Faecal pathogens and ectoparasites associated with small mammals in forest fringes around Sydney, Australia

This dataset contains curated and Hellinger-tranformed sequencing data obtained from DNA extracted from small mammal scats collected in Ku-ring-gai Chase National Park, and surrounding urban reserves. We present the csv files neccesary to obtained the published results of our study, in which we aimed to analyse the influence of host species identity and traits (i.e. sex, body mass index; BMI), and seasonality on the presence of faecal pathogenic fungi and bacteria as well as the ectoparasites associated with small mammals inhabiting forest reserves near urban areas in the Sydney region, New South Wales (NSW), Australia. Across samples, we identified 12 pathogenic fungi, nine bacterial pathogens, and 15 ectoparasite taxa. The most abundant representatives of each group were Malassezia japonica (fungi), Escherichia coli (bacteria), and Siphonaptera (fleas). Host traits influenced pathogen and ectoparasite occurrence in distinct ways. Host sex affected flea prevalence, with males more frequently infested than females. Host body mass index had no detectable effect on pathogen or ectoparasite presence. Host species was a strong predictor with Rattus fuscipes being more likely to carry fleas and mites, whereas Antechinus stuartii had a higher likelihood of harbouring fungal and bacterial pathogens in their scats. Seasonality also shaped pathogen and ectoparasite dynamics. Pathogenic fungi, bacteria, and ticks were more common in the autumn (wet season), whereas flea prevalence was highest in spring. Overall, our findings underscore the importance of broad-scale assessments of pathogen communities in wildlife species that live near humans, as such work is critical for identifying potential vectors and emerging zoonoses.

Methods
We live-trapped small mammals at five sites in Ku-ring-gai Chase National Park, each paired with one surrounding urban reserve. The urban reserves were located at least 100 m outside the KNP boundary and separated from one another by at least 1.5 km. At each site, we installed 50 Elliot traps in a sampling grid spaced at 10 m intervals. At the forest sites, the sampling grids were 5 × 10 trap rectangles. At the urban reserves, this grid arrangement was not always feasible due to the irregular shape of some sites. We sampled a pair of sites simultaneously. Each trap was baited with a mixture of peanut butter, vanilla essence, and oats. We sampled the sites during both spring (September 2023) and autumn (April 2024). Upon capture, any small mammals showing signs of stress (16% of captures) were immediately released. The remaining individuals were taxonomically identified. After processing the captured individuals, any fresh scats present were collected in 1.5 ml Eppendorf tubes from inside the Elliot traps. We did not collect scats from recaptured individuals. The scat samples were immediately placed in a cooler box with ice packs following collection and then transferred to a -80 °C freezer upon arriving back in the laboratory. We extracted DNA from the scat samples after each sampling season using the DNeasy PowerSoil Pro kit from Qiagen®. We processed 108 scat samples that were sequenced with the ITS 1F and ITS 2R and used the DADA2 pipeline to process the sequencing data and identify the fungal species using the UNITE database for fungi version 10.0. These data were then assigned to functional guilds using the package FunguildR. Sixty of these samples were also sequenced using the 16s primers 16s V3-V4 primers 341F and 806R. The DADA2 pipeline was used to curate this data, where the taxonomical assignment was done using the reference database SILVA taxonomic training data version 138.2. The identified bacterial taxa were classified into ecological functional groups using the FAPROTAX and the NJC19 databases in microeco R package, and only human pathogenic bacteria were filtered for further analyses. For the ectoparasites, each captured small mammal was inspected for the presence of ectoparasites for two minutes. Upon collection, the ectoparasites from each mammal were submerged in 100% ethanol in an Eppendorf tube. Individual ectoparasites were examined under a stereomicroscope for identification. 

Data analyses. We calculated the overall prevalence of fungal and bacterial pathogens, as well as ectoparasites for each host species and season. We then developed binomial generalised linear models to predict the probability of presence for each pathogen group, and per ectoparasite order (i.e., Ixodida, Mesostigmata and Siphonaptera). We generated a global model in MuMIn to assess the influence of host BMI and species, as well as season on the presence/absence of each of the above mentioned pathogens and ectoparasite groups.