Estimating recent migration and population-size surfaces

Abstract

In many species a fundamental feature of genetic diversity is that genetic similarity decays with geographic distance; however, this relationship is often complex, and may vary across space and time. Methods to uncover and visualize such relationships have widespread use for analyses in molecular ecology, conservation genetics, evolutionary genetics, and human genetics. While several frameworks exist, a promising approach is to infer maps of how migration rates vary across geographic space. Such maps could, in principle, be estimated across time to reveal the full complexity of population histories. Here, we take a step in this direction: we present a method to infer maps of population sizes and migration rates associated with different time periods from a matrix of genetic similarity between every pair of individuals. Specifically, genetic similarity is measured by counting the number of long segments of haplotype sharing (also known as identity-by-descent tracts). By varying the length of these segments we obtain parameter estimates associated with different time periods. Using simulations, we show that the method can reveal time-varying migration rates and population sizes, including changes that are not detectable when using a similar method that ignores haplotypic structure. We apply the method to a dataset of contemporary European individuals (POPRES), and provide an integrated analysis of recent population structure and growth over the last ∼3,000 years in Europe.

Dutch population structure across space, time and GWAS design

“…An IBD haplotype sharing matrix is conceptually similar to a ChromoPainter coancestry matrix 21 , but trades some sensitivity to be more explicitly interpretable. As IBD segment length is inversely related to age 22,23 , different length intervals can inform on structure at different time depths. Total pairwise IBD between Dutch individuals mirrored the structure observed with ChromoPainter (Figure 3a), with 8 distinct clusters identified in the IBD sharing matrix that broadly segregated with geography and recapitulated some of the important splits obtained from fineSTRUCTURE, most strikingly the west-east split in North Brabant.…”

“…To identify population structure captured by IBD sharing patterns we performed PCA on these matrices using the prcomp function in R version 3.2.3 34 and clustered the IBD matrices using a Gaussian mixture model implemented in the R package mclust 35 . We note that while previous work 21 has shown that IBD matrices underperform the linked ChromoPainter matrix in identifying population structure, they are arguably more interpretable for visualising temporal change as they can be subdivided into cM bins corresponding to different time periods, a feature leveraged by emerging work on local population structure 23 . Patterns in IBD sharing that identify population subgroups in older (shorter) cM bins which are preserved in more recent (longer) bins are interpreted as persistent population structure that has been influenced by mating patterns in old and recent generations.…”

Spatial Population Genetics: It's About Time

“…The lengths of segments of genome inherited in common by two individuals from t generations ago scales roughly with 1/t, these can carry substantial information about ancestors from only tens of generations ago. This fact has been used by Al-Asadi et al (2019) to create estimated maps of population density and migration rate for different, recent periods of time. However, the different strata cannot be so cleanly disentangled -e.g., although 1cM-long shared haplotypes are older on average than 2cM-long ones, the distributions of ages of these shared haplotypes overlap considerably.…”

“…Today's large genomic datasets are facilitating the estimation of the timing and extent of shared ancestry at a much finer geographic and temporal resolution than was previously possible (Li & Durbin, 2011, Palamara et al, 2012, Harris & Nielsen, 2013. And, advances in theory in continuous space (Felsenstein, 1975, Barton & Wilson, 1995, Barton et al, 2002, Hallatschek, 2011, Barton et al, 2013b, datasets of unprecedented geographical scale (e.g., Leslie et al, 2015, Aguillon et al, 2017, Shaffer et al, 2017, new computational tools for simulating spatial models (Haller & Messer, 2018, Haller et al, 2019, and new statistical paradigms for modeling those data (Petkova et al, 2016, Ringbauer et al, 2017, 2018, Bradburd et al, 2018, Al-Asadi et al, 2019 are together bringing an understanding of the geographic distribution of genetic variation into reach.…”

When are populations not connected like a circuit? Identifying biases in gene flow from coalescent times

“…;Excoffier, Dupanloup, Huerta-Sánchez, Sousa, & Foll, 2013), shared haplotype lengths (MAPS; Al-Asadi,Petkova, Stephens, & Novembre, 2019), differences in derived alleles (Psi;Peter & Slatkin, 2013), or directly model the temporally heterogeneous scenarios (X-origin; He, Prado, & Knowles, 2017) would be informative. Although resistance methodologies are robust in inferring departures from the IBD expectation in space, they do not provide biologically interpretable estimates of gene flow.…”