Graph Theory and Network Models in Landscape Genetics

Murphy, Melanie A.; Dyer, Rodney J.; Cushman, Samuel A.

doi:10.1002/9781118525258.ch10

Cited by 16 publications

(23 citation statements)

References 92 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Both the population topology and the selection of links will thus determine the configuration of a population network. Links in a population network must represent a process connecting nodes, and therefore, the appropriate set of links depends on the research question (Murphy, Dyer, & Cushman, ). Here, I specifically focus on the use of population networks to determine those links along which explanatory and response variables are calculated.…”

Section: Selection Of Links In Link‐based Analysesmentioning

confidence: 99%

“…This is an important consideration, as the choice of links can have a large effect on the results of link‐based landscape genetic analyses (Keller, Holderegger, & Van Strien, ; Naujokaitis‐Lewis, Rico, Lovell, Fortin, & Murphy, ). Although there are many different ways to select sets of links (Murphy et al., ), most link‐based landscape genetic studies simply calculate response and explanatory variables for links between all possible pairs of populations (Appendix ; but see Murphy, Dezzani, Pilliod, & Storfer, ; Angelone, Kienast, & Holderegger, ; Van Strien et al., ; Coster et al., ; Watts et al., ), which leads to a “saturated” population network (Figure a). However, the power of link‐based analyses in landscape genetics could be improved by using “pruned” networks (i.e., saturated networks from which links have been removed) opposed to saturated networks (Wagner & Fortin, ).…”

Section: Selection Of Links In Link‐based Analysesmentioning

confidence: 99%

“…For instance, with graph theoretical metrics (e.g., degree, betweenness centrality, clustering coefficient) nodes, links, or the network as a whole can be characterized (Barthélemy, ; Boccaletti, Latora, Moreno, Chavez, & Hwang, ). Such metrics can provide valuable ecological information (Murphy et al., ), such as estimates of the sensitivity of population networks to the removal of habitats (e.g., Garroway et al., ). Such metrics can also be used as measures of habitat connectivity for node‐based landscape genetic analysis (Koen, Bowman, & Wilson, ).…”

Section: Selection Of Links In Link‐based Analysesmentioning

confidence: 99%

See 2 more Smart Citations

Consequences of population topology for studying gene flow using link‐based landscape genetic methods

Strien

2017

Ecology and Evolution

View full text Add to dashboard Cite

Many landscape genetic studies aim to determine the effect of landscape on gene flow between populations. These studies frequently employ link-based methods that relate pairwise measures of historical gene flow to measures of the landscape and the geographical distance between populations. However, apart from landscape and distance, there is a third important factor that can influence historical gene flow, that is, population topology (i.e., the arrangement of populations throughout a landscape). As the population topology is determined in part by the landscape configuration, I argue that it should play a more prominent role in landscape genetics. K E Y W O R D Sdistance matrices, maximum dispersal distance, population networks

show abstract

Section: Selection Of Links In Link‐based Analysesmentioning

confidence: 99%

Section: Selection Of Links In Link‐based Analysesmentioning

confidence: 99%

Section: Selection Of Links In Link‐based Analysesmentioning

confidence: 99%

See 1 more Smart Citation

Consequences of population topology for studying gene flow using link‐based landscape genetic methods

Strien

2017

Ecology and Evolution

View full text Add to dashboard Cite

show abstract

“…These approaches require a priori stratification of samples into putative populations. Newer approaches like population graph approaches (Dyer, , ; Dyer & Nason, ; Murphy, Dyer, & Cushman, ) have been largely applied in population‐based frameworks, often where sampling locations, not genetically discrete populations, define the vertices of the graph. Individual‐based analyses in landscape genetics can help overcome problems with predefining populations, and many landscape genetic statistics can be adapted to individual‐based measures of genetic differentiation.…”

Section: Individually Based Landscape Genetics Modelsmentioning

confidence: 99%

Disentangling genetic structure for genetic monitoring of complex populations

Milligan

Archer

Ferchaud

et al. 2018

Evolutionary Applications

View full text Add to dashboard Cite

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.

show abstract

“…Here the graph is composed of vertices representing population distributions in a multilocus genetic space, and edges representing interdependencies between populations due, for example, to gene flow (Excoffier et al, 1992). The primary application to landscape genetics has been identification of conditional independence between populations to remove edges followed by analysis of graph structure metrics such as centrality or connnectness (Dyer, 2007;Murphy et al, 2016).…”

Section: Probabilistic Graph Modelsmentioning

confidence: 99%

Probabilistic graph models for landscape genetics

Milligan

2017

Preprint

View full text Add to dashboard Cite

Landscape genetics combines population genetics, landscape ecology, and spatial analysis to identify landscape and genetic factors that influence genetic and genomic variation. Progress in the field depends on a strong conceptual foundation and the means of identifying mechanistic connnections between environmental factors, landscape features, and genetic or genomic variation. Many existing approaches and much of the software commonly in use was developed for population genetics or statistics and is not entirely appropriate for landscape genetics. Probabilistic graph models provide a statistically rigorous and flexible means of constructing models directly applicable to landscape genetics. Probabilistic graph models also allow construction of mechanistic models, which are crucial elements in testing hypotheses. Sophisticated software exists for the analysis of graph models; however, much of it does not handle the types of data used for landscape genetics, model structures involving autoregressive spatial interaction between variables, or the scale of landscape genetics problems. Thus, an important priority for the field is to develop suitably flexible software tools for graph models that overcome these problems and allow landscape geneticists to explore meaningfully mechanistic and flexible models. We are developing such a library and applying it to examples in landscape genetics.Keywords: landscape genetics, population genetics, graph models, Bayesian inference, open source software, software development Landscape genetics combines population genetics, landscape ecology, and spatial analysis to 8 identify the mechanisms by which landscape and environmental factors influence genetic and ge-9 nomic variation. From the outset, the field has focused on the twin ecological and evolutionary 10

show abstract

Graph Theory and Network Models in Landscape Genetics

Cited by 16 publications

References 92 publications

Consequences of population topology for studying gene flow using link‐based landscape genetic methods

Consequences of population topology for studying gene flow using link‐based landscape genetic methods

Disentangling genetic structure for genetic monitoring of complex populations

Probabilistic graph models for landscape genetics

Contact Info

Product

Resources

About