Making use of the social network in conservation genomics: Integrating kinship and network analyses to understand connectivity

Abstract

Inferring and quantifying recent barriers to connectivity is increasingly important for conservation and management in a world undergoing rapid environmental change. Traditional measures of genetic differentiation can take many generations to reflect a new barrier to connectivity. Although methods that use the linkage disequilibrium signal in mixed genetic samples are able to reflect recent levels of gene flow, they are not suitable for use in situations with low levels of genetic differentiation. Kinship‐based methods, those that assess the spatio‐temporal distribution of related individuals, have been used in this context, but a formal statistical framework for such approaches has been lacking. In this issue of Molecular Ecology Resources, Escoda, et al. adapt the assortativity coefficient, AC, to analyse the networks of kin relationships in the Pyrenean desman (Galemys pyrenaicus) across potential barriers to dispersal. Their modified AC quantifies the proportion of missing kin relationships across putative dispersal barriers with respect to the expected proportion if there was no barrier. This application highlights that AC can be used to test the null hypothesis that a putative barrier has no effect on gene flow, in which case AC is not significantly different from 0. The method represents a useful step forward in conservation genomics by developing and adapting tools to assess contemporary connectivity using genomic data.

Pedigree‐based assessment of recent population connectivity in a threatened rattlesnake

“…Recent advances have been made to infer the probable location of parents based on the location of full‐sibling pairs, but such pairs can be rare in many data sets (Kormann et al, 2012). Others have used extended kin pairs as qualitative evidence of connectivity, or to detect the presence of low‐permeability barriers (Escoda et al, 2017), but do not explicitly quantify levels of dispersal (Carroll & Gaggiotti, 2019; Vandergast et al, 2019). Here, we have shown that it is possible to use both close and distant kin relations to generate quantitative estimates of dispersal rates.…”

“…used extended kin pairs as qualitative evidence of connectivity, or to detect the presence of low-permeability barriers (Escoda et al, 2017), but do not explicitly quantify levels of dispersal (Carroll & Gaggiotti, 2019;Vandergast et al, 2019). Here, we have shown that it is possible to use both close and distant kin relations to generate quantitative estimates of dispersal rates.…”