Reproducible Peak Clusters on Differential Mouse Mortality Curves and Their Relation to the Gompertz Model

Abstract

It is shown that differentiation of mouse mortality curves (number of animals that died at a certain age plotted versus their lifespan) results in the appearance of eight clearly distinguished clusters of peaks corresponding to increased mortality rates. Smoothing of the original mortality curves and subsequent transformation of the differential mortality curves according to the Gompertz model makes the peaks and the corresponding clusters less pronounced and drives the logarithm of the force mortality curve toward a straight line. The positions of the clusters on the lifespan axis (expressed in days) were calculated as weighted means by dividing the sum of the products of multiplication of the peak heights and their position on the lifespan axis by the sum of the peak heights within a cluster. To prove that the peaks and their clusters are not random, we have demonstrated that the positions of the clusters on the lifespan axis do not depend on the extent of mortality curve smoothing or the group of mice analyzed.