Temporal variability and flooding influence the ecological niche of Biomphalaria intermediate hosts for Schistosoma mansoni in rural Uganda
Data files
Nov 12, 2025 version files 474.82 KB
-
data.csv
161.74 KB
-
dict.xlsx
15.59 KB
-
polygons.R
11.74 KB
-
README.md
2.60 KB
-
sites_snails_models.RData
247.40 KB
-
snail_models.R
33.68 KB
-
spatial_autocorrelation.R
2.07 KB
Abstract
Understanding the niches of intermediate hosts and vectors for environmentally transmitted pathogens is crucial for identifying endemic areas, assessing habitat suitability, and targeting interventions. This study focuses on intermediate hosts of intestinal schistosomes, with over 700m people at risk of lifelong infection. We compared habitat suitability and species interactions across 674 sites in 52 villages in rural Uganda between 2022–2024, capturing a severe flooding event. Spatiotemporal models incorporating a polygon-based method to account for space with time as a fixed effect were developed to analyse snail abundance for Biomphalaria sudanica and B. stanleyi. B. sudanica was associated with marshy sites near lake shorelines and presence of hyacinths, while B. stanleyi was more likely found in deeper waters with Vallisneria plants. However, cohabitation was common for both species. Habitat suitability for each species fluctuated temporally, and more starkly with extreme flooding, resulting in switching of species dominance. Our study suggests that events consistent with climate change may influence habitat suitability without necessitating an expansion of environmental areas. Our models enable tracking of dynamic ecological niches that, if replicated elsewhere and for other intermediate hosts or vectors, can be used to better target environmental and community interventions as environmental conditions change.
This repository contains the code for running the modelling pipeline in the paper "Temporal variability and flooding influence the ecological niche of Biomphalaria intermediate hosts for Schistosoma mansoni in rural Uganda".
Journal: Proceedings of The Royal Society B
Article DOI: 10.1098/rspb.2025.2083.
Study summary
This study, conducted within the SchistoTrack cohort in rural Uganda, specifically the districts of Buliisa, Pakwach, and Mayuge, analysed the ecological niche and species interactions of Biomphalaria sudanica and B. stanleyi, intermediate hosts of Schistosoma mansoni, across 674 water sites from 2022–2024. Data were collected during four malacology surveys spanning dry and rainy seasons, including a major flooding event in 2022. Spatiotemporal models using polygon-based methods were developed to assess snail abundance and ecological niche variability across sites and time.
Packages and package versions
R version 4.2.1 (2022-06-23)
tidyverse_1.3.1
glmmTMB_1.1.10
lmtest_0.9-40
DHARMa_0.4.5
pROC_1.18.0
broom.mixed_0.2.9.4
performance_0.13.0
zeallot_0.1.0
MuMIn_1.48.4
caret_6.0-92
deldir_1.0-6
sp_1.5-0
maptools_1.1-4
sf_1.0-7
geosphere_1.5-20
factoextra_1.0.7
cluster_2.1.3
spdep_1.3-10
Files
Data to run the snail models are included in a .csv and .RData file. It is recommended to use the .RData file so that the structure of the dataframe remains intact (e.g. for variables that are factors, etc.).
dict.xlsx: Data dictionary including variable definitions and relevant units.snail_models.R: Code to run variable selection and GLMM models to identify the ecological niche of each species. Code for model validation using stratified k-fold and leave-one-out cross-validation methods. Needs to importsites_snails_models.RData.polygons.R: Code for the construction of village- and cluster-based polygons. Data not included as they include household GPS locations and identify study participants that need to remain anonymous.spatial_autocorrelation.R: Code to compute Moran's I to identify potential spatial autocorrelation. Data includes GPS locations and are thus not included (as above).sites_snails_models.RData: Dataframe used for the snail models (anonymised villages and does not include any GPS locations). Recommended to load this when runningsnails_models.R.data.csv: Same assites_snails_models.RDatabut in.csvform for easier viewing.
