Frontiers
Browse
Table1.DOCX (38.11 kB)

Table1.DOCX

Download (38.11 kB)
dataset
posted on 2018-04-04, 04:32 authored by Maria V. Fernández, John Budde, Jorge L. Del-Aguila, Laura Ibañez, Yuetiva Deming, Oscar Harari, Joanne Norton, John C. Morris, Alison M. Goate, NIA-LOAD family study group, NCRAD, Carlos Cruchaga

Gene-based tests to study the combined effect of rare variants on a particular phenotype have been widely developed for case-control studies, but their evolution and adaptation for family-based studies, especially studies of complex incomplete families, has been slower. In this study, we have performed a practical examination of all the latest gene-based methods available for family-based study designs using both simulated and real datasets. We examined the performance of several collapsing, variance-component, and transmission disequilibrium tests across eight different software packages and 22 models utilizing a cohort of 285 families (N = 1,235) with late-onset Alzheimer disease (LOAD). After a thorough examination of each of these tests, we propose a methodological approach to identify, with high confidence, genes associated with the tested phenotype and we provide recommendations to select the best software and model for family-based gene-based analyses. Additionally, in our dataset, we identified PTK2B, a GWAS candidate gene for sporadic AD, along with six novel genes (CHRD, CLCN2, HDLBP, CPAMD8, NLRP9, and MAS1L) as candidate genes for familial LOAD.

History