of genetic data.

genetic screening; autism; cytogenetics; genetic markers; linkage mapping; genetic susceptibility; nucleic acid sequence; genetic polymorphism; DNA methylation; family genetics; neurogenetics; gene mutation; molecular pathology; chromosomes; gene expression; genotype; linkage disequilibriums; tissue /cell culture; molecular cloning; clinical research; nucleic acid probes; human subject; pulsed field gel electrophoresis; fluorescent in situ hybridization;

DESCRIPTION (provided by applicant): Age-related macular degeneration (AMD) is the most common cause of severe vision loss among individuals over age 50 in the U.S. The socioeconomic impact is considerable and will only get worse as the U.S. population ages. Unfortunately, treatment options remain limited because the etiology of this devastating disease is complex and largely unknown. Considerable evidence implicates a combination of genetic and environmental factors in the pathogenesis of AMD. We have demonstrated that underlying genetic variation is critical to the development of AMD and hypothesize interactions with environmental factors may trigger both the development and progression of the disease. Numerous independent genomic screens, including our collaborative effort with the University of Pittsburgh, have consistently implicated chromosomes 1,10, and 16. Thus it is time to begin detailed mapping and /or gene identification studies in these high priority regions of interest. Our recent identification of the T1277C polymorphism in CFH on chromosome 1 as an AMD susceptibility allele requires detailed follow-up and shows the feasibility of this approach in identifying the genes on chromosomes 10 and 16. Even with this exciting CFH result, the number of genes to be interrogated on chromodomes 10 and 16 is large. To further prioritize within this gene pool, we will invoke the genomic convergence, which combines genetic linkage results, allelic/genotypic/haplotypic association results, and gene expression results to prioritize and test candidate genes for their involvement in AMD. Our specific aims are: to extend our current patient and family data set to include a data set of 1000 cases and 1000 controls;to perform a detailed examination of CFH and other genes in the Regulator of Complement Activity complex on chromosome 1;to perform detailed association analyses across the critical regions of chromosomes 10 and 16, similar to our successful approach on chromosome 1;to generate gene expression data from RPE cells from patients and controls focused specifically on all the genes lying within the chromosome 10 and 16 regions;test candidate genes for association with AMD;and to test for gene-gene and gene-environmental interactions incorporating already identified risk factors such as smoking and the apolipoprotein E (APOE) polymorphisms. The knowledge derived from this study will further our understanding of AMD and will be crucial for future studies to develop and test interventions.

DESCRIPTION (provided by applicant): Alzheimer Disease (AD) is the most common form of dementia in the elderly and currently affects more than four million people In the United States. Identifying the risks associated with the amyloid precursor protein, presenilin 1, presenilin 2, and apolipoprotein E genes has opened new avenues of research but because AD is a complex genetic disorder our understanding of the genetic basis of AD is far from complete. Nearly half of the genetic effect has not been explained and these unidentified genes are keys to defining the cause of this devastating disorder. The present application is a comprehensive effort to integrate statistical (linkage and association studies) and molecular (gene expression studies) genomic approaches toward identifying the remaining AD genes. We recently completed the largest AD genomic screen (455 families, 726 sibpairs) and identified 14 potential regions (MLOD or MLS is greater than or equal too 1.00) as locations for novel AD genes. The most interesting novel region is on 9p22 (MLOD = 3.45; MLS 3.30 in the overall and MLOD = 3.97; MLS = 4.42 in the autopsy confirmed subset[CONF]). We also found evidence supporting the recently reported linkage to chromosome 10q22 (MLOD=2.65; MLS=2.12 in the overall and MLOD=2.94; MLS=1.96 in the CONF). Combined with our continued evidence for a locus on chromosome 12, these three regions represent the highest priority locations for additional AD genes. To perform gene discovery in AD we will: (1) Extend our Family Resources to increase power and develop an independent confirmnation dataset; (2) Identify Locational Candidate Genes by confirming and narrowing the linkage regions and using the human genome sequence to enumerate all the genes within each region; (3) Identify Functional Candidate Genes using microarray and SAGE expression analysis on hippocampal tissue; (4) Test Locational and Functional Candidate Genes using association analyses on a large set of discordant sib pairs. Candidate genes identified through both specific aims 2 (location) and 3 (function) will be given highest priority for detailed examiantion. This approach integrates our family resources and our statistical and molecular expertise to identify the remaining genes in AD.

[unreadable] DESCRIPTION: Autism is a neurodevelopmental disorder characterized by impairments in reciprocal social interaction and communication and the presence of restricted and repetitive patterns of interest or behavior. With the improved surveillance and a broadening of the diagnostic criteria, the most recent prevalence study suggests that autism affects as many as 1 in 300 children in the US. Treatments are few and most have little impact on the very significant morbidity. Little is known about the etiology of autism, but it does have a strong genetic component. Despite this significant genetic effect studies over the past decade have clearly shown that the underlying genetics is complex with the likelihood that several genes acting independently as well as interactively significantly raise the risk of autism. With this realization the field of autism genetics is at a critical juncture. To move forward we must embrace new and creative paradigms to successfully dissect the genetic etiology of this disease. During the current funding period we have emphasized both innovative and established genomic approaches to begin teasing apart the complex weave of autism genetics. In our renewal we will expand and build on previous results embracing the paradigm that the wedding of new genomic technology with novel statistical methodology will bring about success. Specifically we propose to 1) Broaden our ascertainment scheme to include the full range of the autism spectrum disorder phenotype, 2) Identify the chromosome 19 autism gene, 3) Investigate a newly defined linkage to chromosome 12 in large extended multigenerational autism families, 4) Extend our studies of the GABA receptor subunits genes, 5) Identify clinically homogeneous subsets of patients and families and use the refined dataset to fine map ASD chromosomal regions and in candidate gene analyses, 6) Test for evidence of new gene/gene interactions to fully explain the spectrum of autism risk. These efforts will be integrated to address an important problem in childhood disease, the genetics of autism spectrum disorders. [unreadable] [unreadable] [unreadable]

The purpose of the Informatics Component is to foster interactions among the components of the Duke Cardiovascular Genomics Resource Center (DCGRC), as well as interactions among the other Programs for Genomic Applications (PGAs), cardiovascular researchers and the general public. The primary tool of sharing information about resources available through the CGRC will be a website with links to each of the components. Through those links we will administer and provide data, data summaries, analytic software, and results data analysis from each of the components to the target audiences, as well as the comprehensive educational modules for each of the components developed by Education Component. As part of our mission to provide accessible, readily usable data, the informatics components will develop and maintain a database server of all data generated by the study. This state-of-the-art database server will be accompanied by component-specific software tools to allow flexible access to raw data as well as integrated data. The informatics component will also perform statistical analyses as needed throughout the project, making, making the results of those analyses available on the website to allow interaction among the components and PGAs in the interpretation of those results.

DESCRIPTION (provided by applicant): Identifying the four genes [amyloid precursor protein, presenilin I and 2 and apolipoprotein El already associated with Alzheimer disease (AD) has opened new avenues of research and greatly enhanced our understanding of this common and complex disease. Still our understanding of the genetic basis of AD is far from complete. Nearly half of the genetic effect has not been explained. Our recently completed Collaborative Alzheimer Project (CAP) genomic screen, in the largest set of AID families to date (455 families, 726 sibpairs), identified a novel gene localization to 9p21 (MLOD = 3.43; MLS = 3.31 in the overall dataset and MLOD = 3.94; MLS = 4.41 in the subset of autopsy confirmed cases). Preliminary follow-up analyses have demonstrated an allelic association with several single nucleotide polymorphisms in this region (P<0.02), further strengthening our evidence for an AD gene on 9p. A similar localization has been confirmed in an independent study of an isolated Arab population. The present application is a focused effort to integrate statistical and molecular genomic approaches to identify the 9p gene. Breathtaking advances in human genetics, including the completion of the draft human genome sequence and the identification and mapping of millions of single-nucleotide polymorphisms (SNPs), give us the necessary tools to accomplish this goal. We will integrate statistical and molecular methods to identify and test candidate genes on 9p and isolate the gene that modulates the risk of AD. This will be done using complementary datasets of multiplex families, discordant sibpairs, and case-control pairs. We will initially genotype five anchor microsatellite markers in all newly obtained families (-250). Second, using a combination of in silico and laboratory discovery, we will identify SNPs within genes and determine the minimal set of SNPs encoding the haplotypes for each candidate gene. Third, we will perform high throughput genotyping on the minimal SNP set for each candidate gene through all three datasets. We will analyze these data using statistical methods appropriate for each dataset to look for association between a candidate gene and AD. Finally, we will carry out detailed mutation and expression analysis to fully characterize the most likely candidate genes and isolate the 9p AD gene.

Autism is a neurodevelopmental abnormality characterized by significant disturbances in social, communicative, and behavioral functioning. Epidemiologic data has implicated a strong genetic component in autism. The International Molecular Genetic Study of Autism Consortium (IMGSAC) first reported significant results for markers in the region 7q31-35, with a peak maximum lod score (MLS) of 3.55, for their genomic autism screen. Since these initial results, multiple groups, ours included, have reported supporting evidence for an autism locus (or loci) on the long arm of human chromosome 7. Additional support in our data comes from a multiplex autism family (Duke 7543) with a cytogenetic abnormality (Inv (7)(q22q31.2)) overlapping the region of potential linkage as well as other individuals with cytogenetic abnormalities on 7q and autism. During the past three years we have mapped and investigated candidate genes that lie in the critical autism region. Despite this convergence of linkage and cytogenetic studies and efforts by numerous groups, definitive identification of the chromosome 7q gene has remained elusive. In order to identify this gene we propose an integrated yet multiprong approach. First, we will fine map, haplotype, and carry out association studies on the 15 MB to either side of the peak linkage signal for autism using a SNP each 100kb (total 300 SNPs) in 391) multiplex autism families (approximately 1600} individuals available at the start of this project). We will do the bulk of this high-density SNP mapping using newly developed primer extension protocols using pooled patient DNA in conjunction with DHPLC. Candidate genes that lie within peak SNP regions (p<= 0.05) will then be evaluated within the entire data set including singleton families. Secondly, we will simultaneously evaluate and carry out association studies and molecular analyses of genes surrounding the breakpoints in family 7543. Lastly, we will thorougl_y evaluate other proposed chromosome 7q candidate genes, beginning with the reelin gene, for which we have recently obtained significant evidence for association in the joint DUMC, AGRE and Boston/VUMC dataset. All candidate genes will be evaluated by haplotype mapping studies, mutational analysis, transcription and expression as well as methylation and cytogenetic abnormalities. Using this systematic and integrated approach will allow us to ultimately identify the chromosome 7 autism gene, representing a fundamental step forward in solving the genetic riddles of autism.

This Core will be responsible for handling the large-scale genotyping. This Core will be responsible for all the high-throughput genotyping of candidate genes and genetic mapping. Mass Genotyping of individual DNA samples for mapping or association studies of candidate genes will primarily use the TaqMan assay in conjunction with the ABI Prism 700HT Sequence Detection System for single nucleotide polymorphisms (SNPs). In those cases (5-10%) where TaqMan does not work satisfactorily the TaqMan genotyping assay will be supplemented with the Oligonucleotide Ligation Assay (OLA). For high-density SNP frequency mapping of pooled patient samples we will use primer extension in conjunction with denaturing high performance liquid chromatography (DHPLC) detection. The PI's and investigators will identify the autism patients for biological sampling, DNA banking, and genotyping. The Projects, in consultation with this Core, will also determine which genes and SNPs are to be genotyped. In the case of new SNPs, genotype and allele frequency data will be databased and reported to the research community. For genes of particular interest, this Core will provide the necessary DNA sequencing. It will also be responsible obtaining serotonin levels when requested by the Projects.

This Core will be responsible for identifying all the participants in the studies of the Center. This core will identify potential participants, contact them for participation, and enroll them in the study. The core will help arrange appointments either in the clinics or in the home. They will also enroll parents. The Core will also perform numerous assessments on each of the participants, and draw blood samples for DNA extraction. There will be substantial coordination between the Duke and Vanderbilt arms of the Core. In addition, this Core will provide the bioinformatics support to database all the relevant clinical, medical history, family history, and genotypic information collected as part of the Center. The PEDIGENE@ system, already in place at both the DUMC and VUMC sites, will be used as the primary database system. PEDIGENE@ is already being used for numerous other projects and has been field tested for over 10 years.

DESCRIPTION (provided by applicant): In the past, the field of Alzheimer disease (AD) genetics has benefited from the development of innovative paradigms that incorporate the latest genomic technologies combined with pristine patient data to dissect its complex etiology. Our group has successfully used this paradigm in both the identification of the APOE risk effect and more recently the glutathione S-transferase Omega-1 (GSTO1) age at onset (AAO) effect in AD. There is a new appreciation of the power of incorporating clinical phenotypes and developing clinical subphenotypes in attacking complex disorders. In addition, molecular genetic methods have continued to advance rapidly. Whole genome association (WGA) is a new approach that allows the direct evaluation of 300,000- 1,000,000 SNPs from across the genome for association with AD and provides the opportunity to perform a much more detailed examination of the genome than linkage studies. However, while the information content of WGA is extraordinarily high, the initial false positive rate using standard analyses is also high. Investigating each of the thousands of markers that will reach nominal significance is an ominous and inefficient task. One solution to this problem is the genomic convergence approach, which integrates disparate data types to sift through the volumes of existing data to prioritize the best candidate genes for intensive analysis. We have already demonstrated the utility of this approach with the identification of the GSTO1 gene. Thus we are proposing a WGA study of AD and will filter the results using existing linkage, candidate gene, and our recently generated microarray and Serial Analysis of Gene Expression (SAGE) data. A small set of candidate genes identified in multiple of these studies will be the focus of intensive follow-up analysis. Of particular importance will be our ability to follow-up using detailed clinical data on movement and psychiatric symptoms in a newly collected case-control dataset. Our unique position will enable us to marry the most powerful of new genomic approaches, WGA, to existing information to elucidate additional genetic effects contributing to this important neurodegenerative disease. The knowledge derived from this study will further our understanding of AD and will be crucial for future studies to develop and evaluate interventions. The knowledge derived from this study will further our understanding of the genetic etiology of AD. This understanding will be crucial for future studies to develop early interventions and more focused treatments, which will help alleviate the suffering of those with the disease and their families.

DESCRIPTION (provided by applicant): Age-related macular degeneration (AMD) is a significant health problem that affects millions of individuals and is the most common cause of severe vision loss among individuals over age 50 in the U.S. (Rein et al. 2006). The influence of genetic variation on AMD is strong and through the application of recent technological advances the genetic etiology of risk for AMD is being deconstructed. Independent studies, including our own, have identified and confirmed variations in multiple genes that affect risk to AMD, including CFH, HTRA1/ARMS2, C2/CFB, and C3 (DeAngelis et al. 2007; Edwards et al. 2005; Haines et al. 2005; Jakobsdottir et al. 2005; Klein et al. 2005; Maller et al. 2006; Rivera et al. 2005; Schaumberg et al. 2007; Schmidt et al. 2006; Hageman et al. 2005; Yates et al. 2007; Maller et al. 2007). Variation in these genes explain a significant portion of the genetic risk for AMD and ongoing studies are continuing to identify additional such genes. Also important are environmental risk factors such as smoking, hormone therapy and diet that contribute to AMD risk both independently and through their interactions with genes (Schmidt et al. 2006; Schaumberg et al. 2007; Wang et al. 2009). Again, ongoing studies are teasing apart these contributions. However, risk is just one of the many facets of the overall genetic architecture of AMD. Disease progression and treatment response are two critical elements also influenced by genetic variation (Shuler, Jr. et al. 2007; Seddon et al. 2007; Francis et al. 2009). The goal of this proposal is to increase our understanding of the genetic etiology of progression and treatment response in AMD, both of which have been understudied. Identifying the genes underlying clinical outcomes is directly relevant to better directing current treatments and developing new and better treatments and regimens for those suffering this disabling disorder.

ABSTRACT Age-related macular degeneration (AMD) is the most common cause of severe vision loss among individuals over age 50 in the U.S. with millions of individuals around the world suffering severe vision loss. The influence of genetic variation on AMD is strong and through recent technological advances the genetic etiology of risk for AMD is being deconstructed. Independent studies have identified and confirmed variations in multiple genes that strongly affect risk to AMD, including CFH, HTRA1/ARMS2, C2/CFB, and C3 explaining a significant portion of the genetic risk for AMD. Initial efforts at genome-wide association studies have identified and/or confirmed several additional loci of more modest individual effect (CFI, LIPC, TIMP3), with many more loci providing suggestive associations. However, a substantial portion of the genetic architecture remains unexplained and detailed examination of effects specific to subtypes of AMD have been lacking. To address these deficiencies very large sample sizes of well characterized cases and controls and families are needed. Over the past year we have formed the AMDgene consortium to combine both samples and expertise. The initial goal of the consortium was a meta-analysis of existing GWAS data in a combined dataset of over 9,000 cases and 49,000 controls. Preliminary findings have identified new genome-wide significant loci. We have chosen an approach that maintains the primary data at each site, which promotes continued engagement by all participating sites, is cost and time efficient, and avoids potential consent, ethics, and privacy issues of sharing data collected under a wide variety of informed consent. The primary goal of this proposal is to support the AMDgene consortium effort through the following specific aims (1) Coordinate the activities of the AMDgene Consortium; (2) Add new datasets and augment current datasets; (3) Perform detailed meta-analyses on existing and new datasets:; and (4) Perform detailed secondary analyses on these data.

DESCRIPTION (provided by applicant): Through genetic analysis of case-control cohorts there has been remarkable progress in identifying genes for Age-Related Macular Degeneration. Most of these identifications have been accomplished through the interrogation of common variants. We will initiate the identification of rare variants in a founder population, the Amish, and use the whole exome chip to assess the association of Age-related Macular Degeneration with coding variants. We will further refine the Age-Related Macular Degeneration phenotype through the use of modern imaging, the OCT, to visualize the early anatomic signs of Age-Related Macular Degeneration. We hypothesize there are endophenotypes associated with specific genotypes that can be used to determine Age-related Macular Degeneration progression. These endophenotypes are hypothesized to be influenced by a combination of common and rare variants. Following the completion of these Aims, we will have phenotypically defined the early signs of Age-Related Macular Degeneration aiding our understanding of this disease. Moreover, we will relate these signs to genotypes to define the role of genetics in the progression of Age-Related Macular Degeneration and a new risk profile incorporating genotypic information.

DESCRIPTION (provided by applicant): Health disparities marginalize many racial and ethnic minorities within the American health care system. Despite steady improvement in the overall health of the United States, individuals within these underserved groups continue to be more vulnerable to lapses in care and are at increased risk for health problems. Health disparities have had an especially profound effect on the overall health of Hispanics/Latinos and Blacks in the United States. Genetic advances hold extreme promise, but also the potential to further increase health disparities. To promote equitable dissemination of the benefits of genomic medicine, genomic and translational research must be performed in diverse populations. Unfortunately, minority populations are underrepresented in most research, including genetic research. The negative effects of this underrepresentation can already be seen in some of the current applications of genomic medicine. In addition, the paradigm shift in healthcare may also fuel health disparities. The Conference to Remedy Health Disparities aims to learn, identify, and discuss many of the challenges and possible solutions for addressing health disparities in genomic medicine. The conference will place greater emphasis on issues specific to the Latino/Hispanic and Black communities will bring together a wider range of stakeholders who would not otherwise have the opportunity to interact and will include data that has emerged since 2008, specifically with regard to the unique issues raised by sequencing. We also plan to tie in the results of our discussion groups with extant initiatives and policies set forth by NIH health disparity agencies and publish the results of the forum to make it widely available to researchers and health professionals unable to attend the conference.

DESCRIPTION (provided by applicant): This proposal, entitled the Consortium for Alzheimer's Sequence Analysis (CASA) describes plans to analyze whole exome and whole genome sequence data generated from subjects with Alzheimer's disease (AD) and elderly normal controls. These data will be generated by the National Human Genome Institute Large-Scale Sequence Program. The goal of the planned analyses is to identify genes that have alleles that protect against or increase susceptibility to AD. This is a multiple PI proposal, a collaboration between five senior AD genetics investigators (Farrer, Haines, Mayeux, Pericak-Vance, Schellenberg). CASA has 4 cores and 3 projects. Core A is the Administrative core that will coordinate all aspects of CASA. Core B is the Analysis Statistics and Innovation core that will design and assist analysis by other cores/projects and devise novel methods of statistical analysis. Core C is that Data Management and Information Transfer Core that will implement analyses designed by Core B and the projects. Core C will provide high-performance computing for CASA. These three cores are mandated by FOA PAR-12-183. Core D is the In Silico Functional Genomics Core that will annotate AD-associated variants and perform pathway and interaction analyses. Project 1 will evaluate variants detected in the sequence data for association with AD to identify protective and susceptibility alleles. Project 2 will evaluate sequence data from multiplex AD families to identify variants associated with AD risk and protection, and evaluate variant co-segregation with AD. Project 3 will focus on structural variants (insertion-deletions, copy number variants, and chromosomal rearrangements). The project will use existing methods and develop and implement novel approaches for detecting structural variants. The projects and cores are highly interdependent. For example, structural variants identified by Project 3 will be integrated with single nucleotide AD-associated variants identified by Projects 1 and 2. Likewise variants identified by Project 1 will be tested in the family-based data sets. Core B will assist all projects in designing analyses and Core C will implement Project analyses. Core D will annotate and help interpret results from all projects.

DESCRIPTION (provided by applicant): Despite steady improvement in the overall health of the United States, individuals within underserved groups continue to be more vulnerable to lapses in care and are at increased risk for health problems. Health disparities have had an especially profound effect on the overall health of Hispanics/Latinos and Blacks in the United States. Genetic advances hold extreme promise, but also the potential to further increase health disparities. To promote equitable dissemination of the benefits of genomic medicine, genomic and translational research must be performed in diverse populations. Unfortunately, minority populations are underrepresented in most research, including genetic research. Upwards of 96% of participants in genetic studies are of European ancestry. The negative effects of this underrepresentation can already be seen in some of the current applications of genomic medicine. In addition, changes in healthcare infrastructure resulting from genetically tailored care may also fuel health disparities. The Conference to Remedy Health Disparities aims to learn, identify, and discuss many of the challenges and possible solutions for addressing health disparities in genomic medicine, particularly with regard to policy (the theme of this year's conference). The conference will place greater emphasis on issues specific to the Latino/Hispanic and Black communities; will bring together a wider range of stakeholders who would not otherwise have the opportunity to interact; and will include recent data, specifically with regard to the unique issues raised by new technologies such as sequencing and epigenetics. We also plan to disseminate the content of the conference on the web to make it widely available to researchers and health professionals unable to attend the conference. Also, we will continue to record interviews of a selection of attendees and speakers about this topic, which we also plan to also make available for large scale dissemination.

PROJECT SUMMARY To identify new treatment targets, we and others have examined the genomics of Alzheimer's disease (AD). However, genomic successes so far have arisen from studying primarily non-Hispanic White (NHW) participants, and the study of minority populations has been minimal. What few studies have been done in minority populations have suggested that the genetic architectures overlap, but only partially. Thus, studying minority populations not only serves to test generalization of the NHW findings but also provides a unique opportunity for discovery of novel targets and pathways. To begin addressing these issues, we propose here the Puerto Rico Alzheimer Disease and Related Disorders Initiative (PRADI). We will whole-genome sequencing (WGS) Caribbean Hispanic Puerto Rican (CHPR) AD multiplex families to identify novel AD variation in CHPRs, and to generalize existing AD genetic discoveries to this underrepresented population. This initiative will increase our knowledge about genetic variation, particularly for the Caribbean Hispanic population of Puerto Rico (CHPR). The Puerto Rican (PR) population is the 2nd largest Hispanic/Latino population in the continental US. The prevalence of AD in the Caribbean Hispanic population of the island of PR is estimated in 65,000. The PR population is a highly mixed population with average ancestry values of ~64% European, ~21% African, and ~15% Native American. The unique genetic make-up of the PR AD population will be critical in new discovery as well in replication of findings from the Alzheimer Disease Sequencing Consortium (ADSP) CHDR data and the Alzheimer's Disease Genetics Consortium (ADGC) African American (AA) data. Thus, discovery of genetic contributions to AD risk and protective variants in CHPR would have a substantial influence on our understanding of AD and towards our goal of identifying new treatment targets. Through this proposal in response to PAR-15-356 we will address this important issue by conducting genomic studies of AD in PR. Specifically we propose a family-based study in PR that parallels the family-based efforts in the ADSP Discovery phase and that will enhance and extend both current ADSP and ADGC efforts to a broader AD community. We aim to 1) Characterize the genetic epidemiology of AD in PR 2.) Generalize and refine known risk and protective loci in familial PR AD. 3.) Perform variant discovery in our PR AD families and case control data 4.) Leverage multi-ethnic populations (PR, DR and AA) to discover novel AD risk/protective effects by calculation of local ancestry, admixture mapping and bioinformatics analysis and 4.) Perform multi-locus analyses providing insight into functional implications of the risk and protective loci. Our overall goal is to identify targets for therapeutic development that will either prevent or significantly delay the onset of AD.

? DESCRIPTION (provided by applicant): Alzheimer disease (AD) is the leading cause of dementia in the elderly and occurs in all ethnic and racial groups. A multitude of genetic studies in AD have identified multiple AD associated genes and loci, but a large portion of the genetic influence on AD remain unknown. The Alzheimer Disease Sequencing Project (ADSP) will use large-scale sequencing efforts to increase our knowledge about the genetic variation that influences AD, particularly rare genetic variants that enhance AD risk or protect against AD. Substantially underrepresented in these efforts, however, is the generalization of current and future findings in African Americans (AA). AA have a higher prevalence of dementia than non-Hispanic Whites (NHW). Despite steady improvement in the overall health of the U.S. population, individuals within these underserved groups continue to be vulnerable to lapses in care and are at increased risk for health problems. Health disparities have had an especially profound negative effect on the overall health of AA. AA have disproportionally higher health-risk factors, limited access to health services, and ultimately poorer health outcomes and life expectancies than NHW. The determinants of the health disparities seen in AA are many, including public health policy, clinical practices, and social, economic, cultural and/or language factors. One promising avenue for reducing health disparities is the use of precision medicine to improve disease prediction, prevention, diagnosis, and treatment. However, genomic medicine relies on participation from diverse populations in both research and clinical applications. Through this application, we will address this important issue by conducting genomic studies of AD in AA. In response to RFA AG-16-002, we are proposing a set of experiments that will complement and extend the activities and results of the ADSP Discovery and Replication phases, thereby addressing the issue of health disparities in AD research. Specifically we propose a family-based study in AA that parallels the family-based efforts in the ADSP Discovery phase and that will enhance and extend current ADSP efforts to a broader AD community. Specifically, we will 1) Expand our existing AA family dataset; 2.) Generalize and refine ADSP risk and protective loci in familial AA AD. 3.) Prioritize variants by admixture mapping and bioinformatics analysis and 4.) Perform multi-locus analyses providing insight into functional implications of the known risk and protective loci and identifying possible additional genic targets. Our overall goal is to identify targets for therapeutic development that will either prevent or significantly delay the onset of AD.

Age-related macular degeneration (AMD) is one of the leading causes of blindness in the elderly. It has a significant impact on the independence, quality of life, and healthcare costs for those afflicted and the additional social cost on caregivers and family members is incalculable. There is substantial variability in the AMD phenotype and the primary treatment, to repeatedly inject anti-VEGF antibodies into the eye of those severely affected, is effective in only a subset of individuals. Thus a better understanding of the underlying causes of AMD is needed to help guide development of more universal and effective treatments and potential preventive measures for AMD. AMD is strongly influenced by genomic variation. Through the initial funding period of this proposal, we supported the development of the International AMD Genomics Consortium (IAMDGC), which brought together 26 research groups from around the world. The IAMDGC has increased the known genomic loci from 12 to 52 and successfully performed a new and larger genotyping study focused on rarer variation using a high-density genome-wide SNP chip with exome content. This collaborative effort has also spawned numerous additional interesting avenues of research that we now need to explore in more detail through the renewal of this highly successful project. While the current genotypic dataset of over 50,000 samples has much left to be mined, expansion of the available samples, with a particular focus on families and minority samples, is necessary if we are to achieve our stated goal of completely defining the genetic architecture of AMD. To address these unresolved issues we propose four specific aims: 1) Expand the IAMDGC resource with additional datasets and expansion of current datasets, with a focus on family data and diverse genetic ancestry; 2) Expand the range of clinical diagnostic measures (e.g. fundus photos, OCT measures), biomarker, comorbidity, and covariate data associated with the samples; 3) Use an analytical hub infrastructure to perform detailed analyses of these data; and 4) Support the logistics and administration of the IAMDGC.

ABSTRACT: Alzheimer disease (AD) is the most common form of dementia in older individuals. Both genetic and environmental factors contribute to AD risk, yet despite huge research efforts, a significant portion of the genetic etiology of AD remains unexplained. Population-wide studies of unrelated AD cases and controls have identified several common genetic risk factors and through the Alzheimer?s Disease Sequencing Project (ADSP) whole exome (WES) and whole genome sequencing (WGS) data are being analyzed to identify AD risk modulators. However, the primary focus of almost all of these studies has been on identifying variants that increase risk; studies designed to identify variants that may protect from AD are few and usually underpowered. Thus, additional strategies are required to identify functional variants that protect against or delay the development of AD. The Amish provide a powerful and unique opportunity to identify variants protecting against AD whilst controlling for some confounding factors such as level of education, lifestyle and diet. In addition, the large Amish pedigrees offer an enrichment strategy for identifying rare variants since Mendelian transmissions from parents to offspring, coupled to inbreeding loops, maximize the chance that multiple copies of rare variants exist. The primary goal of this project is to identify genetic variations offering protection against AD. The project will achieve this goal by pursuing three specific aims: (1): Generation of a family-based Amish AD Protective Variant dataset. We will collect DNA and phenotype data from Amish families in Ohio and Indiana by examining and following 800 known and newly identified cognitively normal individuals age 80+ and their 1st and 2nd degree relatives. We will perform SNP genotyping on all samples and WGS on a subset of 200 cognitively normal individuals; (2): Identification of AD protective variants. Sibships with multiple individuals who are age 80+ and cognitively normal will be analyzed for genetic linkage, IBD segment sharing, and association. Single-marker analyses will be supplemented by gene-wise analysis and pathway (gene set-based) analysis; (3): Perform functional validation of candidate protective variants. These experiments will include screening for effects on gene expression, impact on A? or tau processing, and effects on cellular function.

PROJECT SUMMARY The Alzheimer?s Disease Sequencing Project (ADSP) is a national sequencing initiative focused on identifying genetic risk and protective factors for Alzheimer?s Disease (AD) in an effort to identify new pathways for prevention and new targets for drug development. The projects? discovery phase included whole exome sequencing (WES) of 10,914 unrelated cases (N=5,778) and controls (N=5,136) and whole genome sequencing (WGS) of 1,019 familial samples. A majority of these samples are non-Hispanic white (NHW) in origin, making the addition of ethnically diverse samples to the study critical to identification of both shared and novel genetic risk factors for AD between populations. This ethnic diversity was emphasized in the ?ADSP Follow-Up Study (FUS) Phase? planning stage with a directive that additional existing cohorts with unrelated AD cases that ?encompass the richest possible ethnic diversity? be given the highest priority for inclusion. To fulfill the goals of this RFA and this FUS Phase Mandate, this proposal identifies seven existing elderly cohorts of African-American (AA) and pan-HI ancestry with a total of 10,430 samples (N=2,322 AA AD cases and 1,843 AA controls and 2,928 Hispanic AD cases and 2,875 Hispanic controls) for WGS and processing in collaboration with existing NIH-funded AD infrastructure. Combining these cohorts with existing African America (AA) and Hispanic (HI) sequencing from the Washington Heights-Hamilton Heights-Inwood Community Aging Project (WHICAP), the Alzheimer?s Disease Genetics Consortium (ADGC) and the ADSP will provide large ethnically diverse datasets for both validation of ADSP discovery phase findings and discovery of novel risk and/or protective variants for AD. Importantly, these data will allow for admixture mapping, a powerful method of gene mapping for diseases that show differential risk by ancestry, by comparing allele frequency differences between populations. They will also become an invaluable resource for the AD research community at-large, and will help to address the health disparities that contribute to AA and HI populations having higher rates of AD than NHW. Thus, we will address these important issues by creating a large dataset of AA and pan-HI AD cases and controls for study. Specifically we propose to: 1) increase the ethnic diversity of the ADSP by assembling samples from existing cohorts with AA and HI AD cases and controls; 2) collaborate with the National Cell Repository for Alzheimer?s Disease (NCRAD) in assemblage, storage, and distribution of DNA on these cohorts; 3) generating genome-wide SNP array data and WGS for all collected samples; and 4) collaborate with the NIA Genetics of Alzheimer?s Disease Data Storage Site (NIAGADS) and The Genome Center for Alzheimer?s Disease (GCAD) in processing, quality controls, storage and distribution of the final datasets. Our overall goal is to enhance the discovery of AD risk factors by facilitating research on AD in ethnically diverse datasets.

The Alzheimer's Disease Sequencing Project (ADSP) is a national sequencing initiative focused on identifying genetic variants that either protect from or increase the risk for late onset Alzheimer Disease (LOAD), with the goal of accelerating development of effective therapeutics. The ADSP Discovery phase includes whole exome sequencing (WES) of 10,571 unrelated non-Hispanic White (NHW) cases (N=5,606) and controls (N=4,965), and whole genome sequencing (WGS) of 578 NHW and Hispanic (HI) familial samples. The Discovery Extension Phase of the project added WGS on 434 new familial samples and 3,343 NHW, AA, and HI cases and controls (collectively the ADSP-DEP). Preliminary analyses of these data confirm known LOAD genes and point toward several new LOAD-related genes and their functional relationships to APP processing, neuroinflammation, endocytosis, and cholesterol metabolism. The current Follow-Up Study phase (ADSP-FUS) will generate >15,000 additional WGS focused on samples that `encompass the richest possible ethnic diversity', which will include primarily HI and African-American (AA) datasets. This combination of diverse datasets derived from case-control, cohort, and family study designs requires an intensive and comprehensive analytical effort to uncover the wealth of information sequestered in the WGS. We (the Collaboration on Alzheimer Disease REsearch [CADRE]) hypothesize that protective and risk genomic variants will provide potential therapeutic targets for Alzheimer disease. Thus, the primary goal of this proposal is to integrate comprehensive genomic analysis of the combined ADSP data (WGS, WES, SNP array) with extant biological data to identify the highest priority variants and loci as candidates for downstream functional analysis. By leveraging the data derived from the AA and HI admixed populations, we use their increased diversity to accelerate and define likely targets. This goal will be met by: 1) Characterizing genomic variation in LOAD using ethnically diverse datasets. We will supplement the ADSP-DEP and ADSP-FUS with additional, separately funded WES and WGS data; 2) enhancing discovery and fine-mapping using admixture analyses in ethnically diverse datasets; and 3) Prioritizing variants and genes by integrating statistical and biological information. For the variants we identify, we will generate additional information from structural and gene expression data and integrate all data into a genomically driven comprehensive biological network that will be used to prioritize loci for functional testing as therapeutic targets.

PROJECT SUMMARY The Alzheimer's Disease Sequencing Project (ADSP) is a national sequencing initiative focused on identifying genetic risk and protective factors for AD. The projects' discovery phase includes whole exome sequencing (WES) of 10,061 unrelated non-White Hispanic (NHW) cases (N=5,096) and controls (N=4,965), and whole genome sequencing (WGS) of 584 NHW and Hispanic familial samples. The 'Discovery Extension Phase' of the project added WGS on ~430 additional familial samples. An initial 'Follow-Up Study (FUS)' Phase focused on examining candidate variants from the discovery phase, and identification of novel variants through combined analysis of diverse datasets is ongoing and focuses on additional existing cohorts with unrelated AD cases that 'encompass the richest possible ethnic diversity' as well as highly valuable set of autopsy confirmed cases and controls. In total we have already included in FUS over 14000 samples for sequencing including >2800 autopsy confirmed cases and controls, >6800 Hispanic (HI) cases and controls and > 5790 African American (AA) cases and controls. In this FUS2 application, we are proposing sequencing of an additional 4243 samples that will both increase our power to find effects but will also enhance our analysis by inclusion of several rich and unique sample sets. Additionally, these datasets will become an invaluable resource for the AD research community at-large. In summary ADSP will provide large ethnically diverse as well as unique datasets for both validation and generalization of ADSP discovery phase findings and discovery of novel risk and protective variants/genes for late onset AD (LOAD). Additionally, these datasets will become an invaluable resource for the AD research community at-large. Specifically we propose to: 1. Increase the diversity and further enrich the clinical phenotype data of the ADSP including data sets targeted to find protective effects through assembling samples from existing cohorts from unique populations that fulfill this goal. 2. Collaborate with the National Cell Repository for AD (NCRAD) in assembling DNA, blood and brain tissue on these existing cohorts, which will serve as a central resource for the AD research community. 3. Generate genome-wide SNP array data through the John P. Hussman Institute (HIHG) Center for Genome Technology (CGT) and whole genome sequencing data through the Uniformed Services University of the Health Sciences (USUHS) for all collected samples 4. Collaborate with NIA Genetics of AD Data Storage Site (NIAGADS), the Genome Center for AD (GCAD) and the University of Pennsylvania and the HIHG Center for Genetic Epidemiology and Statistical Genetics Quality Control Teams in processing, storage and dissemination of final data sets. Our overall goal is to enhance the discovery of AD risk and protective factors by facilitating research on AD in ethnically diverse and phenotypically rich datasets.

ABSTRACT Alzheimer disease (AD) is the leading cause of dementia in older adults and occurs in all ethnic and racial groups. A multitude of studies have identified multiple AD associated genes and loci, but a large portion of the genetic contribution to AD remain unknown. The Alzheimer Disease Sequencing Project (ADSP) is using large-scale sequencing efforts to increase our knowledge about the genetic variation that influences AD, particularly rare genetic variants that enhance AD risk or protect against AD. A major initiative within the ADSP is the Follow Up Study (ADSP-FUS), an expansion of gene discovery efforts to include diverse and unique population. Over the past 12 months, the scope of the ADSP-FUS has expanded to include a number of new cohorts for sequencing. The inclusion of these new cohorts, which have a broad span of clinical-phenotype information, significantly enriches the value of the ADSP in achieving its goal of increasing knowledge of genetic variation in AD across different populations. This supplement is designed to complete pre-statistical harmonization activities in six new cohorts that will be included in the larger harmonization of clinical-phenotype data in all of the ADSP datasets. The final ADSP harmonized dataset, which includes all cohorts (and future cohorts) will create an invaluable, much needed, legacy resource for the NIA. We will perform comprehensive pre-harmonization activities for the six new FUS cohorts (Age, Gene/Environment Susceptibility (AGES) Study; Longitudinal Study of Aging in India-Diagnostic Assessment of Dementia (LASI-DAD); Gwangju Alzheimer and Related Dementias (GARD) Study; Long Life Family Study (LLFS); Aspirin in Reducing Events in the Elderly (ASPREE) Trial Cohort; Iberian Peninsula Cohort) that are not currently funded through other sources. Given that these cohorts vary in their focus (i.e., not all are dementia cohorts) there is a wide span of clinical-phenotype information which makes harmonization challenging. Creating a pre-statistical harmonization workflow in which these data are prepared and used by the ADSP-Harmonization Consortium will expedite the delivery of harmonized clinical-phenotype data to the larger AD community. To complete the pre-statistical harmonization efforts, we will: (a) Identify, collect and organize clinical-phenotype data from the new cohorts, (b) Review and document procedures for data collection for all relevant clinical-phenotype variables, and (c) Perform preliminary quality control analyses for cohort specific data. The successful completion of the proposed pre-statistical harmonization will yield harmonization ready data for the six cohorts that will be utilized in the statistical harmonization of all ADSP datasets. Just as important, this supplement will establish an infrastructure for implementing pre-statistical harmonization that will be integrated into the ADSP-Harmonization Consortium. In the long run, by enhancing harmonization of the ADSP cohorts we are helping make available valuable sequence data to the AD research community to speed gene discovery and identify targets for AD therapies.

Age-related macular degeneration (AMD) is a leading cause of vision loss in older Americans and severely impacts the independence, quality of life, and healthcare costs for those afflicted and their families. Genetic variation has a major influence on AMD, but only about half of the heritability is currently understood. Understanding the genetic architecture of AMD is critical for developing better treatments for AMD. The International AMD Genomics Consortium (IAMDGC) has assembled 33 research groups and over the past four years of this grant has enabled significant progress by extending the number of known risk loci and implicating new biological pathways. This renewal extends these efforts to multiple genetic ancestries, study designs, and more detailed phenotypic data. We propose the following aims: 1) Continue to expand the IAMDGC resource with new datasets. We have added seven new collaborators and now have access to data from >100,000 participants. 2) Use universal hubs to process and share genomic, phenotypic, and biomarker data. Regeneron Pharmaceuticals has agreed to conduct whole exome sequencing on approximately 40,000 participants at no cost to the grant. By statistical imputation on the remaining GWASed samples, we will create an extremely large dataset. We will continue to house the data in two analytic hubs (US and Europe) to simplify access and provide computational and analytic support. 3) Perform detailed analyses on the extensive resulting dataset. The dataset (87,542 cases/controls and 13,766 related individuals in nearly 6,000 families) enables testing of numerous genetic hypotheses underlying clinical subtypes, biomarkers, effects of rare variants, and variability in the genetic architecture across ancestries. The initial processing and analysis of the combined genomic data will be overseen through this application and results will be available to all members. We have an efficient process allowing members to propose additional studies and the broader research community to access these data and computational and analytical support through the appropriate analytic hub. 4) Support the logistics and administration of the IAMDGC. Successful collaboration requires constant communication and support. We will continue our yearly IAMDGC-specific face-to-face meeting, a second half- day meeting for those attending the ARVO annual meeting, and regular teleconference calls. Our goal is to greatly advance the understanding of AMD pathophysiology (using genomics as our foundational guide) and thus speed the development of better treatments and/or preventions of AMD.

ABSTRACT Alzheimer disease (AD) is the leading cause of dementia in older adults and occurs in all ethnic and racial groups. Genetic studies of AD have mostly been performed in non-Hispanic Whites (NHW) of Northern European (NE) ancestry. Only recently have efforts in AD started to expand into other populations, such as African-Americans (AA) and Hispanics (HI), and have a ready demonstrated differences in both risk effect size (e.g., APOE in AA and HI) and risk loci (e.g., ABCA7 in AA). Further evaluation demonstrates that genetic ancestry (as opposed to environmental/cultural factors) likely underlie at least part of this heterogeneity. Individuals with the Amerindian (AI) ancestry remain one of the most underrepresented groups in AD. Importantly, the NHW datasets did not differentiate among the Europeans (EU), whereas recent investigations showed that these pan-European results only partially overlap with the findings from populations from the Iberian Peninsula (IP) with Southern European (SE) ancestry. Caribbean and South American Hispanic populations are admixed with both AI and SE, thus making their study a critical scientific objective. Our proposed study enables testing the generalization of findings from NHW to these other ancestries, as well as identify AD risk/protective factors correlated specifically with AI and SE ancestry. Our results will allow for a better and more complete understanding of the genetic architecture of AD which will help improve disease prediction, prevention, diagnosis, and treatment in AI, admixed Hispanic populations, and beyond. To accomplish these goals, we propose three aims. In Aim 1 we will characterize known AD loci in admixed populations with AI and SE ancestry. This includes expanding collections, generalizing known AD loci to AI/SE populations, and variant discovery through admixture mapping and fine-mapping. In Aim 2 we will extend our Puerto Rican dataset by expanding PR multiplex families. This will allow more powerful linkage analyses, longitudinal neurocognitive and biomarker data, and the initiation of a brain donation program. Finally, in Aim 3 we will perform functional follow-up of variants using bioinformatics approaches, assessment of AD biomarkers, and assessment of cellular function using IPSc.