Resultados de investigación

This package contains macroecological data on the trophic niche of terrestrial bears (Ursidae), as well as paleoecological data on the trophic niche of the European brown bear (Ursus arctos) based on stable isotope analyses., Along with the raw data we provide R code to reproduce the analysis from the paper. For details about how to run the R code please have a look at the README.txt file in the package. For details about the methods (e.g., sample collection or background of particular analyses) we refer to the original publication., German Academic Exchange Service | DAAD (No. 91568794), EEA and Norway Grants (No. POL-NOR/198352/85/2013), Peer reviewed

Full GSEA results for all genes across all variables and DESeq2 model iterations., Peer reviewed

All versions are available at DOI 10.6084/m9.figshare.25671048, This package contains macroecological data on the trophic niche of terrestrial bears (Ursidae), as well as paleoecological data on the trophic niche of the European brown bear (Ursus arctos) based on stable isotope analyses., Along with the raw data we provide R code to reproduce the analysis from the paper. For details about how to run the R code please have a look at the README.txt file in the package. For details about the methods (e.g., sample collection or background of particular analyses) we refer to the original publication., German Academic Exchange Service | DAAD (No. 91568794), EEA and Norway Grants (No. POL-NOR/198352/85/2013), Leibniz Association (No. P52/2017), Peer reviewed

All versions are available at DOI 10.5281/zenodo.13270451, This dataset is the one used in our preprint "Coevolution and temporal dynamics of species interactions shape species coexistence". R codes to analyse these data can be found here: https://github.com/f-duchenne/Evolution_pheno_vs_morpho, Peer reviewed

Accurate forecasting of patient arrivals at emergency departments (EDs) is vital for efficient resource allocation and high-quality patient care. In this study we investigate the relevance of exogenous variables, namely tourism, weather, calendar and demographic variables, in forecasting ED visits in the reference hospital in Palma de Mallorca, a city with significant seasonal population fluctuations due to tourism. Using a machine learning approach, we develop a model that predicts ED visits based solely on these exogenous variables. We test different machine learning algorithms (random forests, support vector machines, and feedforward neural networks) with different combinations of input variables and compare their symmetric mean average percentage errors (SMAPEs). Our findings reveal that calendar information, resident, and tourist population data are statistically significant for the accuracy of the predictions, while the addition of weather data does not provide any further improvement. Comparison of non-time-series with time-series prediction models reveals that the latter provide better accuracy for short prediction horizons (e.g., shorter than a week). Furthermore, time-series models become less or equally accurate to models relying only on exogenous variables for long prediction horizons (e.g., fortnight or month). Our study highlights the importance of carefully selecting predictive variables to ensure short- and long-term, robust and reliable forecasts. This demonstrates that, despite their lower complexity, non-time-series models with well-chosen input variables can be as effective as time-series models when predicting for long time horizons., Peer reviewed

All versions are available at DOI 10.5281/zenodo.7542872, Data1_Proceed.csv contains the data to analyze the abundance of fish herbivore individuals as a function of the species and the type of shoal. Data2_Proceed.xlsx contains the data to analyze the foraging activity of the herbivorous fish found in our stuty. We analyzed the relationship between bite rates per fish min-1 and the species, shoal type and shoal size. Minguito-Frutos_Bunuel_etal_2023_ProceedingsB.R contains the R reproducible code to run all the analyses carried out in this study., Peer reviewed

Subfigure on the left shows the SMAPE for the validation dataset obtained from the RF model for different values of the n_estimators hyperparameter. Subfigure on the right shows the SMAPE for the validation dataset obtained from the SVR model for different values of the degree hyperparameter. The vertical grey lines indicate the optimal values., Peer reviewed

Here nn corresponds to the module torch.nn. The main hyperparameter is the number of training epochs. To tune this hyperparameter, we ran the model for 500 epochs and checked the loss function of both training and validation sets at each epoch. In order to minimize overfitting, we set the number of epochs to 130 (see S7 Fig. Hyperparameter tuning for the FNN model)., Peer reviewed

Database of urban areas and their rural surroundings for CORDEX-CORE (and CORDEX-EUR-11 for the European domain), covering the cities selected within the framework of the CORDEX FPS URB-RCC. A detailed explanation of the dataset is provided in the preprint article submitted to Scientific Data., Peer reviewed

The figure shows the loss function at different training epochs for both the training (blue curve) and validation (red curve) dataset. The dashed vertical grey line indicates the optimal value, corresponding to the minimum of the validation curve., Peer reviewed