Disentangling genetic structure for genetic monitoring of complex populations

Brook G. Milligan, Frederick I. Archer, Anne Laure Ferchaud, Brian K. Hand, Elizabeth M. Kierepka, Robin S. Waples

Research output: Contribution to journalArticlepeer-review

11 Scopus citations

Abstract

Genetic monitoring estimates temporal changes in population parameters from molecular marker information. Most populations are complex in structure and change through time by expanding or contracting their geographic range, becoming fragmented or coalescing, or increasing or decreasing density. Traditional approaches to genetic monitoring rely on quantifying temporal shifts of specific population metrics—heterozygosity, numbers of alleles, effective population size—or measures of geographic differentiation such as FST. However, the accuracy and precision of the results can be heavily influenced by the type of genetic marker used and how closely they adhere to analytical assumptions. Care must be taken to ensure that inferences reflect actual population processes rather than changing molecular techniques or incorrect assumptions of an underlying model of population structure. In many species of conservation concern, true population structure is unknown, or structure might shift over time. In these cases, metrics based on inappropriate assumptions of population structure may not provide quality information regarding the monitored population. Thus, we need an inference model that decouples the complex elements that define population structure from estimation of population parameters of interest and reveals, rather than assumes, fine details of population structure. Encompassing a broad range of possible population structures would enable comparable inferences across biological systems, even in the face of range expansion or contraction, fragmentation, or changes in density. Currently, the best candidate is the spatial Λ-Fleming-Viot (SLFV) model, a spatially explicit individually based coalescent model that allows independent inference of two of the most important elements of population structure: local population density and local dispersal. We support increased use of the SLFV model for genetic monitoring by highlighting its benefits over traditional approaches. We also discuss necessary future directions for model development to support large genomic datasets informing real-world management and conservation issues.

Original languageEnglish
Pages (from-to)1149-1161
Number of pages13
JournalEvolutionary Applications
Volume11
Issue number7
DOIs
StatePublished - Aug 2018

Keywords

  • density
  • dispersal
  • genetic monitoring
  • isolation by distance
  • multiple merger coalescent
  • population structure
  • spatial Λ-Fleming-Viot model
  • Λ-coalescent

Fingerprint

Dive into the research topics of 'Disentangling genetic structure for genetic monitoring of complex populations'. Together they form a unique fingerprint.

Cite this