Biology subjects

Tuljapurkar, S.

Publications and source records attributed to Tuljapurkar, S..

7 recordsLinked to original sources

bioRxiv2023

Migration in uncertain times

Understanding the conditions that promote the evolution of migration is important in ecology and evolution. When environments are fixed and there is one most favorable site, migration to other sites lowers overall growth rate and is not favored. Here we ask, can environmental variability favor migration when there is one best site on average? Previous work suggests that the answer is yes, but a general and precise answer remained elusive. Here we establish new, rigorous inequalities to show (and use simulations to illustrate) how stochastic growth rate can increase with migration when fitness (dis)advantages fluctuate over time across sites. The effect of migration between sites on the overall stochastic growth rate depends on the difference in expected growth rates and the variance of the fluctuating difference in growth rates. When fluctuations (variance) are large, a population can benefit from bursts of higher growth in sites that are worse on average. Such bursts become more probable as the between-site variance increases. Our results apply to many ([≥] 2) sites, and reveal an interplay between the length of paths between sites, the average differences in site-specific growth rates, and the size of fluctuations. Our findings have implications for evolutionary biology as they provide conditions for departure from the reduction principle, and for ecological dynamics: even when there are superior sites in a sea of poor habitats, variability and habitat quality across space determine the importance of migration.

ecology↗

bioRxiv2019

Skewed distributions of LRS: beyond mean and variance

Many field studies find that lifetime reproductive success (LRS) is highly skewed and often multimodal among individuals. Field biologists generate invaluable data on survival and reproductive rates, as a function of age and stage, that are used to parameterise structured models. These models often perform well at predicting population growth and mean LRS, but we do not know whether they accurately predict observed distributions of individual LRS. If the models fail to recreate these distributions, their use may be limited because the LRS is central to understanding life history evolution. We present powerful tools to generate distributions of LRS from age and/or stage structured models. Our methods reveal that structured models do perform well at generating distributions that agree with observations. Our approach also reveals why such skewed distributions arise, and helps resolve a debate about detecting signatures of selection in skewed distributions of LRS.

ecology↗

bioRxiv2019

‘Early Rectangularization’ of under-5 mortality in sub-Saharan Africa: compression, convergence, inequality

Summary paragraphThe epidemiological transition from young to old deaths in high-income countries reduced mortality at all ages, but a major role was played by a decline of infant and child mortality from infectious diseases1,2 that greatly increased life expectancy at birth2,3. Over time, declines in infectious disease continue but chronic and degenerative causes persist4,5, so we might expect under-5 deaths to be concentrated in the first month of life. However, little is known about the age-pattern of this transition in early mortality or its potential limits. Here we first describe the limit using detailed data on Denmark, Japan, France, and the USA-- developed countries with low under-5 mortality. The limiting pattern of under-5 deaths concentrates in the first month, but is surprisingly dispersed over later ages: we call this the early rectangularization of mortality. Then we examine the progress towards this limit of 31 developing countries from sub-Saharan Africa (SSA)--the region with the highest under-5 mortality6. In these countries, we find that early deaths have large age-heterogeneities; and that the age patterns of death is an important marker of progress in the mortality transition at early ages. But a negative association between national income and under-5 mortality levels, confirmed here, does not help explain reductions in child mortality during the transition.

epidemiology↗

bioRxiv2018

Demographic determinants of biometric heritability

The response of quantitative characters to selection depends on their transmission from parents to offspring. A common estimate of this transmission is the biometric heritability defined as the slope of the regression of offspring phenotype on same-aged mid-parent phenotype (i.e. the ratio of the phenotypic parent-offspring covariance over the parental phenotypic variance). This slope is often interpreted as the percentage of phenotypic variation due to additive genetic effects after accounting for confounding factors such as environment, litter or parental effects. However, researchers seldom account for the possible influence of selection on this estimate. Here we study the effect on biometric heritability of fertility and viability selection, as well as phenotype ontogeny (growth) and inheritance from parents to offspring. We present exact formulas for the elasticities of biometric heritability in age-phenotype-structured integral projection models (IPMs), and illustrate these for two iteroparous long-lived species. We find that both viability and fertility selection can strongly affect heritability, mediated by growth and inheritance. Generally, demographic processes that result in parents reproducing at large phenotypes, regardless of their own birth phenotype, decrease heritability. Analysed at equilibrium, our models imply that a heritable character can show no response to selection, if parental phenotypes affect offspring phenotypes and if phenotypes develop with age. Our results further highlight the importance of accounting for demographic processes when estimating heritability.

ecology↗

bioRxiv2018

An Advancing Front of Old Age Human Survival

Old age mortality decline has driven recent increases in lifespans, but there is no agreement about trends in the age-pattern of old deaths. Some hypotheses argue that old-age deaths should have become compressed at high ages, others that old-age deaths should have become more dispersed with age, and yet others are consistent with little change in dispersion. However, direct analyses of old-age deaths presents unusual challenges: death rates at the oldest ages are always noisy; published life tables must assume an asymptotic age pattern of deaths; and the definition of \"old age\" changes as lives lengthen. Here we use robust percentile-based methods to overcome these challenges and show, for 5 decades in 20 developed countries, that old-age survival follows an advancing front, like a traveling wave. The front lies between the 25th and 90th percentiles of old-age deaths, advancing with nearly constant long-term shape but annual fluctuations in speed. The existence of this front leads to several predictions that we verify, e.g., that advances in life expectancy at age 65 are highly correlated with the advance of the 25th percentile, but not with distances between higher percentiles. Our unexpected result has implications for biological hypotheses about human aging, and for future mortality change.

epidemiology↗

bioRxiv2018

Quantifying genetic, environmental and individual stochastic variability in Plantago lanceolata

Predicting ecological and evolutionary population dynamics requires understanding how genetic and environmental parameters influence variation in survival and reproduction among individuals. However such a focus often neglects the stochastic events that individuals experience throughout their lives that also influence survival and reproduction. With an illustrative example, we quantify and illustrate the influence of such non-selective demographic variability on population dynamics using size-structured matrix models of an experimental population of Plantago lanceolata. Our analysis shows that variation in survival and reproduction among individuals explained by environment, genes, and their interaction was modest compared to the stochastic variation in lifespan and reproduction. We illustrate how expectations on population growth, based on expected lifetime reproduction and generation time, can be misleading when variance in reproduction among individuals of the same genotype (full sibs) was large. Such large within genotype variance can lower population growth, fitness. Our results accompany recent investigations that call for more focus on stochastic variation in survival and reproduction, rather than dismissal of this variation as uninformative noise.

evolutionary biology↗

bioRxiv2017

Drivers of diversity in individual life courses: Sensitivity of the population entropy of a Markov chain

Individuals differ in their life courses, but how this diversity is generated, how it has evolved and how maintained is less understood. However, this understanding is crucial to comprehend evolutionary and ecological population dynamics. In structured populations, individual life courses represent sequences of stages that end in death. These sequences can be described by a Markov chain and individuals diversify over the course of their lives by transitioning through diverse discrete stages. The rate at which stage sequences diversify with age can be quantified by the population entropy of a Markov chain. Here, we derive sensitivities of the population entropy of a Markov chain to identify which stage transitions generate--or contribute--most to diversification in stage sequences, i.e. life courses. We then use these sensitivities to reveal potential selective forces on the dynamics of life courses. To do so we correlated the sensitivity of each matrix element (stage transition) with respect to the population entropy, to its sensitivity with respect to fitness {lambda}, the population growth rate. Positive correlation between the two sensitivities would suggest that the stage transitions that selection has acted most strongly on (sensitivities with respect to {lambda}) are also those that contributed most to the diversification of life courses. Using an illustrative example on a seabird population, the Thick-billed Murres on Coats Island, that is structured by reproductive stages, we show that the most influential stage transitions for diversification of life courses are not correlated with the most influential transitions for population growth. Our finding suggests that observed diversification in life courses is neutral rather than adaptive. We are at an early stage of understanding how individual level dynamics shape ecological and evolutionary dynamics, and many discoveries await.

evolutionary biology↗