Loading...
Loading...
Mathematical datasets, statistical benchmarks, probability, optimization, operations research
2,489 datasets
Statistical ensemble modeling forecasts current and future circum-Arctic Mean Annual Ground Temperature (MAGT) and Active Layer Thickness (ALT). The MAGT forecasts indicate permafrost currently prevails over an area of 15.1 ± 2.8 million square kilometers. The models were developed by Juha Aalto and published in 2020, showing region-specific changes in ground thermal regime due to climate change.
Data supports a Bayesian spatial capture-recapture (SCR) model using Local Evaluation of the State-Space (LESS) to estimate animal density. The study maps wolverine (Gulo gulo) density across the species' entire range in Norway, covering over 200,000 km². The LESS method reduced model computation time by up to 57-fold in simulations.
78 ingroup taxa of Cyatheaceae ferns were analyzed to reconstruct large-scale historical biogeographical patterns. The study tests vicariance hypotheses related to the break-up of Gondwana and identifies limited transoceanic dispersal events. It uses DNA sequence data analyzed with Bayesian inference and Lagrange for maximum likelihood biogeographical reconstruction.
Dryad hosts simulation data from a study on optimum design and resource allocation for family-based association mapping. The research used simulation, resample model averaging, and cross-validation to analyze factors like population size and phenotyping intensity. Results indicate predictive power and QTL detection accuracy depend heavily on these design parameters.
Experimental data from a scientific reevaluation of Drosophila melanogaster courtship song rhythms. The dataset was generated by David L. Stern to investigate claims of periodic song pulse intervals, concluding no evidence for such rhythms. It includes manually annotated and automatically segmented recordings from multiple fly strains.
A collection of quantified structural components from 48 internal saphenous vein samples stained with Masson's trichrome. It was created by Pablo Hernández-Morera to develop and test digital image processing and statistical methods for analyzing vessel wall changes. The methods segment smooth muscle fibers and extracellular matrix, with results validated against expert medical judgment.
Graham Jones provides data and models for inferring evolutionary histories of allopolyploid species, which arise from hybridization. The work evaluates two new Bayesian models, AlloppMUL and AlloppNET, implemented in the BEAST framework. It includes simulation results and demonstrations on empirical data from the plant genera Pachycladon and Silene.
This dataset supports the evaluation of the multispecies coalescent model's statistical fit using posterior predictive simulation. It includes empirical data from Tamias chipmunks and Myotis bats, where hybridization and gene flow are suspected. The data was used to test the P2C2M R package, which employs probability-based summary statistics to assess model fit.
A collection of spatiotemporal distribution data for polar bears, Atlantic walruses, and ringed seals in the Kara Sea, derived from a hierarchical Poisson point process model applied to heterogeneous survey data. It includes relative density estimates linked to environmental covariates like ice concentration and distance to coast over a 17-year study period. The data was compiled by Jussi Mäkinen from scientific literature and open repositories to analyze distributional shifts.
Aggregating simulation results used to characterize the scaling behavior of the *BEAST method in Bayesian phylogenetics. The data enables quantitative prediction of how increasing the number of loci impacts computational performance and statistical accuracy, comparing *BEAST to concatenation and other species tree methods.
Featuring 81,924 measures of cortical thickness derived from 1,748 anatomic MRI scans of 792 healthy twins and siblings. It was used to map dynamic genetic contributions to brain development across childhood and adolescence, with up to eight longitudinal time points per subject.
A dataset supporting the inference of speciation and extinction rate variation using the fossilized birth-death process. It includes phylogenies for both extant and extinct species, implemented and tested within the BAMM computational framework. The data was authored by Jonathan S. Mitchell and last updated in June 2020.
Data from a study evaluating Bayesian clustering methods (BAPS and STRUCTURE) for detecting hybridization and introgression in a reintroduced red wolf population. The dataset uses 17 microsatellite loci from individuals with known ancestry percentages (50–100% red wolf), based on a detailed pedigree. It provides an empirical test case for assessing admixture estimation accuracy in endangered species management.
A study of locomotor activity, rumen temperature, and heart rate in free-living Svalbard reindeer across the Arctic year. Lomb-Scargle periodogram analyses with high statistical power found diel or circadian rhythmicity persisted throughout the year, including during Polar Night and midnight sun. The data reveal profound seasonal changes in foraging, metabolic activity, and rhythm power as adaptations to extreme High Arctic conditions.
Presenting predictions of invasion risk for the biomass crop Miscanthus × giganteus across the continental United States. It combines outputs from simulation-based statistical models and bioclimatic models to produce aggregated risk maps. The data was created by Ranjan Muthukrishnan and published in 2020.
Encompassing simulation results from a study comparing Bayesian methods for estimating genomic breeding values of threshold traits. The study introduced three new methods (BayesTA, BayesTB, BayesTCπ) and evaluated their accuracy improvements over standard methods, with one scenario showing a 30.4 percentage point gain. It was authored by Qin Zhang and published in 2020.
Aggregating behavioral and neurophysiological measures from a single-case proof-of-principle study investigating feedback update intervals in restorative brain-computer interfaces for stroke rehabilitation. The primary behavioral measure, the Action Research Arm Test, showed a 36% clinically significant increase over the training course. Neurophysiological measures include motor evoked potentials and maximum voluntary contraction.
This dataset contains simulation results and empirical genetic data from the Sceloporus scalaris lizard species group, used to evaluate Bayes factor methods for species delimitation. It was created by Jared A. Grummer and published in 2020. The data supports the recognition of all scalaris group taxa and three previously undescribed lineages as independent evolutionary lineages.
Data from a study by Tracy A. Heath presents a hierarchical Bayesian model for calibrating species divergence time estimates using multiple fossil age constraints. The model employs a Dirichlet process prior as a hyperprior on calibration density parameters and was evaluated using both simulated and biological data. The dataset supports the analysis of node age estimates under this hierarchical approach.
Data from a study evaluating Bayesian species delimitation methods for cryptic mycorrhizal fungi. It includes DNA sequence data for 7 nuclear and mtDNA loci from 147 fungal isolates obtained from the Australian orchid genus Caladenia. The analysis found strong support for eight distinct Serendipita species.