A0458
Title: Testing microbiome associations with survival times
Authors: Yijuan Hu - Emory University (United States) [presenting]
Abstract: Finding microbiome associations with possibly censored survival times is an important problem, especially as specific taxa could serve as biomarkers for disease prognosis or as targets for therapeutic interventions. The existing methods are restricted to testing associations at the community level and do not provide results at the individual taxon level. An ad hoc approach testing each taxon with a survival outcome using the Cox proportional hazard model may not perform well in the microbiome setting with sparse count data and small sample sizes. The linear decomposition model (LDM) has been previously developed for testing continuous or discrete outcomes that unify community-level and taxon-level tests into one framework. The LDM is extended to test survival outcomes. The use of the Martingale residuals or the deviance residuals obtained is proposed from the Cox model as continuous covariates in the LDM. Further tests that combine the results of analyzing each set of residuals separately are constructed. Using simulated data, it is shown that the LDM-based tests preserved the false discovery rate for testing individual taxa and had good sensitivity. An analysis of data on the association of the gut microbiome and the time to acute graft-versus-host disease revealed several dozen associated taxa that would not have been achievable by any community-level test, as well as improved community-level tests by the LDM over existing methods.
