Abstract
In linear regression, a residual measures how far a subject's observation is from expectation; in survival analysis, a subject's Martingale or deviance residual is sometimes interpreted similarly. Here we consider ways in which a linear regression-like interpretation is not appropriate for Martingale and deviance residuals, and we develop a novel time-to-event residual which does have a linear regression-like interpretation. We illustrate the utility of this new residual via simulation of a time-to-event genome-wide association study, motivated by a real study seeking genetic modifiers of Duchenne Muscular Dystrophy. By virtue of its linear regression-like characteristics, our new residual may prove useful in other contexts as well.
Highlights
In this paper we develop a new form of survival analysis residual with linear regression-like (LRL) properties, in a sense to be made clear in what follows
In this paper we argue that Martingale residual (MR) and deviance residual (DR) do not lend themselves to an LRL interpretation, but that we can modify DR to do so
We introduce a new form of residual, which we call an ordinary time-to-event (OTE) residual, so called because it affords an ordinary LRL interpretation in the context of time-to-event data (TE), and we compare OTE with MR, DR and with an ordinary linear regression residual (OLRR)
Summary
In this paper we develop a new form of survival analysis residual with linear regression-like (LRL) properties, in a sense to be made clear in what follows. The simplest way to motivate this work is by describing a target application: discovery of genes that modify the Duchenne muscular dystrophy (DMD) phenotype. DMD involves progressive muscle tissue loss with replacement by fat and fibrotic tissue, and is currently without a cure. Patients typically become reliant on wheelchairs by early to mid-adolescence, but some maintain ambulation substantially longer, and age at loss of ambulation (LOA) is an important clinical indicator of disease progression. A great deal is known about the gene (DMD) that causes DMD, including the fact that modifier genes influence the rate of disease progression in a DMD mouse model [3, 4]; evidence for modifiers exists in humans as well [5,6,7,8]. The discovery of modifier genes in humans has implications both for therapeutics and for the design of DMD clinical trials
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
