Supplementary Figure 5: Exploration of TP53-3p interaction with respect to patient survival.
From: Multi-tiered genomic analysis of head and neck cancer ties TP53 mutation to 3p loss
(a) Number of patients surviving or deceased for various time intervals. (b) Statistical models fit using logistic regression. CIN indicates chromosomal instability, measured by the fraction of deleted genes per tumor genome. (c,d) Performance of each logistic regression model in leave-one-out cross-validation to assess the ability of different combinations of genomic variables to predict patient outcomes. For a description of regression formulation, see the Online Methods. For multivariate Cox analysis of the best model, m5, using the full censored data set, see Supplementary Table 7.
