Functional Expression Regulating N-terminal domain of K v 4.2. The p.F11 residue plays a crucial role in the binding to the potas- sium channel-Interacting protein (KChIP). Mutations in this resi- due are known to disrupt KChIP binding, trafficking, and functional modulation of the K v 4.2 channel (Kunjilwar et al., 2013). Conclusions: The genetic data suggest that K v 4.2, the molec- ular partner of DPP6, is intolerant to mutations. Together, our re- sults as well as the specific protein function, warrant further investigation of this multimeric protein complex in the pathogen- esis of neurodegenerative brain diseases. P3-129 MAPPING FUNCTIONAL REGULATORY VARIANTS ATALZHEIMER’S DISEASE RISK LOCI Mariet Allen 1 , Jenny M. Bredenberg 1 , Joseph S. Reddy 1 , Vivekananda Sarangi 2 , Minerva M. Carrasquillo 1 , Xue Wang 1 , Sarah J. Lincoln 1 , Thuy Nguyen 1 , Kimberly G. Malphrus 1 , Dennis W. Dickson 1 , Yan W. Asmann 1 , Nilufer Ertekin-Taner 1 , 1 Mayo Clinic, Jacksonville, FL, USA; 2 Mayo Clinic, Rochester, MN, USA. Contact e-mail: Allen.Mariet@mayo.edu Background: Genome-wide association studies have identified com- mon variants associated with risk for Alzheimer’s disease (AD), some of which also associate with brain expression levels of near-by genes. Systems biology approaches including transcription profiling and co-expression network analysis can also nominate additional genes and pathways that may be perturbed in AD. Impor- tantly these studies implicate the likely influenced gene(s), and nominate transcriptional regulation as an important factor in AD pathogenesis; however the precise regulatory mechanism and/or functional genetic variants are largely not yet known. We hypothe- size that in-depth sequence analysis of genes and regulatory regions implicated by eQTL and systems biology studies may identify functional regulatory disease risk variants at both known and novel loci. Methods: We previously collected gene expression measures from two brain regions (temporal cortex and cerebellum) for w200 AD subjects and w200 subjects with other pathologies (nonAD) using the Illumina WG-DASL array, and genome-wide genotypes using the Illumina Hap300 chip. We performed eQTL analysis using PLINK, differential expression analysis using R sta- tistical software, and co-expression network analysis using WGCNA. To identify candidate regulatory variants at selected loci, we performed targeted sequencing in up to 142 of these same samples using the Agilent Haloplex HS target enrichment sys- tem. Results: Our prior work indicated that AD risk GWAS variants were associated with expression of CLU, ABCA7 and PILRB. Tran- scription profiling analyses identified 10 genes differentially ex- pressed (DEG) between AD and nonAD subjects across both brain regions studied. Co-expression network analysis identified 8 hub genes with high connectivity in “immune” networks enriched for AD candidate genes. Sequencing is being performed across both coding and non-coding regions for CLU, ABCA7 and PILRB and 12 of the 18 genes nominated by the systems biology approach. Following quality control, identified variants will be tested for as- sociation with gene expression and AD risk, using available data. Annotation databases will be used to further filter significant vari- ants and nominate those most likely to have functional regulatory effects. These variants will be validated using model systems. Conclusions: The identification of functional, AD risk, regulatory variants is expected to provide novel insights into the pathophysi- ology of this disease and nominate therapeutic targets. P3-130 NIA GENETICS OF ALZHEIMER’S DISEASE DATA STORAGE SITE (NIAGADS): ALZHEIMER’S GENOMICS DATABASE Emily Greenfest-Allen 1 , Prabhakaran Gangadharan 1 , Amanda B. Kuzma 1 , Yuk Yee Leung 1 , Liming Qu 1 , Otto Valladares 1 , Christian J. Stoeckert, Jr. 1 , Li-San Wang 2 , 1 University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA; 2 University of Pennsylvania, Philadelphia, PA, USA. Contact e-mail: allenem@pennmedicine. upenn.edu Background: The Alzheimer’s GenomicsDB is a searchable data- base of Alzheimer’s disease (AD)-relevant variant annotations. It provides public access to variant annotations from the Alzheimer’s Disease Sequencing Project (ADSP) and genome-wide association study (GWAS) summary statistics datasets deposited at NIAGADS, a national genetics data repository for AD and related neuropathol- ogies. These data are linked to dbSNP, NHGRI, and 1000 Genomes variant annotations, genes and gene function, pathways, and func- tional genomics datasets, allowing researchers to easily search, discover, and better understand AD-associated variants. Methods: The Alzheimer’s GenomicsDB is powered by a big-data optimized relational database system and uses OBO Foundry ontologies to consistently annotate study design and experimental details for pro- vided datasets, facilitating data harmonization and integration. This system enables efficient and sophisticated real-time data-mining of the GWAS summary statistics datasets and is made accessible on GenomicsDB website by a flexible search interface that allows users to define search strategies via interactive graphical work- flows. Using strategies, users can combine (via Boolean logic or co- location queries), filter, transform, and analyze search results, or make comparisons to sequence annotations or uploaded user data. Strategy results can also be saved and shared privately or published. The Alzheimer’s GenomicsDB also provides resources for exploring sequencing features and genomic regions in depth. Our genome browser allows users to visually inspect GWAS datasets and compare to functional genomics or personal tracks. Detailed gene and variant reports include ADSP variant annotations and highlight AD-risk associated alleles from NIAGADS GWAS data- sets. Results: The NIAGADS Alzheimer’s Genomics Database (https://www.niagads.org/genomics) makes available 43 GWAS summary datasets from 11 NIAGADS accessions and more than >150 million variant annotations, including w30 million (5 million novel) variants identified as AD-relevant by ADSP, for interactive exploration and data-mining. Conclusions: With a newly redesigned, efficient, search interface and comprehensive record pages linking summary statistics to variant and gene annotations, the Ge- nomicsDB is a rich resource and valuable tool for AD research. P3-131 LOOKING FOR THERAPEUTIC ENTRY POINTS FOR ALZHEIMER’S DISEASE: LESSONS LEARNED FROM AGNOSTIC TRANS-CO-REGULATORY NETWORK ANALYSES OF APOE, TREM2, PLCG2 AND ABI3 LOCI Agust  ın Ruiz 1,2,3 , Laura Madrid 4 , Itziar de Rojas 3 , Antonio Gonz alez- P erez 4 , Santos Ma~ nes 5 , Alfredo Ramirez 6 , Bego~ na Hern andez-Olasagarre 3 , Mar  ıa Eugenia S aez 4 and IMI 2 -ADAPTED, 1 Facultat de Medicina i Ci encies de la Salut, Universitat Internacional de Catalunya, Sant Cugat del Vall es, Spain; 2 Fundaci o ACE, Barcelona Alzheimer Treatment and Research Center, Barcelona, Spain; 3 Neuroscience Center, Fundaci o ACE, Poster Presentations: Tuesday, July 24, 2018 P1117