Networks of spatial genetic variation across species

Abstract

Spatial patterns of genetic variation provide information central to many ecological, evolutionary, and conservation questions. This spatial variability has traditionally been analyzed through summary statistics between pairs of populations, therefore missing the simultaneous influence of all populations. More recently, a network approach has been advocated to overcome these limitations. This network approach has been applied to a few cases limited to a single species at a time. The question remains whether similar patterns of spatial genetic variation and similar functional roles for specific patches are obtained for different species. Here we study the networks of genetic variation of four Mediterranean woody plant species inhabiting the same habitat patches in a highly fragmented forest mosaic in Southern Spain. Three of the four species show a similar pattern of genetic variation with well-defined modules or groups of patches holding genetically similar populations. These modules can be thought of as the long-sought-after, evolutionarily significant units or management units. The importance of each patch for the cohesion of the entire network, though, is quite different across species. This variation creates a tremendous challenge for the prioritization of patches to conserve the genetic variation of multispecies assemblages.complex networks | gene flow | habitat fragmentation | population genetics

Network modularity reveals critical scales for connectivity in ecology and evolution

“…1b), geographic distance is a key (yet implicit) component of modularity on spatial networks and may confound our understanding of the mechanisms generating such structure. All empirical examples of modularity in ecology and evolution have neglected this issue 15,16,31 , such that it currently remains unknown whether modularity is simply a by-product of distance effects on movement.…”

“…S1). Fortuna et al 15 argued that such modules represent fundamental scales for populations. Our results provide empirical support for this claim by illustrating that these scales can profoundly influence conclusions on connectivity and metapopulation viability.…”

“…Consequently, our understanding of scale has been largely driven by patch-based 10 (or island-based 11 ) or larger landscape (summarized across all locations or patches in a pre-defined region) paradigms 8 , neither of which are typically defined by the key process of movement. Integrating the concept of modularity from physics 12 , cellular and molecular biology 13 , and social sciences 14 has the potential to fill this conceptual gap and objectively identify critical scales for populations 15 .…”

“…In population biology, spatial modularity occurs where habitat patches (or local populations) are tightly connected to other patches through movement of individuals or their alleles but only weakly connected to the remaining patches in the landscape 15 . In doing so, spatial modularity provides a formal description of the functional aggregation of populations, identifies a potentially critical scale for ecological and evolutionary dynamics (for example, a relevant 'management unit' 9 ) and emphasizes the roles of patches in landscape connectivity based on movement within and between modules 15,16 .…”

“…In doing so, spatial modularity provides a formal description of the functional aggregation of populations, identifies a potentially critical scale for ecological and evolutionary dynamics (for example, a relevant 'management unit' 9 ) and emphasizes the roles of patches in landscape connectivity based on movement within and between modules 15,16 .…”