Combining genetic, historical and geographical data to reconstruct the dynamics of bioinvasions: application to the cane toad Bufo marinus

We developed a spatially explicit model of a bioinvasion and used an approximate Bayesian computation (ABC) framework to make various inferences from a combination of genetic (microsatellite genotypes), historical (first observation dates) and geographical (spatial coordinates of introduction and sampled sites) information. Our method aims to discriminate between alternative introduction scenarios and to estimate posterior densities of demographically relevant parameters of the invasive process. The performance of our landscape-ABC method is assessed using simulated data sets differing in their information content (genetic and/or historical data). We apply our methodology to the recent introduction and spatial expansion of the cane toad, Bufo marinus, in northern Australia. We find that, at least in the context of cane toad invasion, historical data are more informative than genetic data for discriminating between introduction scenarios. However, the combination of historical and genetic data provides the most accurate estimates of demographic parameters. For the cane toad, we find some evidence for a strong bottleneck prior to introduction, a small initial number of founder individuals (about 15), a large population growth rate (about 400% per generation), a standard deviation of dispersal distance of 19 km per generation and a high invasion speed at equilibrium (50 km per year). Our approach strengthens the application of the ABC method to the field of bioinvasion by allowing statistical inferences to be made on the introduction and the spatial expansion dynamics of invasive species using a combination of various relevant sources of information.     

Shifting dispersal modes at an expanding species’ range margin

While it is generally recognized that noncontiguous (long‐distance) dispersal of small numbers of individuals is important for range expansion over large geographic areas, it is often assumed that colonization on more local scales proceeds by population expansion and diffusion dispersal (larger numbers of individuals colonizing adjacent sites). There are few empirical studies of dispersal modes at the front of expanding ranges, and very little information is available on dispersal dynamics at smaller geographic scales where we expect contiguous (diffusion) dispersal to be prevalent. We used highly polymorphic genetic markers to characterize dispersal modes at a local geographic scale for populations at the edge of the range of a newly invasive grass species (Brachypodium sylvaticum) that is undergoing rapid range expansion in the Pacific Northwest of North America. Comparisons of Bayesian clustering of populations, patterns of genetic diversity, and gametic disequilibrium indicate that new populations are colonized ahead of the invasion front by noncontiguous dispersal from source populations, with admixture occurring as populations age. This pattern of noncontiguous colonization was maintained even at a local scale. Absence of evidence for dispersal among adjacent pioneer sites at the edge of the expanding range of this species suggests that pioneer populations undergo an establishment phase during which they do not contribute emigrants for colonization of neighbouring sites. Our data indicate that dispersal modes change as the invasion matures: initial colonization processes appear to be dominated by noncontiguous dispersal from only a few sources, while contiguous dispersal may play a greater role once populations become established.     

msABC: a modification of Hudson's ms to facilitate multi-locus ABC analysis

With the availability of whole-genome sequence data biologists are able to test hypotheses regarding the demography of populations. Furthermore, the advancement of the Approximate Bayesian Computation (ABC) methodology allows the demographic inference to be performed in a simple framework using summary statistics. We present here msABC, a coalescent-based software that facilitates the simulation of multi-locus data, suitable for an ABC analysis. msABC is based on Hudson's ms algorithm, which is used extensively for simulating neutral demographic histories of populations. The flexibility of the original algorithm has been extended so that sample size may vary among loci, missing data can be incorporated in simulations and calculations, and a multitude of summary statistics for single or multiple populations is generated. The source code of msABC is available at http://bio.lmu.de/~pavlidis/msabc or upon request from the authors.     

Contrasting effects of long distance seed dispersal on genetic diversity during range expansion

Currently many attempts are made to reconstruct the colonization history of plant species after the last ice age. A surprising finding is that during the colonization phase genetic diversity did not decrease as much as expected. In this paper we examine whether long distance seed dispersal events could play a role in the unexpected maintenance of genetic diversity during range expansion. This study is based on simulations carried out with a maternally inherited haploid locus using a cellular automaton. The simulations reveal a close relationship between the frequency of long distance seed dispersal events and the amount of genetic diversity preserved during colonization. In particular, when the colonized region is narrow, a complete loss of genetic diversity results from the occurrence of very rare long distance dispersal (LDD) events. We call this phenomenon the 'embolism effect'. However, slightly higher rates of LDD events reverse this effect, up to the point that diversity is better preserved than in a pure diffusion model. This phenomenon is linked to the reorganization of the genetic structure during colonization and is called the 'reshuffling effect'.     

Extreme long-distance dispersal of the lowland tropical rainforest tree Ceiba pentandra L. (Malvaceae) in Africa and the Neotropics

Many tropical tree species occupy continental expanses of rainforest and flank dispersal barriers such as oceans and mountains. The role of long-distance dispersal in establishing the range of such species is poorly understood. In this study, we test vicariance hypotheses for range disjunctions in the rainforest tree Ceiba pentandra, which is naturally widespread across equatorial Africa and the Neotropics. Approximate molecular clocks were applied to nuclear ribosomal [ITS (internal transcribed spacer)] and chloroplast (psbB-psbF) spacer DNA sampled from 12 Neotropical and five West African populations. The ITS (N=5) and psbB-psbF (N=2) haplotypes exhibited few nucleotide differences, and ITS and psbB-psbF haplotypes were shared by populations on both continents. The low levels of nucleotide divergence falsify vicariance explanations for transatlantic and cross-Andean range disjunctions. The study shows how extreme long-distance dispersal, via wind or marine currents, creates taxonomic similarities in the plant communities of Africa and the Neotropics.     

Pollen flow in the wildservice tree, Sorbus torminalis (L.) Crantz. IV. Whole interindividual variance of male fecundity estimated jointly with the dispersal kernel

Interindividual variance of male reproductive success (MRS) contributes to genetic drift, which in turn interacts with selection and migration to determine the short-term response of populations to rapid changes in their environment. Individual relative MRS can be estimated through paternity analysis and can be further dissected into fecundity and spatial components. Existing methods to achieve this decomposition either rely on the strong assumption of a random distribution of pollen donors (TwoGener) or estimate only the part of the variance of male fecundity that is explained by few covariates. We developed here a method to estimate jointly the whole variance of male fecundity and the pollen dispersal curve from the genotypic information of sampled seeds and their putative fathers and geographical information of all individuals in the study area. We modelled the relative individual fecundities as a log-normally distributed random effect. We used a Bayesian approach, well suited to the hierarchical nature of the model, to estimate these fecundities. When applied to Sorbus torminalis , the estimated variance of male fecundity corresponded to an effective density of trees 13 times lower than the observed density ( d_obs / d_ep ~  13). This value is between the value ( ~ 2) estimated with a classical mating model including three covariates (neighbourhood density, diameter, flowering intensity) that affect fecundity and the value ( ~ 30) estimated with TwoGener. The estimated dispersal kernel was close to previous results. This approach allows fine monitoring of ongoing genetic drift in natural populations, and quantitative dissection of the processes contributing to drift, including human actions.     

Dispersal strategies, secondary range expansion and invasion genetics of the nonindigenous round goby, Neogobius melanostomus, in Great Lakes tributaries

Dispersal strategies are important mechanisms underlying the spatial distribution and colonizing ability of all mobile species. In the current study, we use highly polymorphic microsatellite markers to evaluate local dispersal and colonization dynamics of the round goby (Neogobius melanostomus), an aquatic invader expanding its range from lake to river environments in its introduced North American range. Genetic structure, genotype assignment and genetic diversity were compared among 1262 round gobies from 20 river and four lake sites in three Great Lakes tributaries. Our results indicate that a combination of short-distance diffusion and long-distance dispersal, collectively referred to as 'stratified dispersal', is facilitating river colonization. Colonization proceeded upstream yearly (approximately 500 m/year; 2005-2009) in one of two temporal replicates while genetic structure was temporally stable. Contiguous dispersal from the lake was observed in all three rivers with a substantial portion of river fish (7.3%) identified as migrants. Genotype assignment indicated a separate introduction occurred upstream of the invasion front in one river. Genetic diversity was similar and relatively high among lake and recently colonized river populations, indicating that founder effects are mitigated through a dual-dispersal strategy. The remarkable success of round goby as an aquatic invader stresses the need for better diffusion models of secondary range expansion for presumably sessile invasive species.     

Congruent population structure inferred from dispersal behaviour and intensive genetic surveys of the threatened Florida scrub-jay (Aphelocoma coerulescens)

The delimitation of populations, defined as groups of individuals linked by gene flow, is possible by the analysis of genetic markers and also by spatial models based on dispersal probabilities across a landscape. We combined these two complimentary methods to define the spatial pattern of genetic structure among remaining populations of the threatened Florida scrub-jay, a species for which dispersal ability is unusually well-characterized. The range-wide population was intensively censused in the 1990s, and a metapopulation model defined population boundaries based on predicted dispersal-mediated demographic connectivity. We subjected genotypes from more than 1000 individual jays screened at 20 microsatellite loci to two Bayesian clustering methods. We describe a consensus method for identifying common features across many replicated clustering runs. Ten genetically differentiated groups exist across the present-day range of the Florida scrub-jay. These groups are largely consistent with the dispersal-defined metapopulations, which assume very limited dispersal ability. Some genetic groups comprise more than one metapopulation, likely because these genetically similar metapopulations were sundered only recently by habitat alteration. The combined reconstructions of population structure based on genetics and dispersal-mediated demographic connectivity provide a robust depiction of the current genetic and demographic organization of this species, reflecting past and present levels of dispersal among occupied habitat patches. The differentiation of populations into 10 genetic groups adds urgency to management efforts aimed at preserving what remains of genetic variation in this dwindling species, by maintaining viable populations of all genetically differentiated and geographically isolated populations.     

Detecting dispersal of Nuphar lutea in river corridors using microsatellite markers

1. River corridors are well-known for their role in plant dispersal. The buoyancy of seeds, the possibility of dispersal by vegetative fragments, and the frequency and efficiency of dispersal among different river catchments determine linear distribution patterns. Little is known about the relative importance of these factors to observed patterns of genetic variation.
2. One hundred and fifty-six Nuphar lutea individuals from forty-four sampling sites in the river catchments of the Cidlina River, the Mrlina River and the Labe River (Czech Republic) were studied using ten microsatellite markers. Interpretation of patterns in genetic variation allowed several conclusions about dispersal mechanisms.
3. Vegetative long-distance dispersal is probably very limited in this species. Only one multilocus genotype was found in more than one sampling site. The distance between the sites was about 75 km.
4. To explain the distribution of Bayesian based clusters of related multilocus genotypes, both along-river and inter-river long-distance dispersals have to be invoked.
5. A marginally significant tendency for higher genetic diversity in the lower part of the river Cidlina was detected. Continuous downstream dispersal of seeds by water currents could be a valid explanation.
6. Significant positive autocorrelation was found among individuals at within-river distances of up to 25 km. Repeated dispersal of seeds over distances in the range of tens of kilometers is common.     

A simulation model of plant invasion: long-distance dispersal determines the pattern of spread

Mechanisms and consequences of biological invasions are a global issue. Yet, one of the key aspects, the initial phase of invasion, is rarely observed in detail. Data from aerial photographs covering the spread of Heracleum mantegazzianum (Apiaceae, native to Caucasus) on a local scale of hectares in the Czech Republic from the beginning of invasion were used as an input for an individual-based model (IBM), based on small-scale and short-time data. To capture the population development inferred from the photographs, long-distance seed dispersal, changes in landscape structures and suitability of landscape elements to invasion by H. mantegazzianum were implemented in the model. The model was used to address (1) the role of long-distance dispersal in regional invasion dynamics, and (2) the effect of land-use changes on the progress of the invasion. Simulations showed that already small fractions of seed subjected to long-distance dispersal, as determined by systematic comparison of field data and modelling results, had an over-proportional effect on the spread of this species. The effect of land-use changes on the simulated course of invasion depends on the actual level of habitat saturation; it is larger for populations covering a high proportion of available habitat area than for those in the initial phase of invasion. Our results indicate how empirical field data and model outputs can be linked more closely with each other to improve the understanding of invasion dynamics. The multi-level, but nevertheless simple structure of our model suggests that it can be used for studying the spread of similar species invading in comparable landscapes.     

Ant-mediated expansion of an obligate seeder species during the first years after fire

Most obligate seeder species build up a soil seed bank that is associated with massive seed germination in the year immediately after a fire. These species are also shade‐intolerant and disappear when vegetation cover closes, creating unsuitable conditions for seedling recruitment. The only way for these plants to expand their populations is when habitats suitable for seedling recruitment arise (i.e. in years immediately after a fire). However, short primary seed dispersal of obligate seeders does not allow these plants to colonise the suitable habitats, and these habitats can only be colonised by secondary seed dispersion. We hypothesised that Fumana ericoides, an obligate‐seeding small shrub, not only establishes abundantly in the first year after fire, but also expands its local range in the following years due to secondary dispersal by ants while suitable habitats are still available. We tested this hypothesis using experimental studies and a simulation model of potential population expansion in a recently burned area. Results showed that F. ericoides not only established prolifically in the year immediately after fire, but was also able to recruit new individuals and expand its population in the years following the fire, despite a low germination rate and short primary seed dispersal. Ant‐mediated seed dispersal and availability of suitable habitats were key factors in this phenomenon: ants redistributed seeds in suitable habitats while they were available, which accelerated the expansion of F. ericoides because new plants established far away from the core population.     

Geostatistical genetic analysis for inferring the dispersal pattern of a partially clonal species: example of the chestnut blight fungus

Spatial genetic analyses can be used to infer dispersal processes in natural populations. For partially clonal species with alternating sexual and asexual reproduction, the repetition of genotypes must be taken into account in analyses. The methods currently employed to evaluate the relevance of the spatial scale used for the estimation of gene flow are not suitable for these species. We investigated recently developed methods for taking into account repeated genotypes and for determining whether the sampling scale is large enough to capture all the spatial genetic structure existing within a population. We applied these methods to a fungal plant pathogen species, Cryphonectria parasitica, which has caused the death of many American and European chestnut trees since its introduction from Asia at the beginning of the 20th century. These methods were found to be useful for unravelling the effects of clonality and historical gene flow on the spatial genetic structure, and indicated that dispersal processes have probably occurred over a larger spatial scale than previously assumed.     

Phylogeography of the Rhabditis (Pellioditis) marina species complex: evidence for long-distance dispersal, and for range expansions and restricted gene flow in the northeast Atlantic

Pinpointing processes that structure the geographical distribution of genetic diversity of marine species and lead to speciation is challenging because of the lack of obvious dispersal barriers and the likelihood of substantial (passive) dispersal in oceans. In addition, cryptic radiations with sympatric distributions abound in marine species, challenging the allopatric speciation mechanism. Here, we present a phylogeographical study of the marine nematode species complex Rhabditis ( Pellioditis ) marina to investigate processes shaping genetic structure and speciation. Rhabditis ( P .) marina lives on decaying macroalgae in the intertidal, and may therefore disperse over considerable distances. Rhabditis ( P .) marina consists of several cryptic species sympatrically distributed at a local scale. Genetic variation in the COI gene was screened in 1362 specimens from 45 locations around the world. Two nuclear DNA genes (ITS and D2D3) were sequenced to infer phylogenetic species. We found evidence for ten sympatrically distributed cryptic species, seven of which show a strong genetic structuring. A historical signature showed evidence for restricted gene flow with occasional long-distance dispersal and range expansions pre-dating the last glacial maximum. Our data also point to a genetic break around the British Isles and a contact zone in the Southern Bight of the North Sea. We provide evidence for the transoceanic distribution of at least one cryptic species (PmIII) and discuss the dispersal capacity of marine nematodes. The allopatric distribution of some intraspecific phylogroups and of closely related cryptic species points to the potential for allopatric speciation in R. ( P .) marina .     

Long distance dispersal and the fate of a gene from the colonization front

There is an increasing recognition that long distance dispersal (LDD) plays a key role in establishing spatial genetic structure during colonization. Recent works, focused on short distance dispersal, demonstrated that a neutral mutation arising at the colonization front can either ‘surf’ with the wave front and reach high frequencies or stay near its place of origin at low frequencies. Here, we examine how LDD, and more generally the shape of the dispersal kernel, modifies this phenomenon and how the width of the colonization corridor affects the fate of the mutation. We demonstrate that when LDD events are more frequent, the ‘surfing phenomenon’ is less frequent, probably because any alleles can get far ahead from the colonization front and preclude the invasion by others alleles, thus leading to an attenuation of the diversity loss. We also demonstrate that the width of the colonization corridor influences the fate of the mutation, wide spaces decreasing the probability of invasion. Overall, the genetic structure of diversity resulted not only from LDD but also particularly from the shape of the dispersal kernel.     

From the Apennines to the Alps: colonization genetics of the naturally expanding Italian wolf (Canis lupus) population

Wolves in Italy strongly declined in the past and were confined south of the Alps since the turn of the last century, reduced in the 1970s to approximately 100 individuals surviving in two fragmented subpopulations in the central-southern Apennines. The Italian wolves are presently expanding in the Apennines, and started to recolonize the western Alps in Italy, France and Switzerland about 16 years ago. In this study, we used a population genetic approach to elucidate some aspects of the wolf recolonization process. DNA extracted from 3068 tissue and scat samples collected in the Apennines (the source populations) and in the Alps (the colony), were genotyped at 12 microsatellite loci aiming to assess (i) the strength of the bottleneck and founder effects during the onset of colonization; (ii) the rates of gene flow between source and colony; and (iii) the minimum number of colonizers that are needed to explain the genetic variability observed in the colony. We identified a total of 435 distinct wolf genotypes, which showed that wolves in the Alps: (i) have significantly lower genetic diversity (heterozygosity, allelic richness, number of private alleles) than wolves in the Apennines; (ii) are genetically distinct using pairwise F(ST) values, population assignment test and Bayesian clustering; (iii) are not in genetic equilibrium (significant bottleneck test). Spatial autocorrelations are significant among samples separated up to c. 230 km, roughly correspondent to the apparent gap in permanent wolf presence between the Alps and north Apennines. The estimated number of first-generation migrants indicates that migration has been unidirectional and male-biased, from the Apennines to the Alps, and that wolves in southern Italy did not contribute to the Alpine population. These results suggest that: (i) the Alps were colonized by a few long-range migrating wolves originating in the north Apennine subpopulation; (ii) during the colonization process there has been a moderate bottleneck; and (iii) gene flow between sources and colonies was moderate (corresponding to 1.25-2.50 wolves per generation), despite high potential for dispersal. Bottleneck simulations showed that a total of c. 8-16 effective founders are needed to explain the genetic diversity observed in the Alps. Levels of genetic diversity in the expanding Alpine wolf population, and the permanence of genetic structuring, will depend on the future rates of gene flow among distinct wolf subpopulation fragments.