Affordable Access

deepdyve-link
Publisher Website

Sequence kernel association test for survival outcomes in the presence of a non-susceptible fraction.

Authors
  • Lakhal-Chaieb, Lajmi1
  • Simard, Jacques2
  • Bull, Shelley3, 4
  • 1 Département de mathématiques et de statistique, Université Laval, 1045 de la médecine, Québec G1V 0A6, Canada. , (Canada)
  • 2 Département de médecine moléculaire, Chaire de recherche du Canada en encogénétique, Université Laval, Québec G1V 0A6, Canada. , (Canada)
  • 3 Dalla Lana School of Public Health, University of Toronto, 6th floor, Health Sciences Building, 155 College Street, Toronto, Ontario M5T3M7 Canada. , (Canada)
  • 4 The Lunenberg-Tanenbaum Research Institute, Sinai Health System, 60 Murray Street, Toronto, Ontario M5T 3L9 Canada. , (Canada)
Type
Published Article
Journal
Biostatistics (Oxford, England)
Publication Date
Jul 01, 2020
Volume
21
Issue
3
Pages
518–530
Identifiers
DOI: 10.1093/biostatistics/kxy075
PMID: 30590388
Source
Medline
Keywords
Language
English
License
Unknown

Abstract

In this work, we propose a single nucleotide polymorphism set association test for survival phenotypes in the presence of a non-susceptible fraction. We consider a mixture model with a logistic regression for the susceptibility indicator and a proportional hazards regression to model survival in the susceptible group. We propose a joint test to assess the significance of the genetic variant in both logistic and survival regressions simultaneously. We adopt the spirit of SKAT and conduct a variance-component test treating the genetic effects of multiple variants as random. We derive score-type test statistics, and we investigate several approaches to compute their $p$-values. The finite-sample properties of the proposed tests are assessed and compared to existing approaches by simulations and their use is illustrated through an application to ovarian cancer data from the Consortium of Investigators of Modifiers of BRCA1 and BRCA2. © The Author 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: [email protected]

Report this publication

Statistics

Seen <100 times