Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
8
datasets available to search
ShareScore release 0.9.0
Dataset results
8 results for “Ecology: statistical”
Codes in R for spatial statistics analysis, ecological response models and spatial distribution models
<p>In the last decade, a plethora of algorithms have been developed for spatial ecology studies. In our case, we use some of these codes for underwater research work in applied ecology analysis of threatened endemic fishes and their natural habitat. For this, we developed codes in Rstudio® script environment to run spatial and statistical analyses for ecological response and spatial distribution models (e.g., Hijmans & Elith, 2017; Den Burg <em>et al.</em>, 2020). The employed R packages are as follows: caret (Kuhn et al., 2020), corrplot (Wei & Simko, 2017), devtools (Wickham, 2015), dismo (Hijmans & Elith, 2017), gbm (Freund & Schapire, 1997; Friedman, 2002), ggplot2 (Wickham et al., 2019), lattice (Sarkar, 2008), lattice (Musa & Mansor, 2021), maptools (Hijmans & Elith, 2017), modelmetrics (Hvitfeldt & Silge, 2021), pander (Wickham, 2015), plyr (Wickham & Wickham, 2015), pROC (Robin et al., 2011), raster (Hijmans & Elith, 2017), RColorBrewer (Neuwirth, 2014), Rcpp (Eddelbeuttel & Balamura, 2018), rgdal (Verzani, 2011), sdm (Naimi & Araujo, 2016), sf (e.g., Zainuddin, 2023), sp (Pebesma, 2020) and usethis (Gladstone, 2022).</p> <p>It is important to follow all the codes in order to obtain results from the ecological response and spatial distribution models. In particular, for the ecological scenario, we selected the Generalized Linear Model (GLM) and for the geographic scenario we selected DOMAIN, also known as Gower's metric (Carpenter <em>et al.</em>, 1993). We selected this regression method and this distance similarity metric because of its adequacy and robustness for studies with endemic or threatened species (<em>e.g.</em>, Naoki <em>et al.</em>, 2006). Next, we explain the statistical parameterization for the codes immersed in the GLM and DOMAIN running:</p> <p>In the first instance, we generated the background points and extracted the values of the variables (<a href="https://zenodo.org/api/files/fdd5446b-dee9-4b52-ad4f-cf556443d3dd/Code2_Extract_values_DWp_SC.R?versionId=c1ea0c61-53fe-4f95-ab88-0c1cb28399cb">Code2_Extract_values_DWp_SC.R</a>). Barbet-Massin <em>et al. </em>(2012) recommend the use of 10,000 background points when using regression methods (<em>e.g.</em>, Generalized Linear Model) or distance-based models (<em>e.g.</em>, DOMAIN). However, we considered important some factors such as the extent of the area and the type of study species for the correct selection of the number of points (Pers. Obs.). Then, we extracted the values of predictor variables (<em>e.g.</em>, bioclimatic, topographic, demographic, habitat) in function of presence and background points (<em>e.g.</em>, Hijmans and Elith, 2017).</p> <p>Subsequently, we subdivide both the presence and background point groups into 75% training data and 25% test data, each group, following the method of Soberón & Nakamura (2009) and Hijmans & Elith (2017). For a training control, the 10-fold (cross-validation) method is selected, where the response variable presence is assigned as a factor. In case that some other variable would be important for the study species, it should also be assigned as a factor (Kim, 2009).</p> <p>After that, we ran the code for the GBM method (Gradient Boost Machine; <a href="https://zenodo.org/api/files/fdd5446b-dee9-4b52-ad4f-cf556443d3dd/Code3_GBM_Relative_contribution.R?versionId=1656bbae-66aa-409e-bb91-d8007dee8f95">Code3_GBM_Relative_contribution.R</a> and <a href="https://zenodo.org/api/files/fdd5446b-dee9-4b52-ad4f-cf556443d3dd/Code4_Relative_contribution.R?versionId=0e1d9352-e6b2-43da-984b-d6853a914258">Code4_Relative_contribution.R</a>), where we obtained the relative contribution of the variables used in the model. We parameterized the code with a Gaussian distribution and cross iteration of 5,000 repetitions (<em>e.g.</em>, Friedman, 2002; kim, 2009; Hijmans and Elith, 2017). In addition, we considered selecting a validation interval of 4 random training points (Personal test). The obtained plots were the partial dependence blocks, in function of each predictor variable.</p> <p>Subsequently, the correlation of the variables is run by Pearson's method (<a href="https://zenodo.org/api/files/fdd5446b-dee9-4b52-ad4f-cf556443d3dd/Code5_Pearson_Correlation.R?versionId=275f8dd4-b056-44d2-bfe5-f6264bc3298b">Code5_Pearson_Correlation.R</a>) to evaluate multicollinearity between variables (Guisan & Hofer, 2003). It is recommended to consider a bivariate correlation ± 0.70 to discard highly correlated variables (<em>e.g.</em>, Awan <em>et al.</em>, 2021).</p> <p>Once the above codes were run, we uploaded the same subgroups (<em>i.e.</em>, presence and background groups with 75% training and 25% testing) (<a href="https://zenodo.org/api/files/fdd5446b-dee9-4b52-ad4f-cf556443d3dd/Code6_Presence&backgrounds.R?versionId=d797b528-782f-4a19-bd61-cfb197f38513">Code6_Presence&backgrounds.R</a>) for the GLM method code (<a href="https://zenodo.org/api/files/fdd5446b-dee9-4b52-ad4f-cf556443d3dd/Code7_GLM_model.R?versionId=e4aca276-d601-49ec-a62c-a9223b05a7ed">Code7_GLM_model.R</a>). Here, we first ran the GLM models per variable to obtain the <em>p</em>-significance value of each variable (alpha ≤ 0.05); we selected the value one (<em>i.e.</em>, presence) as the likelihood factor. The generated models are of polynomial degree to obtain linear and quadratic response (<em>e.g.</em>, Fielding and Bell, 1997; Allouche <em>et al.</em>, 2006). From these results, we ran ecological response curve models, where the resulting plots included the probability of occurrence and values for continuous variables or categories for discrete variables. The points of the presence and background training group are also included.</p> <p>On the other hand, a global GLM was also run, from which the generalized model is evaluated by means of a 2 x 2 contingency matrix, including both observed and predicted records. A representation of this is shown in Table 1 (adapted from Allouche et al., 2006). In this process we select an arbitrary boundary of 0.5 to obtain better modeling performance and avoid high percentage of bias in type I (omission) or II (commission) errors (e.g., Carpenter et al., 1993; Fielding and Bell, 1997; Allouche et al., 2006; Kim, 2009; Hijmans and Elith, 2017).</p> <p>Table 1. Example of 2 x 2 contingency matrix for calculating performance metrics for GLM models. A represents true presence records (true positives), B represents false presence records (false positives - error of commission), C represents true background points (true negatives) and D represents false backgrounds (false negatives - errors of omission).</p> <table align="center"> <tbody> <tr> <td> <p> </p> </td> <td> <p>Validation set</p> </td> </tr> <tr> <td> <p>Model</p> </td> <td> <p>True</p> </td> <td> <p>False</p> </td> </tr> <tr> <td> <p>Presence</p> </td> <td> <p>A</p> </td> <td> <p>B</p> </td> </tr> <tr> <td> <p>Background</p> </td> <td> <p>C</p> </td> <td> <p>D</p> </td> </tr> </tbody> </table> <p>We then calculated the Overall and True Skill Statistics (TSS) metrics. The first is used to assess the proportion of correctly predicted cases, while the second metric assesses the prevalence of correctly predicted cases (Olden and Jackson, 2002). This metric also gives equal importance to the prevalence of presence prediction as to the random performance correction (Fielding and Bell, 1997; Allouche <em>et al.</em>, 2006).</p> <p>The last code (<em>i.e.</em>, <a href="https://zenodo.org/api/files/fdd5446b-dee9-4b52-ad4f-cf556443d3dd/Code8_DOMAIN_SuitHab_model.R?versionId=d951a8f2-d3a4-4804-b862-1b2762061876">Code8_DOMAIN_SuitHab_model.R</a>) is for species distribution modelling using the DOMAIN algorithm (Carpenter <em>et al.</em>, 1993). Here, we loaded the variable stack and the presence and background group subdivided into 75% training and 25% test, each. We only included the presence training subset and the predictor variables stack in the calculation of the DOMAIN metric, as well as in the evaluation and validation of the model.</p> <p>Regarding the model evaluation and estimation, we selected the following estimators:</p> <p>1) partial ROC, which evaluates the approach between the curves of positive (<em>i.e.</em>, correctly predicted presence) and negative (i.e., correctly predicted absence) cases. As farther apart these curves are, the model has a better prediction performance for the correct spatial distribution of the species (Manzanilla-Quiñones, 2020).</p> <p>2) ROC/AUC curve for model validation, where an optimal performance threshold is estimated to have an expected confidence of 75% to 99% probability (De Long <em>et al.</em>, 1988).</p>
Seeing shapes in clouds: the fallacy of deriving ecological hypotheses from statistical distributions
<p>The explanations behind observations of global patterning in species diversity pre-date the field of ecology itself. The generation of new species-area theories, in particular, far outpaces their falsification, resulting in a centuries-old accumulation in species diversity theories. We use historical assessment and new data analysis to argue that one of the earliest recognized and most consistent patterns in species diversity is not strictly an ecological phenomenon and, when ecological mechanism is invoked, the range of potential mechanisms is too numerous for tractable hypothesis falsification. We provide a historical parallel in that the normal distribution once was treated as a pattern assuming a biological mechanism rather than a statistical distribution that can be generated by biological and non-biological forces. Similarly, power law distributions are ubiquitous in aggregated data, such as the species-area relationship. That nearly identical broad-scale aggregation patterns are observed for both ecological and non-ecological data as a function of area suggest that these broad-scale patterns reflect a statistical distribution that, in itself, cannot be used to discern between or among ecological and non-ecological mechanisms. We argue that by seeking processes in such a ubiquitous pattern, ecologists may read ecological mechanism into statistical patterns, and we suggest that falsifying broad-scale diversity distribution hypotheses should be a greater priority than generating or parameterizing new ones.</p>
Statistical Decision Trees for Marine Ecological Data
<p>Both decision trees can be used to choose an appropriate statistical analysis when dealing with ecological (marine) data. For each analysis, an example publication using/explaining the method, some of the R commands required and a tutorial are listed. Please note that (1) these statistical analyses should be seen as suggestions and other approaches are available, (2) papers listed are only provided as examples but appropriate referencing of the method is expected and (3) some analyses have specific assumptions that the users should check before analysing their data. </p>
Does evaluative learning depend on the statistical relationship between stimuli? On the sensitivity of evaluative cue conditioning to ecological contingencies
<p>Evaluative conditioning (EC) is concerned with the transfer of valence from an unconditioned stimulus (US) to a conditioned stimulus (CS). Only recently the notion of EC was extended from individual CSs to categories of CSs that share certain cues. That research shows that a contingency implemented between a cue dimension and US valence has a direct effect on the evaluation of stimuli carrying values of this cue dimension. This phenomenon was coined “evaluative cue conditioning” (ECC). The present research tests whether ECC is sensitive to the contingency between a cue dimension and US valence. The present work thereby investigates the impact of ecological contingencies that define contingency by reference to all other CS-US pairings in the learning environment. Two experiments demonstrate that ECC effects are sensitive to the strength of the ecological contingency, suggesting that ECC is a relative phenomenon that depends on the valence of the CS-US pairings in the reference set. Moreover, these findings call for more research on the contingency sensitivity of standard EC effects using this revised definition of contingency.</p>
Seeing shapes in clouds: the fallacy of deriving ecological hypotheses from statistical distributions
Open the record for dataset details and reuse information.
Data associated with ecological niche models and post-ENM statistical analyses for Trillium species distributions
Open the record for dataset details and reuse information.
Data from: Statistical analysis of the individual variability of 1D protein profiles as a tool in ecology: an application to parasitoid venom
Open the record for dataset details and reuse information.
Data from: Combining statistical inference and decisions in ecology
Open the record for dataset details and reuse information.
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.