Premium
Statistical challenges in high‐dimensional molecular and genetic epidemiology
Author(s) -
Bull Shelley B.,
Andrulis Irene L.,
Paterson Andrew D.
Publication year - 2018
Publication title -
canadian journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.804
H-Index - 51
eISSN - 1708-945X
pISSN - 0319-5724
DOI - 10.1002/cjs.11342
Subject(s) - trait , genetic association , molecular epidemiology , breast cancer , biology , disease , genome wide association study , genetic epidemiology , data science , computational biology , evolutionary biology , medicine , computer science , cancer , gene , genetics , pathology , genotype , single nucleotide polymorphism , programming language
Molecular and genetic association studies conducted in well‐characterized longitudinal cohorts offer a powerful approach to investigate factors influencing disease course or complex trait expression. As measurement technologies continue to develop and evolve, studies based on existing cohorts raise methodological challenges. Five such challenges are illustrated in two long‐term inter‐disciplinary collaborations. In one, molecular genetic prognostic factors in the natural history of node‐negative breast cancer are investigated using a combination of hypothesis‐testing and hypothesis‐generating molecular approaches. In the other, genome‐wide association methods are applied to identify genes for multiple traits in extended follow‐up data from participants of a therapeutic RCT in type 1 diabetes. The Canadian Journal of Statistics 46: 24–40; 2018 © 2017 Statistical Society of Canada