Fig 1.
State of Espírito Santo and its mesoregions.
Study area of Triatoma vitticeps occurrences and their infection by Trypanosoma cruzi-like and DTUs, with the Espírito Santo State extended by a 50 km buffer. Software: QGIS 3.22. Source: GADM data version is 4.1. Evaluable from: https://gadm.org/download_country.html.
Table 1.
Name of the 19 bioclimatic variables available on the WorldClim (https://www.worldclim.org/) version 2.1 climate data for 1970-2000. Bioclimatic variables derived from monthly temperature and precipitation data, representing annual trends, seasonality, and extreme environmental conditions.
Table 2.
Environmental covariables chosen in species distribution modeling. Variables Max Temperature of Warmest Month (BIO5), Annual Precipitation (BIO12), Precipitation of Wettest Month (BIO13), Precipitation of Driest Month (BIO14), Wind Speed Range (max - min) (m/s), Water Vapor Pressure Range (max - min) (kPa), Normalized Difference Vegetation Index (NDVI), and Topographic Diversity were selected based on Spearman’s correlation (–0.7 < ρ < +0.7).
Fig 2.
Spearman’s correlation between the bioclimatic variables.
Variables: BIO 5 (Max Temperature of Warmest Month), BIO 12 (Annual Precipitation), BIO 13 (Precipitation of Wettest Month), BIO 14 (Precipitation of Driest Month), Wind Speed Range (max - min), Water Vapor Pressure Range (max - min) and Topographic diversity. Spearman correlation values -0.7 < ρ < +0.7 were considered adequate for modeling. Software: RStudio, under R programming language version 4.1.2.
Table 3.
Environmental characterization of the occurrence points used for modeling. Environmental characterization indicates that the occurrence points are located in areas with higher temperatures (average > 18°C) and in humid to super-humid regions. The species occurs predominantly in hot, humid, and super-humid environments. Source of the Brazilian Climate Map (1:5,000,000) from IBGE (2002) [55]: https://www.ibge.gov.br/geociencias/cartas-e-mapas/informacoes-ambientais/15817-clima.html.
Fig 3.
Presence and pseudo-absence for Triatoma vitticeps, Trypanosoma cruzi-like, TcII, TcIII, TcIV and Zymodeme 3.
A: Points of occurrence and pseudo-absence of T. vitticeps; B: T. vitticeps infected by Trypanosoma cruzi-like; C: DTU TcII; D: DTU TcIII; E: DTU TcIV; and F: Zymodeme 3 (Z3). Mixed infections involving TcII and TcIII/TcIV could not be distinguished, as the PCR-RFLP assay used does not allow discrimination between these two DTUs. Software: QGIS 3.22. Source: GADM data version is 4.1. Evaluable from: https://gadm.org/download_country.html.
Fig 4.
Presence and pseudo-absence for T. vitticeps infected by the Zymodeme 3 (Z3), Z3/TcIII and Z3/TcIV.
Z3 genotype (A) and performing the combinations of the genotypes Z3 with TcIII (Z3/TcIII) (B) and Z3 with TcIV (Z3/TcIV) (C) databases. Mixed infections involving TcII and TcIII/TcIV could not be distinguished, as the PCR-RFLP assay used does not allow discrimination between these two DTUs. Software: QGIS 3.22. Source: GADM data version is 4.1. Evaluable from: https://gadm.org/download_country.html.
Fig 5.
Species Distribution Modeling of Triatoma vitticeps, Trypanosoma cruzi-like, TcIII, TcIV and Zymodeme 3 (Z3).
A: Species Distribution Model of T. vitticeps; B: T. vitticeps infected by Trypanosoma cruzi-like; C: DTU TcII; D: DTU TcIII; E: DTU TcIV; and F: Zymodeme 3. Environmental suitability ranges from 0% (blue) to 100% (red), with greater suitability in the Central mesoregion. Software: QGIS 3.22. Source: GADM data version is 4.1. Evaluable from: https://gadm.org/download_country.html.
Fig 6.
Species Distribution Modeling of T. vitticeps infected by Zymodeme 3 (Z3), Z3/TcIII and Z3/TcIV.
Z3 genotype (A) and performing the combinations of the Z3 with TcIII (Z3/TcIII) (B) and Z3 with TcIV (Z3/TcIV) (C) databases. Environmental suitability ranges from 0% (blue) to 100% (red), with greater suitability in the Central mesoregion. Software: QGIS 3.22. Source: GADM data version is 4.1. Evaluable from: https://gadm.org/download_country.html.
Table 4.
Overall performance of algorithms in the decision classification tree. Performance of the J48, RepTree, and LMT algorithms in the analysis of Triatoma vitticeps and Trypanosoma cruzi-like occurrence, highlighting the LMT algorithm, which identified 84% of infected triatomines despite an overall accuracy of 61.2%. Software: WEKA 3.8.6.
Table 5.
Classification decision tree of Triatoma vitticeps infected and not infected by Trypanosoma cruzi-like. Classification of the vector as infected or uninfected based on the relationship between true positives and true negatives (TP/TN) and false positives and false negatives (FP/FN).
Fig 7.
Classification model based on Logistic Model Tree (LMT) algorithm decision tree.
Classification performed for Triatoma vitticeps infected with Trypanosoma cruzi-like based on the Logistic Model Tree (LMT) decision tree algorithm, with variables: wind speed, maximum temperature, range temperature, month in blue, NDVI in green and species richness in yellow. The values on the lines between the nodes represent the range of values used to classify individuals as infected or not with T. cruzi. Decision tree generated with Weka 3.8.6 software.