Controlling for non-independence in comparative analysis of patterns across populations within species.

摘要: How do we quantify patterns (such as responses to local selection) sampled across multiple populations within a single species? Key this question is the extent which species represent statistically independent data points in our analysis. Comparative analyses and higher taxa have long recognized need control for non-independence of that arises through shared common ancestry among them (phylogenetic non-independence), quantitative genetic studies individuals linked by pedigree. Analyses lacking pedigree information fall middle, not only deal with ancestry, but also impact exchange migrants between (gene flow). As result, phenotypes measured one population are influenced processes acting on others, may be good guide either strength or direction selection. Although many examine species, few consider such non-independence. Here, discuss sources comparative analysis, show why phylogeny-based approaches widely used cross-species unlikely useful species. We outline (intraspecific contrasts, generalized least squares, linear mixed models autoregression) been context, explain their specific assumptions. highlight power ‘mixed models’ contexts where problems arise, these allow incorporation both gene flow. suggest what can done when ideal solutions inaccessible, wider range intraspecific methods call simulation error rates associated alternative approaches.