P1-051 THE DISTRIBUTION OF APOLIPOPROTEIN E GENOTYPE OVERTHE ADULT LIFESPAN AND IN RELATION TO BIRTH ORIGIN Anne B€ orjesson-Hanson 1 , Silke Kern 2 , Kirsten Mehlig 3 , Henrik Zetterberg 4 , Dag Thelle 3 , Ingmar Skoog 1 ,J€ urgen Kern 5 , Kaj Blennow 6 , 1 Neuroscience and Physiology, Sahlgrenska Academy at the University of Gothenburg, M€ olndal, Sweden; 2 Neuroscience and Physiology, Sahlgrenska Academy at the University of Gothenburg, M€ olndal, Sweden; 3 Sahlgrenska Academy at the University of Gothenburg, Gothenburg, Sweden; 4 Neuroscience and Physiology, Sahlgrenska Academy at the University of Gothenburg, Gothenburg, Sweden; 5 Institute of Neuroscience and Physiology, Sahlgrenska Academy at the University of Gothenburg, M€ olndal, Sweden; 6 Sahlgrenska Academy, University of Gothenburg, M€ olndal, Sweden. Contact e-mail: anne.borjesson@neuro. gu.se Background: The cholesterol regulating apolipoprotein E is encoded by the APOE gene, which exists in three different alleles, ε2, ε3 and ε4. APOE ε4 carriership is associated with various conditions such as cardio- vascular disease and Alzheimer‘s disease. The relative proportion of the three APOE alleles has been found to depend on both ethnicity and age. Several studies suggest a slow but continuous decline in the prevalence of the ε4 allele after the age of 60 years, but the age at which the decline begins varies between studies. The prevalence of the APOE alleles has also been shown to vary geographically. The aim of this study was to consider the dependence of the APOE allele proportions in relation to age and ethnicity simultaneously. Methods: The study sample consists of partici- pants from four different population based studies including 4 579 subjects aged 25 to 99 years, and living in the Gothenburg area at the time of recruitment. Gothenburg is situated on the West-Coast of Sweden and has an immigrant population of 22%, out of which 10% origins from Iran, 9% from Iraq and 7% from Finland 26. Blood samples were collected and APOE (gene map locus 19q13.2) genotyping was performed by mini- sequencing. Self-reported country of origin was used as a proxy for ethnicity. Results: A constant distribution of APOE alleles was found up to the age of 90 years, where after a decline in ε4 prevalence was observed. We found a 2 to 3 times higher prevalence of ε4 among individ- uals born in the Nordic countries compared with subjects from continental Europe or the Middle East, a result that was independent of age in the in- terval between 25 and 80. After age 90, the ε4 prevalence decreased from 16% to 9%. Conclusions: We found that the proportions of APOE geno- types varied with latitude in this European and Middle East population, so that the ε4 frequency increased by latitude of the country of birth. Furthermore, we found a stable distribution of APOE alleles over the life span up to age 90, after which the prevalence of ε4 decreased in the population of Nordic origin. The relatively high turning point may be ex- plained by the inclusion of subjects with APOE -related diseases such as dementia. P1-052 DEEP SEQUENCING ALZHEIMER’S DISEASE ASSOCIATED GENES, CLU, PICALM, CR1, ABCA7, BIN1, MS4A, CD2AP, EPHA1, AND CD33 IDENTIFIES POTENTIAL FUNCTIONAL SNPS Anne Braae, James Turton, Jenny Lord, Christopher Medway, Kristelle Brown, Imelda Barber, Kevin Morgan, University of Nottingham, Nottingham, United Kingdom. Contact e-mail: mrxab14@nottingham.ac.uk Background: There is a strong genetic component to late onset Alzheimer’s disease (AD). Genome wide association studies (GWAS) have identified several loci associated with AD risk. However, the functional variants responsible for the association are still unknown. Uncovering these variants would shed further light on disease mechanisms. The nine GWAS associated genes CLU, PICALM, CR1, ABCA7, BIN1, MS4A, EPHA1, CD2AP and CD33 were investigated using next generation sequencing (NGS) to identify potential causative variants in the GWAS loci. Methods: Eight pools of twelve AD individuals, 96 in total, were sequenced using 100 base pair (bp), paired-end, target-enriched library on the Illumina HiSeq 2000. Following initial quality control and data alignment, post-processing was performed and variants were called using pooled data specific software. Var- iants were further quality control checked, and then annotated using En- sembl’s Variant Effect Predictor, ENCODE (Encyclopedia of DNA Elements) and other in silico databases. Results: Over 3000 variants were called within the nine genes of which 760 were novel. The NGS variant fre- quency estimates were strongly positively correlated with estimates from 1000 Genomes project, indicating successful variant calling. The majority of variants (3205) were non-coding, with only 126 exonic variants called. Functional annotation revealed 43 variants in untranslated regions (UTRs), 44 missense variants and 21 variants predicted to affect splicing. Conclusions: Studies focusing solely on coding variants will miss the ma- jority of the variation in loci surrounding GWAS hits. Variants found in this study with predicted functional effects will be investigated further by direct genotyping and functional experiments. P1-053 OLFACTORY RECEPTOR COPY NUMBER VARIATION INVOLVED IN ALZHEIMER’S DISEASE Caroline Van Cauwenberghe 1 , Karolien Bettens 1 , Sebastiaan Engelborghs 2 , Mathieu Vandenbulcke 3 , Rik R. Vandenberghe 4 , Peter P. De Deyn 5 , Marc Cruts 1 , Christine Van Broeckhoven 6 , Kristel Sleegers 6 , 1 Neurodegenerative Brain Diseases Group, VIB and Institute Born-Bunge, University of Antwerp, Antwerpen, Belgium; 2 Institute Born-Bunge, University of Antwerp and Hospital Network Antwerp, Middelheim and Hoge Beuken, Antwerp, Belgium; 3 University of Leuven and University Hospitals Leuven, Leuven, Belgium; 4 University Hospitals Leuven and University of Leuven (KUL), Leuven, Belgium; 5 Institute Born-Bunge, University of Antwerp and Hospital Network Antwerp, Middelheim and Hoge Beuken and University Medical Center Groningen, Antwerpen, Belgium; 6 Neurodegenerative Brain Diseases Group, VIB and Institute Born-Bunge, University of Antwerp, Antwerp, Belgium. Contact e-mail: caroline.vancauwenberghe@molgen.vib-ua.be Background: Olfactory deficits are frequently present in several neurode- generative disorders in particular Alzheimer disease (AD). Parts of the ol- factory system are impaired in the early stages of the disease. A copy number variation (CNV) of an olfactory receptor gene cluster has been re- ported to be associated with age at onset (AAO) in AD patients. Since this CNV could be a risk factor for disease, we further assessed the role of the olfactory receptor gene cluster copy number and the correlation with biomarker profiles in cerebrospinal fluid (CSF) in a Flanders-Belgian AD patient and control cohort. Methods: We screened the CNVof olfactory receptor genes located on chromosome 14q11.2 in 582 AD patients and 448 control individuals by Multiplex Amplicon Quantification (MAQ) technol- ogy. The assay includes 15 target amplicons and 10 control amplicons. CSF levels of b-amyloid peptide (Ab 1-42), total tau protein (T-tau) and tau Poster Presentations: P1 P322