Abstract
Research into longevity and aging involves comparing the size of cohorts at certain points on survival curves. However, this analysis is oversimplified because it provides limited information about the sample structure and the distribution of lifespan as a trait. Here, we introduce a method for estimating lifespan across the entire data range using distribution analysis. More specifically, we propose dividing the lifespan series into intervals, obtaining the frequencies of phenotypes by lifespan within the sample, followed by distribution analysis using the normality criterion. Additionally, to visualize the differences, we propose describing the resulting distributions formally using the normal distribution function and the β-distribution function. We demonstrate that the proposed methodology enables to extract additional information from survival data, providing new insights into the processes that occur in populations in response to genetic interventions and shedding light on their impact on ontogenesis. In particular, we observed that the lifespan distribution in Drosophila may not meet the normality criterion and may take different shapes depending on the line's genotype or in response to genetic interventions. The proposed approach adds a new layer of information to studies of longevity and aging and expands the toolkit of methods used to analyze survival data.
