The dataset contains predictions of occurrence probability for ticks in Great Britain (2014 to 2021) at 1 km spatial resolution + all covariate layers used for modeling. Over seven million electronic health records (EHRs), among which 11,741 EHRs reported tick attachment, were used to evaluate climate, environmental and animal host factors affecting the risk of tick attachment in cats and dogs in Great Britain (GB). The tick presence/absence EHRs for dogs and cats were further overlaid with spatiotemporal time-series of climatic, vegetation, human influence, hydrological and terrain variables (slope, wetness index) to produce a spatiotemporal regression matrix; an Ensemble Machine Learning framework was used to fine-tune hyperparameters for Random Forest (classif.ranger), Gradient boosting (classif.xgboost) and GLM-net (classif.glmnet) algorithms, which were then used to produce a final ensemble meta-learner that predicts the probability of occurrence of ticks across GB with monthly intervals. gb1km_covariates.zip contains ALL covariate layers as GeoTIFFs (time-series) used for modeling ticks dynamics; data_1km_2014_M01.rds = contains all covariates for January 2014 prepared as SpatialGridDataFrame (R data object); Codes of files indicate e.g.: "monthly.tick.prob_savsnet.mar_p_1km_s_2014_2021" = monthly occurrence probability for January based on the training data from 2014 to 2021; "monthly.tick.prob_savsnet.oct_md_1km_s_20211001_20211031" = monthly prediction (model) error derived as the standard deviation from multiple base learners; The dataset is described in detail in the following publication: Arsevska, E., Hengl, T., Singelton, D. et al. (2023?) Risk factors for tick attachment in companion animals in Great Britain: a spatiotemporal analysis covering 2014–2021. Submitted to Parasites & Vectors (in review). The model summary shows: Call: stats::glm(formula = f, family = "binomial", data = getTaskData(.task, .subset), weights = .weights, model = FALSE) Deviance Residuals: Min 1Q Median 3Q Max -1.4749 -0.0557 -0.0471 -0.0430 3.7611 Coefficients: Estimate Std. Error z value Pr(>|z|) (Intercept) -7.64495 0.02095 -364.957 < 2e-16 *** classif.ranger 4.95061 0.63615 7.782 7.13e-15 *** classif.xgboost 189.75543 5.53109 34.307 < 2e-16 *** classif.glmnet 140.24208 5.05375 27.750 < 2e-16 *** --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 (Dispersion parameter for binomial family taken to be 1) Null deviance: 170604 on 7303013 degrees of freedom Residual deviance: 162571 on 7303010 degrees of freedom AIC: 162579 Number of Fisher Scoring iterations: 9 Acknowledgements: We are grateful to data providers in veterinary practice (VetSolutions, Teleos, CVS, and other practitioners). We are grateful to the INRAE MIGALE bioinformatics facility (MIGALE, INRAE, 2020. Migale Bioinformatics Facility, doi: 10.15454/1.5572390655343293E12) for providing computing resources. We are also grateful for the help and support provided by SAVSNET team members Bethaney Brant, Susan Bolan and Steven Smyth. This study was funded mainly by a grant from the Biotechnology and Biological Sciences Research Council, BB/NO19547/1 and British Small Animal Veterinary Association (BSAVA). The research was partly funded by the National Institute for Health Research Health Protection Research Unit (NIHR HPRU) in Emerging and Zoonotic Infections at the University of Liverpool in partnership with Public Health England (PHE) and Liverpool School of Tropical Medicine (LSTM). This work has been partially funded by the “Monitoring outbreak events for disease surveillance in a data science context" (MOOD) project from the European Union’s Horizon 2020 research and innovation program under grant agreement No. 874850 (https://mood-h2020.eu/). The views expressed are those of the authors and not necessarily those of the NHS, the NIHR, the Department of Health or Public Health England.

{"references": ["Aegerter, J., Fouracre, D., Smith, G.C.: A first estimate of the structure and density of the populations of pet cats and dogs across Great Britain. PLOS ONE 12(4), 0174709 (2017). doi:10.1371/journal.pone.0174709"]}