Maximum likelihood estimates under k-allele models with selection can be numerically unstable

The stationary distribution of allele frequencies under a variety of Wright-Fisher k-allele models with selection and parent independent mutation is well studied. However, the statistical properties of maximum likelihood estimates of parameters under these models are not well understood. Under each of these models there is a point in data space which carries the strongest possible signal for selection, yet, at this point, the likelihood is unbounded. This result remains valid even if all of the mutation parameters are assumed to be known. Therefore, standard simulation approaches used to approximate the sampling distribution of the maximum likelihood estimate produce numerically unstable results in the presence of substantial selection. We describe the Bayesian alternative where the posterior distribution tends to produce more accurate and reliable interval estimates for the selection intensity at a locus.

• Publication year

  2009

• Venue

  The Annals of Applied Statistics

• Publication date

  2009-09-01

• Fields of study

  Biology, Mathematics

• Identifiers
  DOI 10.1214/09-AOAS237arXiv 0910.1664
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Exploring Population Genetic Models With Recombination Using Efficient Forward-Time Simulations

  2008cited by this paper
• The global distribution of clinical episodes of Plasmodium falciparum malaria

  2005cited by this paper
• Mathematical Population Genetics : I. Theoretical Introduction

  2004cited by this paper
• SNP haplotypes and allele frequencies show evidence for disruptive and balancing selection in the human leukocyte receptor complex

  2004cited by this paper
• A population genetic study of Borrelia burgdorferi sensu stricto from eastern Long Island, New York, suggested frequency-dependent selection, gene flow and host adaptation.

  2004influential reference
• When can one detect overdominant selection in the infinite-alleles model?

  2003cited by this paper
• Likelihoods and simulation methods for a class of nonneutral population genetics models.

  2001cited by this paper
• Linkage disequilibrium, gene trees and selfing: an ancestral recombination graph with partial self-fertilization.

  2000cited by this paper
• The ancestral graph and gene genealogy under frequency-dependent selection.

  1999cited by this paper
• The role of population size in molecular evolution.

  1999cited by this paper
• An ancestral recombination graph

  1997cited by this paper
• Archaic African and Asian lineages in the genetic ancestry of modern humans.

  1997cited by this paper
• Adaptation and selection

  1981cited by this paper
• Heterosis or neutrality?

  1977cited by this paper
• Monte Carlo Sampling Methods Using Markov Chains and Their Applications

  1970cited by this paper
• THE SICKLE‐CELL and HAEMOGLOBIN C GENES IN SOME AFRICAN POPULATIONS

  1956cited by this paper
• Genetics of human populations.

  1948cited by this paper
• Technology and Medicine,

  year unknowncited by this paper

• Inference from the stationary distribution of allele frequencies in a family of Wright-Fisher models with two levels of genetic variability.

  2018cites this paper
• Stationary Allele Frequency Distributions

  2013cites this paper
• An explicit transition density expansion for a multi-allelic Wright-Fisher diffusion with general diploid selection.

  2012cites this paper
• Efficient Simulation and Likelihood Methods for Non-Neutral Multi-Allele Models

  2012cites this paper
• Inference on the strength of balancing selection for epistatically interacting loci.

  2011cites this paper
• Statistical Applications in Genetics and Molecular Biology Estimation of Selection Intensity under Overdominance by Bayesian Methods

  2011cites this paper