Global risk dynamics of Borrelia miyamotoi in the context of climate change
Data files
Oct 11, 2024 version files 569.22 KB
-
appendix_2.txt
10.09 KB
-
data_for_B._miyamotoi_model.csv
30.98 KB
-
data_for_tick_models.xlsx
525.29 KB
-
README.md
2.86 KB
Abstract
The impact of Borrelia miyamotoi on human health, facilitated by the expanding geographical distribution or increasing population of Ixodes ticks, remained obscure under the context of global climate change. We employed multiple models to evaluate the effect of global climate change on the risk of B. miyamotoi worldwide across various scenarios. The habitat suitability index of four primary vector tick species for B. miyamotoi, including Ixodes persulcatus, Ixodes ricinus, Ixodes pacificus, and Ixodes scapularis, was projected using a boosted regression tree model, considering multiple sharing socioeconomic pathway scenarios within multiple time periods. The modelling analysis reveals that apart from Ixodes scapularis, future global warming will result in a northward shift in other three vector tick species and a gradual reduction in suitable habitats. Random forest models indicate consistent changes in B. miyamotoi and its primary tick species, with potential risk areas shrinking and northward shift, particularly in the eastern USA, northeastern and northern Europe, and northeast Asia. These findings highlight the urgent need for enhanced active surveillance of B. miyamotoi infection in primary vector tick species across projected potential risk areas. The effect of climate change on B. miyamotoi distribution might have potential reference significance for public health decision-making of tick-borne pathogens.
https://doi.org/10.5061/dryad.zkh1893hs
These datasets are the data used in the paper on Environmental Microbiology entitled “Global risk dynamics of Borrelia miyamotoi in the context of climate change”. I. persulcatus, I. ricinus, I. pacificus, and I. scapularis have been identified as primary vectors transmitting B. miyamotoi. Distribution databases of these four vector tick species are compiled from publicly available data. We use an integrated multi-model, multi-scenario framework to assess the impact of climate change on B. miyamotoi at the global scale. Based on publicly available data summarizing the detection of B. miyamotoi in humans, vectors, and animals between 2000 and 2020, ensemble machine learning models are adopted to analyze the main factors affecting the distribution of B. miyamotoi, and to predict the potential risk areas of B. miyamotoi.
Description of the data and file structure
The file named “data_for_tick_models” includes the distribution data of four major vector tick species, namely I. persulcatus, I. ricinus, I. pacificus, and I. scapularis, for B. miyamotoi between 2000 and 2020. The main information contains sampling time, country, longitude, latitude, species, and data source. We removed a small amount of data supplemented from GBIF to satisfy the condition of the CC0 waiver.
The file named “data_for_B. miyamotoi_model” includes the distribution data of B. miyamotoi detected in humans, vectors, and animals between 2000 and 2020. The dataset also provides the number of detection samples and the corresponding detection positivity rate for each sampling location.
We confirmed that all the data that was derived from an open-access study is compatible with the CC0 license waiver required by Dryad.
The supplementary file containing the niche modeling codes for vector tick species and B. miyamotoi is uploaded as “appendix 2.txt”.
Sharing/Access information
NA
Code/Software
The data could be load in R version 4.1.3 by base package. The geographic coordinates of these records were plotted on the created grid-map to obtain corresponding grid codes. After removing duplicates, occurrence grids representing the distribution of the four vector tick species were obtained to correlate all the variables. The boosted regression tree (BRT) model used in this case-control study was developed by R packages dismo and gbm and was employed to predict both current and future potential habitat for four vector tick species. To obtain the optimal predictive model, we employed the random forest (RF) model using R package randomForest and caret to investigate the association between ecological factors and B. miyamotoi occurrence.