Explore the vast range of titles available.

Founder-event speciation, where a rare jump dispersal event founds a new genetically isolated lineage, has long been considered crucial by many historical biogeographers, but its importance is disputed within the vicariance school. Probabilistic modeling of geographic range evolution creates the potential to test different biogeographical models against data using standard statistical model choice procedures, as long as multiple models are available. I re-implement the Dispersal-Extinction-Cladogenesis (DEC) model of LAGRANGE in the R package BioGeoBEARS, and modify it to create a new model, DEC+J, which adds founder-event speciation, the importance of which is governed by a new free parameter, j. The identifiability of DEC and DEC+J is tested on data sets simulated under a wide range of macroevolutionary models where geography evolves jointly with lineage birth/death events. The results confirm that DEC and DEC+J are identifiable even though these models ignore the fact that molecular phylogenies are missing many cladogenesis and extinction events. The simulations also indicate that DEC will have substantially increased errors in ancestral range estimation and parameter inference when the true model includes +J. DEC and DEC+J are compared on 13 empirical data sets drawn from studies of island clades. Likelihood-ratio tests indicate that all clades reject DEC, and AICc model weights show large to overwhelming support for DEC+J, for the first time verifying the importance of founder-event speciation in island clades via statistical model choice. Under DEC+J, ancestral nodes are usually estimated to have ranges occupying only one island, rather than the widespread ancestors often favored by DEC. These results indicate that the assumptions of historical biogeography models can have large impacts on inference and require testing and comparison with statistical methods.

The ENMTools software package was introduced in 2008 as a platform for making measurements on environmental niche models (ENMs, frequently referred to as species distribution models or SDMs), and for using those measurements in the context of newly developed Monte Carlo tests to evaluate hypotheses regarding niche evolution. Additional functionality was later added for model selection and simulation from ENMs, and the software package has been quite widely used. ENMTools was initially implemented as a Perl script, which was also compiled into an executable file for various platforms. However, the package had a number of significant limitations; it was only designed to fit models using Maxent, it relied on a specific Perl distribution to function, and its internal structure made it difficult to maintain and expand. Subsequently, the R programming language became the platform of choice for most ENM studies, making ENMTools less usable for many practitioners. Here we introduce a new R version of ENMTools that implements much of the functionality of its predecessor as well as numerous additions that simplify the construction, comparison and evaluation of niche models. These additions include new metrics for model fit, methods of measuring ENM overlap, and methods for testing evolutionary hypotheses. The new version of ENMTools is also designed to work within the expanding universe of R tools for ecological biogeography, and as such includes greatly simplified interfaces for analyses from several other R packages.

The ability of lineages to disperse long distances over evolutionary timescales may be influenced by the gain or loss of traits adapted to enhance local, ecological dispersal. For example, some species in the southern conifer family Podocarpaceae have fleshy cones that encourage bird dispersal, but it is unknown how this trait has influenced the clade’s historical biogeography, or its importance compared with other predictors of dispersal such as the geographic distance between regions. We answer these questions quantitatively by using a dated phylogeny of 197 species of southern conifers (Podocarpaceae and their sister family Araucariaceae) to statistically compare standard, trait-independent biogeography models with new BioGeoBEARS models where an evolving trait can influence dispersal probability, and trait history, biogeographical history, and model parameters are jointly inferred. We validate the method with simulation-inference experiments. Comparing all models, those that include trait-dependent dispersal accrue 87.5% of the corrected Akaike Information Criterion (AICc) model weight. Averaged across all models, lineages with nonfleshy cones had a dispersal probability multiplier of 0.49 compared with lineages with fleshy cones. Distance is included as a predictor of dispersal in all credible models (100% model weight). However, models with changing geography earned only 22.0% of the model weight, and models submerging New Caledonia/New Zealand earned only 0.01%. The importance of traits and distance suggests that long-distance dispersal over macroevolutionary timespans should not be thought of as a highly unpredictable chance event. Instead, long-distance dispersal can be modeled, allowing statistical model comparison to quantify support for different hypotheses.

Historical biogeography is increasingly studied from an explicitly statistical perspective, using stochastic models to describe the evolution of species range as a continuous-time Markov process of dispersal between and extinction within a set of discrete geographic areas. The main constraint of these methods is the computational limit on the number of areas that can be specified. We propose a Bayesian approach for inferring biogeographic history that extends the application of biogeographic models to the analysis of more realistic problems that involve a large number of areas. Our solution is based on a \"data-augmentation\" approach, in which we first populate the tree with a history of biogeographic events that is consistent with the observed species ranges at the tips of the tree. We then calculate the likelihood of a given history by adopting a mechanistic interpretation of the instantaneous-rate matrix, which specifies both the exponential waiting times between biogeographic events and the relative probabilities of each biogeographic change. We develop this approach in a Bayesian framework, marginalizing over all possible biogeographic histories using Markov chain Monte Carlo (MCMC). Besides dramatically increasing the number of areas that can be accommodated in a biogeographic analysis, our method allows the parameters of a given biogeographic model to be estimated and different biogeographic models to be objectively compared. Our approach is implemented in the program, BayArea.

Aim The tomato family Solanaceae is distributed on all major continents except Antarctica and has its centre of diversity in South America. Its worldwide distribution suggests multiple long-distance dispersals within and between the New and Old Worlds. Here, we apply maximum likelihood (ML) methods and newly developed biogeographical stochastic mapping (BSM) to infer the ancestral range of the family and to estimate the frequency of dispersal and vicariance events resulting in its present-day distribution. Location Worldwide. Methods Building on a recently inferred megaphylogeny of Solanaceae, we conducted ML model fitting of a range of biogeographical models with the program 'BioGeoBEARS'. We used the parameters from the best fitting model to estimate ancestral range probabilities and conduct stochastic mapping, from which we estimated the number and type of biogeographical events. Results Our best model supported South America as the ancestral area for the Solanaceae and its major clades. The BSM analyses showed that dispersal events, particularly range expansions, are the principal mode by which members of the family have spread beyond South America. Main conclusions For Solanaceae, South America is not only the family's current centre of diversity but also its ancestral range, and dispersal was the principal driver of range evolution. The most common dispersal patterns involved range expansions from South America into North and Central America, while dispersal in the reverse direction was less common. This directionality may be due to the early build-up of species richness in South America, resulting in large pool of potential migrants. These results demonstrate the utility of BSM not only for estimating ancestral ranges but also in inferring the frequency, direction and timing of biogeographical events in a statistically rigorous framework.

Aim Species distribution models are used across evolution, ecology, conservation and epidemiology to make critical decisions and study biological phenomena, often in cases where experimental approaches are intractable. Choices regarding optimal models, methods and data are typically made based on discrimination accuracy: a model's ability to predict subsets of species occurrence data that were withheld during model construction. However, empirical applications of these models often involve making biological inferences based on continuous estimates of relative habitat suitability as a function of environmental predictor variables. We term the reliability of these biological inferences ‘functional accuracy.’ We explore the link between discrimination accuracy and functional accuracy. Methods Using a simulation approach we investigate whether models that make good predictions of species distributions correctly infer the underlying relationship between environmental predictors and the suitability of habitat. Results We demonstrate that discrimination accuracy is only informative when models are simple and similar in structure to the true niche, or when data partitioning is geographically structured. However, the utility of discrimination accuracy for selecting models with high functional accuracy was low in all cases. Main conclusions These results suggest that many empirical studies and decisions are based on criteria that are unrelated to models’ usefulness for their intended purpose. We argue that empirical modelling studies need to place significantly more emphasis on biological insight into the plausibility of models, and that the current approach of maximizing discrimination accuracy at the expense of other considerations is detrimental to both the empirical and methodological literature in this active field. Finally, we argue that future development of the field must include an increased emphasis on simulation; methodological studies based on ability to predict withheld occurrence data may be largely uninformative about best practices for applications where interpretation of models relies on estimating ecological processes, and will unduly penalize more biologically informative modelling approaches.

Chloroplasts and mitochondria descended from bacterial ancestors, but the dating of these primary endosymbiosis events remains very uncertain, despite their importance for our understanding of the evolution of both bacteria and eukaryotes. All phylogenetic dating in the Proterozoic and before is difficult: Significant debates surround potential fossil calibration points based on the interpretation of the Precambrian microbial fossil record, and strict molecular clock methods cannot be expected to yield accurate dates over such vast timescales because of strong heterogeneity in rates. Even with more sophisticated relaxed-clock analyses, nodes that are distant from fossil calibrations will have a very high uncertainty in dating. However, endosymbiosis events and gene duplications provide some additional information that has never been exploited in dating; namely, that certain nodes on a gene tree must represent the same events, and thus must have the same or very similar dates, even if the exact date is uncertain. We devised techniques to exploit this information: cross-calibration, in which node date calibrations are reused across a phytogeny, and cross-bracing, in which node date calibrations are formally linked in a hierarchical Bayesian model. We apply these methods to proteins with ancient duplications that have remained associated and originated from plastid and mitochondrial endosymbionts: the α and β subunits of ATP synthase and its relatives, and the elongation factor thermo unstable. The methods yield reductions in dating uncertainty of 14-26% while only using date calibrations derived from phylogenetically unambiguous Phanerozoic fossils of multicellular plants and animals. Our results suggest that primary plastid endosymbiosis occurred ~900 Mya and mitochondrial endosymbiosis occurred ~1,200 Mya.

The Andean mountains of South America are the most species-rich biodiversity hotspot worldwide with c. 15% of the world's plant species, in only 1% of the world's land surface. Orchids are a key element of the Andean flora, and one of the most prominent components of the Neotropical epiphyte diversity, yet very little is known about their origin and diversification. We address this knowledge gap by inferring the biogeographical history and diversification dynamics of the two largest Neotropical orchid groups (Cymbidieae and Pleurothallidinae), using two unparalleled, densely sampled orchid phylogenies (including more than 400 newly generated DNA sequences), comparative phylogenetic methods, geological and biological datasets. We find that the majority of Andean orchid lineages only originated in the last 20-15 million yr. Andean lineages are derived from lowland Amazonian ancestors, with additional contributions from Central America and the Antilles. Species diversification is correlated with Andean orogeny, and multiple migrations and recolonizations across the Andes indicate that mountains do not constrain orchid dispersal over long timescales. Our study sheds new light on the timing and geography of a major Neotropical diversification, and suggests that mountain uplift promotes species diversification across all elevational zones.

Aim: To examine the role of geological history, connectivity and distance in shaping the biogeographical structure of North American desert clades that are restricted to habitat islands (sand dunes and relictual aquatic habitats), using statistical model choice on old and new probabilistic biogeographical models. Location: North America, Mojave, Sonoran and Chihuahuan Deserts. Materials and methods: Dated phytogenies were estimated for three fieldsampled insect clades (Trigonoscuta, Rhaphiomidas and sand treader crickets), and five other literature-sampled clades (the snails Assiminea, Pyrgulopsis and Tryonia; the desert fringe-toed lizard Uma; and the desert pupfish Cyprinodon).BioGeoBEARS was used to statistically compare biogeographical models assuming unconstrained or connectivity-constrained dispersal, with or without founder-event speciation (jump dispersal) permitted. Finally, we introduce and test a novel distance-based dispersal model (+x) where dispersal probability is multiplied by distance to the power x. Results: We observed little concordance between biogeographical patterns and timing of geological events. Model comparisons were decidedly in favour of inclusion of founder-event speciation in the models for most taxa, with only a small taxon, Uma, showing support for the model favouring vicariance. The inclusion of a constrained-dispersal matrix was favoured by three of the eight taxa examined (Cyprinodon, sand treader crickets, and Trigonoscuta). Surprisingly, tests for distance influencing dispersal probability were mostly negative. Main conclusions: Our results do not show support for any one geological event shaping the biogeographical patterns of these desert taxa. Instead, the histories of desert dune and aquatic taxa are largely products of rare jump dispersal events, and can be considered island-like systems. Although results are negative for the distance-based dispersal model, this in itself demonstrates the superiority of explicit statistical model testing over a priori assumption of fixed models in historical biogeography.

Estimating divergence times on phylogenies is critical in paleontological and neontological studies. Chronostratigraphically-constrained fossils are the only direct evidence of absolute timing of species divergence. Strict temporal calibration of fossil-only phylogenies provides minimum divergence estimates, and various methods have been proposed to estimate divergences beyond these minimum values. We explore the utility of simultaneous estimation of tree topology and divergence times using BEAST tip-dating on datasets consisting only of fossils by using relaxed morphological clocks and birth-death tree priors that include serial sampling (BDSS) at a constant rate through time. We compare BEAST results to those from the traditional maximum parsimony (MP) and undated Bayesian inference (BI) methods. Three overlapping datasets were used that span 250 million years of archosauromorph evolution leading to crocodylians. The first dataset focuses on early Sauria (31 taxa, 240 chars.), the second on early Archosauria (76 taxa, 400 chars.) and the third on Crocodyliformes (101 taxa, 340 chars.). For each dataset three time-calibrated trees (timetrees) were calculated: a minimum-age timetree with node ages based on earliest occurrences in the fossil record; a 'smoothed' timetree using a range of time added to the root that is then averaged over zero-length internodes; and a tip-dated timetree. Comparisons within datasets show that the smoothed and tip-dated timetrees provide similar estimates. Only near the root node do BEAST estimates fall outside the smoothed timetree range. The BEAST model is not able to overcome limited sampling to correctly estimate divergences considerably older than sampled fossil occurrence dates. Conversely, the smoothed timetrees consistently provide node-ages far older than the strict dates or BEAST estimates for morphologically conservative sister-taxa when they sit on long ghost lineages. In this latter case, the relaxed-clock model appears to be correctly moderating the node-age estimate based on the limited morphological divergence. Topologies are generally similar across analyses, but BEAST trees for crocodyliforms differ when clades are deeply nested but contain very old taxa. It appears that the constant-rate sampling assumption of the BDSS tree prior influences topology inference by disfavoring long, unsampled branches.