In a recent post, I demonstrated the use of the Kaplan-Meier estimator for estimating survival curves of fictional characters undergoing treatment in a fictional drug trial. Here I illustrate the Kaplan-Meier estimator on real data, data that is unique from normal survival analysis data in that the event under consideration is neither time until death nor time until product failure.
Sharova et. al.  published the measured half-lives of roughly 20,000 mouse genes. If we treat the observation of an mRNA reaching half-life during their experiment as the event, we can estimate a survival curve for mRNA half-lives using the Kaplan-Meier estimator. Since the experiment ran for only 24 hours, those few mRNAs with longer half-lives than 24 hours show up as right-censored data. An example of the source data is:
Plotting this data using R’s Kaplan-Meier estimator yields:
Looking at this plot of the estimated survival curve, we can see that the median mRNA half-life is about seven hours. R’s Kaplan-Meier estimator estimates that the median is 7.08 hours:
1. Lioudmila V. Sharova, Alexei A. Sharov, Timur Nedorezov, Yulan Piao, Nabeebi Shaik, and Minoru S.H. Ko. Database for mRNA Half-Life of 19 977 Genes Obtained by DNA Microarray Analysis of Pluripotent and Differentiating Mouse Embryonic Stem Cells. DNA Res (2009) 16(1): 45-58 first published online November 11, 2008 doi:10.1093/dnares/dsn030