Speaker
Description
Tošić, Božana1*; Gajić, Natalija2; Čabrilo, Borislav1; Bjelić-Čabrilo, Olivera1; Cvijanović, Dušanka1; Miljević, Milan3; Jakovetić, Dušan2
1 University of Novi Sad, Faculty of Sciences, Department of Biology and Ecology, Novi Sad, Serbia
2 University of Novi Sad, Faculty of Sciences, Department of Mathematics and Informatics, Novi Sad, Serbia
3 University of Belgrade, Institute for Biological Research “Siniša Stanković” – National Institute of Republic of Serbia, Department of Genetic Research, Belgrade, Serbia
* bozana.tosic@dbe.uns.ac.rs
DOI: 10.20315/evmc.2025.091
Research on the helminth fauna of small rodents was conducted in the period from 2019 to 2023 in the territory of Vojvodina Province, Serbia, in eight different localities. The rodent sample consisted of striped field mice (A. agrarius) (83), yellow-necked mice (A. flavicollis) (116) and wood mice (A. sylvaticus) (43). The mice were hosts to three helminth groups: nematodes, tapeworms, and digeneans. The aim of the study was to predict the abundance of helminths of selected species of the genera Aonchtotheca, Heligmosmoides, Syphacia and Trichuris, as well as species with zoonotic potential, i.e. Rodentolepis fraterna and Capillaria hepatica, based on various abiotic (Corine Land Use types, environmental variables, altitude, locality, region) and biotic (host species, sex, body mass, body length, spleen mass) factors. A random forest machine learning predictive model for factor importance evaluation was used to select and evaluate important features in predicting parasite abundance. P-values were estimated by using Monte Carlo analysis. The results showed that the prediction of the abundance of C. hepatica is influenced by the body condition index and spleen size of the host, and R. fraterna by the same two factors plus the mean monthly air temperature. The factors singled out as significant for species of the genus Heligmosomoides were numerous, including Corine Land Use types, all bioclimatic variables, and all biotic factors. Factors that significantly influenced the prediction of Syphacia and Trichuris species abundance were related to temperature, body condition index, and spleen mass of the host. As for Aonchotheca species, none of the factors were identified as significant. The obtained data are important from the aspect of using machine learning on these types of data and obtaining a better insight into parasite-host population dynamics, which is of particular importance when it comes to species that have zoonotic potential.
The authors gratefully acknowledge the financial support of the Ministry of Science, Technological Development and Innovation of the Republic of Serbia (Grants No. 451-03-66/2024-03/200125 & 451-03-65/2024-03/200125).
