Premium
P1‐129: CELL‐AUTONOMOUS AND EFFECTS OF NEURONAL BIN1 LOSS IN VIVO
Author(s) -
Rajamohamed Sait Hameetha Banu,
Mcavoy Kathleen,
Marsh Galina,
Peterson Michael,
Reynolds Taylor L.,
Gag Jake,
Geisler Sarah,
Roberts Chris,
Ransohoff Richard,
Crotti Andrea
Publication year - 2019
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2019.06.684
Subject(s) - hippocampal formation , in vivo , hippocampus , biology , neurodegeneration , small hairpin rna , neuroscience , microbiology and biotechnology , endocrinology , medicine , gene knockdown , cell culture , genetics , disease
Initiative (ADNI) participants, we derived 6 PRSs: 3 biomarkerPRSs (PRSA, PRST, PRSN), their combination (PRSA/T/N), a “Traditional” PRSTraditional, and an IGAP PRSIGAP-A/T/N (Figure 1, Table1). Specifically, we retained SNPs with GWAS p-value 5e-05. We calculated the correlation between SNPs, and to avoid collinearity we retained one SNP from each LD-block. Using the relevant endophenotype as the response variable, forward selection was applied for SNP selection and weight estimation. The APOE e4 SNP was forced into every PRS model by design. SNPs in PRSTraditional were selected by forward selection on Cox model. IGAP odds ratios used as PRSIGAP-A/T/N weights. Timedependent area under the curve (AUC) was used to assess the predictive performance of each PRSA, PRST, PRSN, PRSA/T/N, PRSIGAP-A/T/N PRSTraditional, as well as the simultaneous presence of PRSA, PRST, PRSN in a model. Results: Accounting simultaneously for PRSA, PRST, and PRSN in AD AAO prediction, achieves significantly better performance (Wilcoxon p-value < 2.2e-16) than the PRSTraditional, PRSIGAP-A/T/N or univariate PRSA, PRST, PRSN models (Figure 2). When assessed separately, PRSN has the best performance, followed by PRST, and PRSA. Conclusions: Individualized biomarker-specific PRS based on the patient’s biomarker profile may be more effective and informative compared to current PRS approaches. Further validation is warranted in larger samples. Optimized PRS profiling, after validation, may be a key supporting approach for the development of a precision medicine approach to AD.