Biphasic survival analysis of trypanotolerance QTL in mice

A marker-assisted introgression (MAI) experiment was conducted to transfer trypanotolerance quantitative trait loci (QTL) from a donor mouse strain, C57BL/6, into a recipient mouse strain, A/J. The objective was to assess the effect of three previously identified chromosomal regions on mouse chromosomes 1 (MMU1), 5 (MMU5) and 17 (MMU17) in different genetic backgrounds on the survival pattern following infection with Trypanosoma congolense. An exploratory data analysis revealed a biphasic pattern of time to death, with highly distinct early and late mortality phases. In this paper, we present survival analysis methods that account for the biphasic mortality pattern and results of reanalyzing the data from the MAI experiment. The analysis with a Weibull mixture model confirmed the biphasic pattern of time to death. Mortality phase, an unobserved variable, appears to be an important factor influencing survival time and is modeled as a binary outcome variable using logistic regression analysis. Accounting for this biphasic pattern in the analysis reveals that a previously observed sex effect on average survival is rather an effect on proportion of mice in the two mortality phases. The C57BL/6 (donor) QTL alleles on MMU1 and MMU17 act dominantly in the late mortality phase while the A/J (recipient) QTL allele on MMU17 acts dominantly in the early mortality phase. From this study, we found clear evidence for a biphasic survival pattern and provided models for its analysis. These models can also be used when studying defense mechanisms against other pathogens. Finally, these approaches provide further information on the nature of gene actions.