Maximum Likelihood Estimation of Population Growth Rates Based on the Coalescent

Department of Genetics, University of Washington, Seattle, Washington 98195, USA.
Genetics (Impact Factor: 5.96). 06/1998; 149(1):429-34.
Source: PubMed


We describe a method for co-estimating 4Nemu (four times the product of effective population size and neutral mutation rate) and population growth rate from sequence samples using Metropolis-Hastings sampling. Population growth (or decline) is assumed to be exponential. The estimates of growth rate are biased upwards, especially when 4Nemu is low; there is also a slight upwards bias in the estimate of 4Nemu itself due to correlation between the parameters. This bias cannot be attributed solely to Metropolis-Hastings sampling but appears to be an inherent property of the estimator and is expected to appear in any approach which estimates growth rate from genealogy structure. Sampling additional unlinked loci is much more effective in reducing the bias than increasing the number or length of sequences from the same locus.

Download full-text


Available from: Mary K Kuhner, Jul 11, 2014
  • Source
    • "Most methods used to calculate population size changes are based on statistical deviation from a specific model (e.g., constant size, Rogers 1995) by permutations (Excoffier et al. 2005), and fluctuations of effective population size (N e ) need to be assessed independently (Kuhner et al. 1998); however, it is now possible to estimate population sizes and fluctuations through time without a priori limitations from multiple loci (Ho and Shapiro 2011). We tested the population expansion hypothesis using EBSPs (Extended Bayesian Skyline Plots; Heled and Drummond 2008) implemented in BEAST 1.7 with three independent runs of 100 9 10 6 generations and a burn-in of 10 9 10 6 steps. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The Pleistocene was characterized by worldwide shifts in community compositions. Some of these shifts were a result of changes in fire regimes, which influenced the distribution of species belonging to fire-dependent communities. We studied an endangered juniper–oak shrubland specialist, the black-capped vireo (Vireo atricapilla). This species was locally extirpated in parts of Texas and Oklahoma by the end of the 1980s as a result of habitat change and loss, preda-tion, brood parasitism, and anthropogenic fire suppression. We sequenced multiple nuclear loci and used coalescence methods to obtain a deeper understanding of historical population trends than that typically available from microsatellites or mtDNA. We compared our estimated population history, a long-term history of the fire regime and ecological niche models representing the mid-Holocene, last glacial maximum, and last interglacial. Our Bayesian skyline plots showed a pattern of historical population fluctuation that was consistent with changing fire regimes. Genetic data suggest that the species is genetically unstructured, and that the current population should be orders of magnitude larger than it is at present. We suggest that fire suppression and habitat loss are primary factors contributing to the recent decline of the BCVI, although the role of climate change since the last glacial maximum is unclear at present.
    Ecology and Evolution 11/2015; DOI:10.1002/ece3.1811 · 2.32 Impact Factor
  • Source
    • "This effective population size parameter need not be a static number—it can itself be a value that increases or decreases through time (Griffiths & Tavare 1994). By discretizing historical periods and allowing different periods to exhibit patterns of population increase or decrease, one can model complex histories of effective population size increase or decrease that can be compared using likelihood and model comparison methods (Griffiths & Tavare 1994; Kuhner et al. 1998; Pybus et al. 2000; Drummond et al. 2005; Minin et al. 2008). Blair et al. (2013) used such an approach to determine that population sizes in lineages of Asian tree frogs (Polypedates leucomystax) had recently increased, helping them to understand how climate has affected these frogs through time. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Molecular ecology has become one of the key tools in the modern conservationist's kit. Here we review three areas where molecular ecology has been applied to amphibian conservation: genes on landscapes, within-population processes, and genes that matter. We summarize relevant analytical methods, recent important papers from the amphibian literature, and conservation implications for each section. Finally, we include five in-depth examples of how molecular ecology has been successfully applied to specific amphibian systems. This article is protected by copyright. All rights reserved.
    Molecular Ecology 10/2015; 24(20). DOI:10.1111/mec.13391 · 6.49 Impact Factor
  • Source
    • "Effective population size measures genetic diversity present in the population and relates to census population size if certain assumptions are met [Wakeley and Sargsyan, 2009]. Many early coalescent-based phylodynamic methods required strict parametric assumptions about the effective population size trajectory, such as constant through time [Griffiths and Tavaré, 1994] or exponential growth [Drummond et al., 2002, Kuhner et al., 1998]. A major alternative arose with the advent of nonparametric methods, one of the earliest and most influential being the piecewise constant classical skyline model [Pybus et al., 2000]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Phylodynamics seeks to estimate effective population size fluctuations from molecular sequences of individuals sampled from a population of interest. One way to accomplish this task formulates an observed sequence data likelihood exploiting a coalescent model for the sampled individuals' genealogy and then integrating over all possible genealogies via Monte Carlo or, less efficiently, by conditioning on one genealogy estimated from the sequence data. However, when analyzing sequences sampled serially through time, current methods implicitly assume either that sampling times are fixed deterministically by the data collection protocol or that their distribution does not depend on the size of the population. Through simulation, we first show that, when sampling times do probabilistically depend on effective population size, estimation methods may be systematically biased. To correct for this deficiency, we propose a new model that explicitly accounts for preferential sampling by modeling the sampling times as an inhomogeneous Poisson process dependent on effective population size. We demonstrate that in the presence of preferential sampling our new model not only reduces bias, but also improves estimation precision. Finally, we compare the performance of the currently used phylodynamic methods with our proposed model through clinically-relevant, seasonal human influenza examples.
Show more