Survival data for: Phylogenetic reconstruction of ancestral aging rates in the primate lineage
Data files
Apr 20, 2026 version files 166.03 KB
-
primate_life_tables.tsv
160.18 KB
-
README.md
5.84 KB
Abstract
Median lifespans of primates show nearly 10-fold variation, ranging from ∼8 years in marmosets to ∼80 years in humans. The molecular mechanisms that govern this variation and how they evolved remain poorly understood. In this study, we implemented a novel phylogenetic Gompertzian survival framework to leverage the evolutionary history of primates in order to estimate parameters of aging for 38 captive primate species. We find that baseline hazards (at the time of sexual maturity) and adult aging rates display significant variation, and aging rates are more evolutionarily conserved than baseline hazards (Pagel lambda=0.96 vs lambda=0.34, respectively). Furthermore, we find that aging rates and baseline hazards do not show a pattern of phylogenetic covariation, suggesting that differential evolutionary pressures may act on these traits. Aging rates were strongly correlated with body weights, with a notable exception within the Ape family, where aging rates remained approximately unchanged despite large differences in body weight. Based on the reconstruction of ancestral aging rates, we estimate that the ancestor of Apes was likely aging at a similar rate to modern humans.
Dataset DOI: 10.5061/dryad.m905qfvd5
Description of the data and file structure
File: primate_life_tables.tsv - contains adult mortality rates (post sexual maturity) for primates in captivity across multiple primate centers and zoos, aggregated at 1-year intervals. This file is generated using as a part of data analysis pipeline.
The hazard of mortality (qx) for the time interval (t to t+dt) is estimated as the number of death events during the time interval divided by the effective number of individuals at risk:
qx = Ndied/Nrisk
Where Nrisk = Nalive + 0.5 * Ncensored is the average number of survivors at the midpoint of the time interval[17]. We used a two-year interval to obtain a sufficiently large number of death events for species with small populations.
Code/software
Codebase: https://github.com/eugenemel/PrimateAgingCodebase/
primate_life_tables.tsv Variables
| Column | Description |
|---|---|
species |
Common species name |
sex |
Sex of the cohort: Female, Male, or Both |
source |
Data source: PAD (Primate Aging Database) or studbook records (STUD_BOOOK_2010/2014/2015) |
int.start |
Start age (in years) of the life table interval |
int.end |
End age (in years) of the life table interval |
age_groups |
Interval notation (e.g., [5,6)) |
lost |
Number of individuals censored (lost to follow-up) during the interval |
events |
Number of deaths during the interval |
int.length |
Length of the interval in years (typically 1) |
int.midpoint |
Midpoint age of the interval |
enter |
Number of individuals entering (at risk at the start of) the interval |
rate |
Mortality rate (hazard) for the interval |
surv |
Cumulative survival probability at the end of the interval |
sd |
Standard deviation of the cumulative survival estimate |
log_rate |
Natural log of the mortality rate |
time |
Time point for the interval (same as int.midpoint) |
sexMature |
Age at sexual maturity (years) for the species |
sci_name |
Scientific (Latin) name of the species |
Subject Level Data Access
The primary source of species lifespan information for this study was the Primate Aging Database (PAD). PAD is a curated database that tracks demographic and physiological information for a large number of captive primates across multiple centers and sites in the United States and Europe \cite{Kemnitz2019-og}.
Complete lifespan information was known for 4,083 subjects, 3,966 were alive, and 1,010- were lost to follow-up. Some of the factors that contributed to the loss to follow-up included site closures and lack of tracking after zoo transfers. Additional data were collected from published studbooks: Western Lowland Gorilla (\textit{Gorilla gorilla gorilla}) (n = 1,291, died=264), Black howlers (\textit{Alouatta pigra}) (n = 510, died=239), and Orangutans (\textit{Pongo abelii}) (n = 1,629, died=699) \cite{Harris2010-nr, Wilms2015-ah, Elder2016-zv}.
Median lifespans of primates show nearly 10-fold variation, ranging from ∼8 years in marmosets to ∼80 years in humans. The molecular mechanisms that govern this variation and how they evolved remain poorly understood. Based on a decades-long multi-site curation effort, we have compiled lifespan data for 39 captive primate species, estimated their Gompertzian aging parameters, and reconstructed ancestral aging parameters for major primate clades.
This dataset contains adult life-tables aggregated at 1yr intervals for these primates. It was generated using a data analysis pipeline, which can be found at https://github.com/eugenemel/PrimateAgingCodebase
https://doi.org/10.5281/zenodo.19392919
How to request subject-level data from Primate Aging Database
To request subject-level data, please contact administrators at PAD via their website: https://primatedatabase.org/authentication/registration and request access to the "PrimateAgingRates.zip" dataset.
Kemnitz, J. W. Database for indices of aging in nonhuman primates. Innov. Aging 3, S957–S957 (2019).
Harris, K. International Studbook Southern Black Howler Monkey Studbook (2010). (2010).
Wilms, T. & Others. International Studbook for the western lowland gorilla Gorilla g. gorilla (Savage & Wyman, 1847). Frankfurt, Germany: Frankfurt Zoo (2015).
Elder, M. International studbook of the orangutan. (2016).