Understanding Images: How dispersal shapes spatial patterns of genetic diversity

Authors: Stepfanie M. Aguillon and Nancy Chen.

Competing Interests: Stepfanie M. Aguillon and Nancy Chen are authors of the article discussed in this blog.

Image Caption: This image shows an adult Florida Scrub-Jay from the study population.

Image Credit: Reed Bowman, Archbold Biological Station, Venus, Florida, United States of America.

The movement of organisms across the landscape (dispersal) can influence various biological processes—from individual social interactions to species’ ranges. Dispersal is particularly important in creating patterns of spatial genetic structure, such as isolation-by-distance, whereby individuals that are geographically close are often more closely related genetically. A considerable amount of theory exists on isolation-by-distance, starting with Sewall Wright and Gustave Malécot in the 1940s, with the basic idea being that genetic drift can cause allele frequencies to diverge faster than dispersal can homogenize them among geographically distant populations. Isolation-by-distance is increasingly taken into account to ensure that it is not mistaken for other processes (e.g., local adaptation). Even though isolation-by-distance is frequently observed in nature, we have few empirical systems that can directly demonstrate how limited dispersal creates this pattern.

A unique bird provides an opportunity to better understand isolation-by-distance

In the featured article from the August issue of PLOS Genetics, we capitalize on the unique biology and long-term monitoring efforts of the Florida Scrub-Jay (Aphelocoma coerulescens) to explore how isolation-by-distance patterns are generated [1]. Florida Scrub-Jays are long-lived birds that disperse extremely short distances from their site of birth to their breeding site and then rarely move once established on a breeding territory.

A population at Archbold Biological Station in Venus, Florida, USA has been monitored intensively since 1969, with all birds marked with uniquely identifying colored leg bands (as shown in the featured image). We thus have a wealth of data on dispersal behavior and familial relationships for hundreds of individuals in our study population. It is incredibly difficult to accurately measure dispersal distances in the wild, so having good direct estimates of dispersal in the Florida Scrub-Jay make it an especially useful system for studying isolation-by-distance.

Moreover, to better understand the genomic consequences of population decline in these birds, in a separate study, Nancy Chen and colleagues have previously genotyped nearly every individual in the population over the past two decades [2]. The unique combination of pedigree, dispersal, and genomic data available for this population make it especially suitable for studying how dispersal shapes patterns of spatial genetic structure, and the extremely limited dispersal in these birds means we can detect isolation-by-distance on an easily observed spatial and temporal scale.

Patterns of isolation-by-distance in the Florida Scrub-Jay

In our study, we detected isolation-by-distance between individuals within the population at Archbold at an extremely small spatial scale (<10 km). We assessed isolation-by-distance separately in autosomal (not on a sex chromosome) and sex-linked markers by comparing genetic relatedness and geographic distance between all possible pairs of individuals in the population. Though dispersal is limited overall, we know from field observations that females tend to disperse farther than males—a trend typical for birds—and we saw a stronger signal of isolation-by-distance in males, which is consistent with female-biased dispersal. Differences in dispersal could also lead to stronger patterns on the Z sex chromosome, as the Z has a smaller effective population size and spends more time in males compared to the autosomes—in birds, males are the homogametic (ZZ) sex. Though we did not find evidence for more isolation-by-distance on the Z, the high levels of immigration into the population may have obscured the pattern, and we had lower statistical power on the Z (as we had fewer Z-linked markers).

Since we have a fairly complete pedigree for the Archbold population, we could further break down the pattern of isolation-by-distance by genealogical relationship. We found that close genealogical relatives—such as parent-offspring pairs and full-siblings—tend to breed closer together geographically. Though these close relatives drive the strength of the relationship at short distances, we still find isolation-by-distance even when they are removed from the dataset—suggesting that more distantly related individuals do contribute to the pattern.

A fairly complete understanding of how dispersal shapes patterns of spatial genetic structure

We took one additional step and performed simulations that could recapitulate the observed isolation-by-distance patterns. First, using only the dispersal curves, we were able to generate the observed distribution of distances between pairs of related individuals of all possible sex combinations (out to second cousins), thus demonstrating that limited dispersal alone could result in close genealogical relatives living close together geographically. We then incorporated even more information into simulations of genetic ancestry—including sex-specific parameters for dispersal, relatedness, and immigration, as well as isolation-by-distance in immigrants—to show that the expected patterns of isolation-by-distance match those that we observe in the population. Interestingly, we were able to recapitulate the observed isolation-by-distance patterns over a short timescale (<10 generations). Thus, we show how limited dispersal can create patterns of isolation-by-distance at a small spatial scale over just a few generations. Additional studies using blocks of identity-by-descent—segments of the genome that are shared through a common ancestor—will help us to further understand how dispersal shapes the movement of genomic information across the landscape.

References

1. Aguillon SM, Fitzpatrick JW, Bowman R, Schoech SJ, Clark AG, Coop G, Chen N (2017) Deconstructing isolation-by-distance: The genomic consequences of limited dispersal. PLoS Genet 13(8): e1006911. https://doi.org/10.1371/journal.pgen.1006911

2. Chen N, Cosgrove EJ, Bowman R, Fitzpatrick JW, Clark AG (2016) Genomic Consequences of Population Decline in the Endangered Florida Scrub-Jay. Curr Biol. 26(21): 2974-2979. doi: 10.1016/j.cub.2016.08.062