Alien mammal introductions can reshape global viral sharing networks

Alien mammal introductions can reshape global viral sharing networks | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Alien mammal introductions can reshape global viral sharing networks Andrea Tonelli, Gregory Albery, Moreno Di Marco This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7622438/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Alien mammals introduced beyond their native distribution ranges can bring novel pathogens into the colonised communities and alter pathogen transmission dynamics. Past studies identified immunological and ecological drivers of cross-species viral transmission, but this knowledge has rarely been applied to alien species, leading to an underestimation of their role in disease emergence. Here we predict the viral sharing network resulting from the establishment of 67 alien mammals introduced globally in the last 50 years, using a trait- and phylogeny-informed viral sharing model. We show that the introduction of alien mammals can result, on average, in six novel viral sharing events per introduction (95% CI = 5.03–6.98), potentially reshaping the viral sharing networks of local communities. Phylogenetic relatedness emerged as the strongest predictor of viral sharing between alien and native species, with additional contributions from trait-based, dietary, and habitat similarities. Predicted viral sharing was concentrated in the Global North, reflecting potential geographic biases in both introduction records and viral surveillance. Our approach provides a quantitative tool to estimate viral hazards driven by established alien species that can be used to support risk assessment frameworks and international policy on biological invasions. Biological sciences/Ecology/Ecological epidemiology Biological sciences/Ecology/Ecological networks Biological sciences/Ecology/Macroecology Figures Figure 1 Figure 2 Figure 3 Figure 4 Main The rate of alien species introduction has increased steadily over the past decades 1 , representing a threat to ecosystem stability and human well-being 2 , 3 . Alien species introduced into new regions – either deliberately or accidentally – may carry novel pathogens into colonised environments 4 , 5 , potentially leading to spillover into naïve populations of native species, including humans. Approximately half of the vertebrates included in the list of the world’s worst invasive alien species 6 have been linked to wildlife diseases 7 , and pathogen-mediated competition is one of the ways by which alien species can displace native fauna 8 . For instance, in the UK, native red squirrel populations ( Sciurus vulgaris ) declined up to 25 times faster when in contact with squirrelpox virus (SQPV)-seropositive alien grey squirrels ( Sciurus carolinensis ), compared to those exposed to seronegative populations 9 . Even when introduced species do not carry novel pathogens into the colonised environments, they may become infected with local pathogens and affect existing transmission networks, thereby altering pathogen dynamics within the local community 10 . Through such direct and indirect pathways, biological invasions are increasingly recognised as drivers of emerging infectious diseases globally, including zoonoses of concern for human health 11 , 12 . This evidence underscores the importance of assessing the disease-related impacts of alien species introductions for both biodiversity conservation and public health. The contribution of alien species to pathogen transmission dynamics, including those of public and animal health concern, is frequently underreported 13 , 14 and projections of emerging zoonoses often neglect the role of alien species as potential drivers of disease emergence and spread 10 . This is in part due to incomplete data on pathogen distributions in wildlife and the inherent difficulty of predicting cascading effects following species introductions. Assessments of how alien species alter the prevalence and distribution of pathogens remain rare and typically limited to well-studied examples (e.g., brushtail possums as reservoirs of bovine tuberculosis in New Zealand 15 ). As a result, the risk of pathogen spillover is seldom integrated into formal risk assessment frameworks for invasive alien species 16 , and quantitative tools to estimate the potential for pathogen transmission following alien species introductions are still lacking. Addressing this gap requires predictive approaches that can assess disease hazards and inform international policy on biological invasions 12 . Building on existing theory on pathogen sharing networks 17 – 20 , we aim to identify quantifiable patterns that drive cross-species viral sharing between alien and native mammals. We used a viral sharing model 21 to predict the probability of viral sharing between alien and native mammal species following the establishment of alien populations introduced globally over the past 50 years (Fig. 1 ). This temporal window allowed us to assess alien species that are already established in their non-native ranges, while remaining early enough to explore potential but still unaddressed viral transmission in local communities 22 . We modelled viral sharing probability as a function of predictors that characterise species’ pairwise similarity and eco-geographical opportunity for pathogen sharing, approximating the immunological and ecological barriers to cross-species transmission: phylogenetic distance, geographic overlap of suitable habitat, life-history trait similarity, and foraging similarity. We also accounted for levels of research efforts, to account for knowledge biases across different areas and different species. We predicted the probability of viral sharing across unique pairs of alien and native mammals that overlap in geographic areas where the alien species got established. We highlight that introduction of alien mammalian species can lead to hundreds of novel opportunities for cross-species viral transmission in alien ranges, especially between closely related host species with similar traits and dietary habits that share the same habitat. At the same time, we found research efforts to have an important effect on our predictions, highlighting potential knowledge gaps affecting wildlife disease research in the context of alien species introductions. Results We used an extreme gradient boosting model to uncover viral sharing patterns within a sample of 1,116 mammal species which share at least one virus with co-occurring species. The model was trained on a sharing network made of 78,241 pairs of spatially overlapping mammals, derived from 1,920 mammal–virus associations gathered from a published database of host virus associations 23 (see Methods). We then applied the viral sharing model to 67 species that established alien populations outside their native ranges in the last 50 years, in order to predict potential viral sharing between alien and native mammals following introduction. Our viral sharing model achieved moderate performance during nested cross-validation (Supplementary Table 1), with a mean AUC of 0.83 (s.d. = 0.01) and a mean recall of 0.71 (s.d. = 0.01). When tested on an independent set of 343 native-alien pairs known to share at least one virus (see Methods) the model showed a mean out-of-sample recall of 0.71 (s.d. = 0.04), being able to correctly predict viral sharing for 244 of those pairs (median, 95% CI = 235–253). Global drivers of cross-species viral sharing We obtained variable importance scores during model training by estimating the average gain (improvement in accuracy) contributed by each predictor variable (Supplementary Fig. 1). Phylogenetic similarity between species emerged as the strongest determinant of viral sharing (median gain = 0.37), followed by research effort, quantified as the sum of species-level citations of the pair. Foraging and trait similarities contributed moderately and at comparable levels, while the extent of geographic overlap had a relatively small influence on model performance. We explored patterns of viral sharing as a function of the interaction between pairs of predictor variables, holding all other variables constant (Fig. 2 ; Supplementary Fig. 2). As expected, higher phylogenetic similarity (i.e., smaller patristic distance) strongly increased the probability of viral sharing across all combinations with the other predictors. Phylogenetic similarity and the extent of geographic overlap showed an additive positive effect, with sharing probability peaking at the highest phylogenetic similarity values (above 0.90, approximately the average similarity between congeneric mammals) and geographic overlap above 10 million Km 2 . Sharing probability showed a sharp increase around a phylogenetic similarity of ~ 0.45, which roughly corresponds to the similarity between different bat families, beyond which sharing probability remained high, with geographic overlap contributing only a minor additive effect. Interestingly, viral sharing probability showed a spike at the lowest values of phylogenetic similarity (~ 0.02, which roughly corresponds to the divergence of Xenarthra and Afrotheria from other mammalian lineages), where it increased evenly along increasing geographic overlap. The interaction between foraging similarity and geographic overlap showed a mild additive trend, with sharing probability peaking at both high and low foraging similarity and around 10 million Km2 of geographic overlap, before declining in regions where the availability of training data was scarce. Foraging similarity had a limited contribution to sharing probability in combination with phylogenetic similarity, only showing a noticeable effect at low values of phylogenetic similarity, producing a spike in sharing probability (p ≈ 0.6) at the lowest phylogenetic similarity values and high foraging similarity. Sharing probability increased with increasing trait similarity, and plateaued for lower values, generally showing minor effects in combination with other predictors. Novel viral sharing driven by the introduction of alien mammals We built an alien-native network that included 3,153 unique pairs of alien and native mammals that overlapped geographically in areas where the alien species had been introduced within the past 50 years. Among the 67 alien species considered, the majority were even-toed ungulates (37.3%), followed by rodents (25.4%), carnivores (20.9%), insectivores (6.0%), lagomorphs (4.5%), pangolins, armadillos and primates (collectively 6.0%). Most species (58.2%) had multiple introduction events, both across different locations and at different points in time. The ten species with the highest number of introduction events accounted for approximately half of all introductions: Dama dama (n = 65), Cervus nippon (n = 45), Myocastor coypus (n = 29), Eutamias sibiricus (n = 19), Neovison vison (n = 13), Oryctolagus cuniculus (n = 11), Procyon lotor (n = 9), Callosciurus erythraeus (n = 7), Funambulus pennantii (n = 7), Ondatra zibethicus (n = 7). Reported introduction events spanned all continents, with the highest concentrations in continental Europe (28.3%) and New Zealand (22.9%). We applied the viral sharing model to identify potential novel viral sharing events between 67 alien mammals and the 1,090 native mammals found in their alien ranges. The model predicted a median of 616 novel viral sharing events (95% CI = 582–650). The ten species with the highest predicted sharing accounted for approximately 62% of all novel viral sharing events (Fig. 3 ). About 54% of novel viral sharing events were associated with alien species whose native ranges encompass multiple biogeographic realms, and about 55% involved species from the palearctic region. We used SHapley Additive exPlanations (SHAP) to assess the contribution of individual predictors to the probability of viral sharing between alien and native mammal species. SHAP values quantify the marginal contribution of each feature to the model’s output for each observation, offering an interpretable framework for understanding complex machine learning predictions 24 . SHAP values largely confirmed the trends found for the patterns that drive viral sharing in native ranges, where phylogenetic similarity between species was the most important explainer of viral sharing (sum of |SHAP| = 253.72) once research effort was accounted for (Supplementary Fig. 3). Importantly, increasing values of geographic overlap, foraging and trait similarity gave important contributions to sharing probability even for distantly related species, sometimes showing non-linear effects with phylogenetic similarity (Supplementary Fig. 4). We predicted that each introduction event would generate, on average, approximately six sharing events with native mammals (95% CI = 5.03–6.98). As a comparison, mammals in our training dataset had, on average, approximately 27 observed sharing links (95% CI = 24.51–29.04) in their native ranges. Novel predicted sharing events cluster strongly in continental Europe and Japan, two regions with a high density of reported alien introductions, and include long-distance connections across multiple biogeographic realms, especially between Europe, Asia, and the Americas (Fig. 4 ). Consistently with the general trends of alien species introduction and reporting, the main hotspot of alien-driven viral sharing was continental Europe, with 201 novel sharing events between unique alien-native pairs and an average of approximately 11 sharing events per introduction. Other areas that were predicted as important hotspots of novel alien-native viral sharing were Japan with 72 events, and Africa with 51 events (42 of which were linked to the introduction of the Asian house rat, Rattus tanezumi ). New Zealand was the landmass with the highest proportion of predicted sharing events on the number of alien-native encounters due to its low mammalian species richness. Other areas with a high proportion of realized viral sharing (i.e., the proportion of predicted alien-native viral sharing events on all alien-native encounters) included Australia and Japan, together with several smaller islands such as Bornholm, Green Island, La Palma, Lolland, Saint Martin, and Shetland (Supplementary Fig. 5). The effect of sampling bias When we repeated the predictions while holding the sum of citations for each alien–native mammal pair at the median value, the predicted number of potential novel viral sharing events was 651 (median, 95% CI = 606–696). Under these conditions, the spatial distribution of predicted viral sharing events showed a general reduction of both the number (Supplementary Fig. 6) and the proportion of sharing events (Supplementary Fig. 7). These patterns were especially evident across Japan, New Zealand and North America. The analysis of SHAP values highlighted research effort as one of the strongest drivers of predicted viral sharing (Supplementary Fig. 8). Particularly, virological research effort was a very strong predictor for several well-studied species associated to adverse impact on conservation and/or human health, such as the raccoon ( Procyon lotor ), the European rabbit ( Oryctolagus cuniculus ) and the European fallow deer ( Dama dama ). Discussion We showed that the introduction of alien mammal populations beyond their native ranges can create novel opportunities for viral sharing with native species. By applying a trait- and phylogeny-informed viral sharing model, we demonstrated that the probability of cross-species viral transmission is shaped by evolutionary and ecological compatibility between alien and native hosts, extending the existing theory of pathogen sharing in the context of biological invasions. Consistently with previous studies on pathogen sharing 18 , 20 , 25 , 26 and parasite acquisition by alien species 27 , we found phylogenetic similarity to be the main driver of viral sharing. This pattern likely reflects conserved molecular and immunological traits between closely related species, which reduce cross-species barriers to viral transmission 28 . For example, the Asian house rat ( Rattus tanezumi ), introduced to South Africa as a ship stowaway, was among the alien species with the highest number of potential viral sharing events. According to our model outputs, this was mostly driven by its close phylogenetic proximity to native African rodents and its strong ecological overlap with other small mammals in the region. Our findings also underscore the importance of ecological similarity as a complementary predictor of mammal-mammal viral sharing. In cases where phylogenetic distance was high, similarities in diet and habitat use increased predictive accuracy, serving as important proxies for potential interspecific contact. Indeed, some viral sharing links were better explained by the extent of geographic overlap, foraging and trait similarity than by phylogenetic relatedness alone. This was the case for the North American raccoon ( Procyon lotor ), a habitat and dietary generalist whose novel sharing events in Japanese islands were mostly driven by high geographic and ecological overlap with native mammals even when phylogenetic distance was high (such as with rodents and bats), once the effect of research effort was excluded. This shows the importance of considering ecological proxies of species interactions, such as similar habitat use and diet preferences, to gather a more comprehensive picture of mammal-mammal viral sharing potential. Our results imply that novel alien-driven viral sharing can substantially reshape the structure of viral sharing networks in colonised areas worldwide. While our analysis focussed on predicting novel direct links between overlapping species in alien ranges, the ecological consequences are likely to extend far beyond individual species pairs. As host species are embedded within broader ecological networks, transmission can potentially go further than the single alien-native interaction, affecting a broader part of the local community. Furthermore, introduced species may undergo shifts in their realised or fundamental niches 29 , potentially leading to further adaptation and range expansion 30 . As alien species expand their distributions, our analysis suggests that the probability of viral sharing with co-occurring species will increase, further affecting the cross-species transmission network within colonised ecosystems. While phylogeography and ecological compatibility were key predictors of viral sharing, research effort (representing knowledge bias) gave an important contribution to model performance. Species with more known viral associations were more likely to be involved in predicted sharing events, consistently with biases in the known virome of mammals 31 that influenced both individual predictions and broader spatial patterns of viral sharing 20 , 26 . Most of the alien viral sharing events predicted by our model were localised in the Northern Hemisphere. This pattern likely reflects a dual bias: one in viral discovery efforts, and another in the underlying distribution of introduction records. The latter is consistent with broader patterns observed for terrestrial vertebrate records 32 and invasive species reporting 33 . A previous assessment of invasive alien mammals impacting human health also detected a potential overrepresentation of wealthier countries, identifying European countries, Japan and Australia as the main global hotspots 34 . By contrast, tropical regions – where alien species detection capacity and management are limited – may represent under-recognised hotspots of alien species introductions and cross-species transmission risk. Addressing these knowledge gaps across taxa and geographic areas could potentially reshape the perceived hazard of cross-species transmission driven by alien mammals at the global scale. Biological invasions are inherently stochastic, and cross-species viral spillover events are even more so – driven by a complex interplay of ecological, evolutionary, and environmental factors. Yet, analytical tools that integrate ecological and evolutionary insights into predictive frameworks may offer valuable information for assessing and anticipating disease risks associated with alien species introductions. Given the threats posed by emerging infectious diseases and the ongoing biodiversity crisis, these tools can support evidence-based decision-making at the interface of wildlife conservation and public health and facilitate the integration of disease hazard evaluations into international policy frameworks for invasive alien species. Methods The viral sharing network We obtained host-virus associations for 1,249 mammalian species from the Vertebrate Virome in One Network (VIRION 23 ). We discarded viral entities that were not ratified by the International Committee on the Taxonomy of Viruses (ICTV 35 ), to avoid introducing potentially incorrect information and pseudoreplications. We followed Albery et al. 21 to obtain a unipartite mammal–mammal network from the bipartite network of mammal–virus associations. In this unipartite network, edges indicate whether a pair of mammalian species share at least one virus. Unobserved edges that involved species with at least one link were treated as pseudo-negatives and encoded as pseudo-absences (0s) during model training. This filtering step minimised bias from potentially undersampled species lacking observed viral sharing due to limited surveillance, thereby reducing overrepresentation of false negatives. Self connections and duplicate links were filtered out, and 49 marsupials and 1 monotremes were also excluded as they were extreme phylogenetic outliers. As we were interested in the patterns that drive viral sharing in the context of biological invasions, when the introduction of an alien species leads to novel mammal co-occurrences, we restricted the network to only include sympatric species pairs (i.e., those with overlapping geographic ranges). This left us with 1,116 mammal species for which complete data on biological traits, geographic ranges, and phylogeny were available, resulting in 78,241 unique mammal-mammal pairs. Of these, 23.8% shared at least one virus and 6.2% shared more than one virus. Predictors of viral sharing We retrieved the distribution of the 1,116 mammal species using Area of Habitat (AOH) maps 36 . These maps provide more refined information on species presence than range polygons, by filtering out areas of the range that are not suitable for the species, thus reducing the risk of spurious geographic overlap due to overestimated range extents and suitability. We resampled AOH rasters (100m resolution) to a resolution of 5km with the terra R package, using a sum of cell values aggregating function. Presence cells were retained using a ≥ 25% threshold of cell coverage, or the maximum coverage for species without any cell above the threshold. This value was selected as a compromise to avoid representing marginal species co-occurrence in a cell while avoiding penalising habitat specialists which might rarely have high coverage of suitable habitat in any cell. As a sensitivity analysis, we applied a > 5% cell coverage threshold to a subset of 100 randomly sampled species, resulting in 32,697 overlaps with other mammals. Geographical overlap under this threshold was highly collinear with that obtained using the 25% threshold (Pearson’s r > 0.99); this indicates that changing the threshold is unlikely to significantly impact our results. We then calculated the spatial overlap between ranges, quantified as the number of overlapping cells within the intersection of the AOHs. For alien species, the alien range polygons were obtained from the global Distribution of Alien Mammals database (DAMA 37 ). The ranges of alien species that had multiple distinct introduction ranges in the selected time period were merged to obtain a single disjointed range. We converted the alien ranges into AOH maps, following the same procedure used in Lumbierres et al. 36 for full comparability with native AOH maps, and followed the same steps described above to obtain alien-native geographic overlaps. For each unique pair of mammals, we calculated biological traits and foraging dissimilarities between species using the Gower’s distance 38 , a metric used to measure dissimilarity in multidimensional trait space 39 . Gower’s distance can handle continuous, categorical and ordinal variables (e.g., foraging classes), making it appropriate for our mixed dataset. Biological and foraging traits of the species were retrieved from the Coalesced Mammal Database of Intrinsic and Extrinsic traits (COMBINE 40 ). Only traits with > 30% original data completeness and low imputation error (as reported in COMBINE) were selected. Biological trait dissimilarities were computed using body mass, longevity, age at first reproduction, gestation length, litter size, litters per year, and weaning age. Foraging dissimilarities were computed using trophic level, foraging stratum, and the percentages of the species’ diet composed of vertebrates, invertebrates and plants. We also obtained pairwise patristic distances for all mammals from PHYLACINE 41 . We then applied min–max scaling to each distance metric, rescaling them to range from 0 to 1. Finally, we obtained biological, foraging, and phylogenetic similarity by subtracting 1 from the respective scaled distances. Data sources, description and rationale for each predictor is provided in Table 1 . Table 1 Description, rationale and data sources for the predictor variables used to model viral sharing. Variable Description Rationale Data source Geographic overlap Number of cells (resolution = 25 Km 2 ) in overlap within the intersection of ranges. (Log 10 transformed) Greater geographic overlap provides more opportunities for viral sharing Native ranges: AOH from 36 ; Invasive ranges: AOH calculated from 37 Phylogenetic similarity Relative phylogenetic similarity quantified as: \(\:1-\frac{patristic\:distance}{max\left(patristic\:distance\right)}\) Closely related species have similar physiology and immune systems which may make them susceptible to the same viruses 41 Foraging similarity \(\:1-\frac{forag\:dist\:-\:min\left(forag\:dist\right)}{max\left(forag\:sit\right)\:-\:min\left(forag\:dist\right)}\) Where forag dist is the Gower distance between two species based on trophic level, foraging stratum, and the percentages of vertebrates, invertebrates and plants in the diet Having similar dietary niches may increase the ecological opportunities for viral sharing 40 Trait similarity \(\:1-\frac{trait\:dist\:-\:min\left(trait\:dist\right)}{max\left(trait\:dist\right)\:-\:min\left(trait\:dist\right)}\) Where trait dist is the Gower distance between two species based on body mass, longevity, age at first reproduction, gestation length, litter size, litters per year, weaning age Life-history traits correlate with determinants of pathogen susceptibility and exposure 40 Sum of citations Sum of the number of virus-related papers for the two species. (Log 10 transformed) Accounting for research effort 42 We also controlled for bias in research efforts across species by collecting publication counts for each species. We obtained species-level virus-related paper counts to account for virological sampling effort in mammals, through an automated screening of Web of Science papers using the R packages httr 43 and jsonlite 44 . To account for variation in research effort, we summed paper counts for each mammal pair and applied a Log 10 transformation. Modelling viral sharing probability We trained an extreme gradient boosting model using the xgboost 45 algorithm in R, with binary viral sharing as the response variable. Predictor variables included geographic overlap (Log 10 transformed), phylogenetic similarity, trait similarity, foraging similarity and sum of citations (Log 10 transformed). Model training, tuning, and validation were conducted using a repeated nested cross-validation 46 , 47 with 10 inner folds for hyperparameter tuning and 10 outer folds for model evaluation. At each model iteration, the best combination of hyperparameters was selected based on the average recall calculated over the assessment inner folds. The best model was then evaluated against the outer test sets to estimate performance metrics. The whole process was repeated 10 times. The range of hyperparameter values selected during model tuning is provided in Supplementary Table 2. We extracted permutation-based variable importance from the trained model to identify the main determinants of viral sharing between mammals and visualised the pairwise interaction effect of predictors on sharing probability when all other predictors were held at their median values. The modeling workflow was implemented using the tidymodels 48 R package. Predicting viral sharing for alien-native pairs of mammals We applied the fully trained viral sharing model to a set of selected alien species introduced in areas outside their native range in the last 50 years for which we had complete information on biological and foraging traits, geographic range and phylogeny (n = 67). These 67 species were involved in a total of 315 introduction events globally, 79.4% of which resulted in the alien population of the species becoming invasive according to DAMA 37 . The potential alien-native viral sharing comprised 3,153 unique pairs of 67 alien mammals and 1,090 native mammals that overlap in geographic areas where the alien species has been introduced. Predictions of novel sharing events were repeated for each final formulation of the models obtained during the nested cross validation routine to obtain estimates of uncertainty around predicted edges. We also obtained model predictions for alien-native pairs in which the sum of citations of each alien-native mammal pair was held at the median value to evaluate the effect of research effort on our predictions and the spatial patterns of predicted viral sharing events. Of all alien-native mammal pairs, 317 pairs (9.8%) already overlapped in their native ranges and shared at least one virus, while 343 pairs (10.9%) shared at least one virus without overlapping in their native ranges. Since the latter group was excluded from our training set (not having overlap), we selected it as an independent validation dataset for our viral sharing model. We then analysed SHapley Additive exPlanations (SHAP) using the kernelshap 49 package in R to unveil the contribution of each variable to the predicted viral sharing probability between any pair of overlapping alien and native mammals. SHAP values provide a tool for interpreting machine learning predictions by quantifying the marginal contribution of each feature to the predicted outcome for individual observations. We also used SHAP values to separately assess the average contribution of sampling effort to the predicted sharing probabilities associated with the different alien species and highlight potential knowledge gaps, both taxonomic and geographic, in the available host–virus association data. References Seebens, H. et al. No saturation in the accumulation of alien species worldwide. Nat. Commun. 8 , 14435 (2017). Roy, H. E., Pauchard, A., Stoett, P. & Renard Truong, T. IPBES Invasive Alien Species Assessment: Full Report . https://zenodo.org/doi/10.5281/zenodo.7430682 (2024) doi:10.5281/ZENODO.7430682. Pyšek, P. & Richardson, D. M. Invasive Species, Environmental Change and Management, and Health. Annu. Rev. Environ. Resour. 35 , 25–55 (2010). Torchin, M. E., Lafferty, K. D., Dobson, A. P., McKenzie, V. J. & Kuris, A. M. Introduced species and their missing parasites. Nature 421 , 628–630 (2003). Schatz, A. M. & Park, A. W. Patterns of host–parasite coinvasion promote enemy release and specialist parasite spillover. J. Anim. Ecol. 92 , 1029–1041 (2023). of the World’s Worst Invasive Alien Species: A Selection From The Global Invasive Species Database. in Encyclopedia of Biological Invasions (eds Simberloff, D. & Rejmanek, M.) 715–716 (University of California Press, 2019). doi:10.1525/9780520948433-159. Hatcher, M. J., Dick, J. T. A. & Dunn, A. M. Disease emergence and invasions. Funct. Ecol. 26 , 1275–1287 (2012). Strauss, A., White, A. & Boots, M. Invading with biological weapons: The importance of disease-mediated invasions. Funct. Ecol. 26 , (2012). Rushton, S. P. et al. Disease threats posed by alien species: the role of a poxvirus in the decline of the native red squirrel in Britain. Epidemiol. Infect. 134 , 521–533 (2006). Chinchio, E. et al. Invasive alien species and disease risk: An open challenge in public and animal health. PLOS Pathog. 16 , e1008922 (2020). Zhang, L. et al. Biological invasions facilitate zoonotic disease emergences. Nat. Commun. 13 , (2022). Roy, H. E. et al. Alien Pathogens on the Horizon: Opportunities for Predicting their Threat to Wildlife. Conserv. Lett. 10 , 477–484 (2017). Chinchio, E., Romeo, C., Crotta, M. & Ferrari, N. Knowledge gaps in invasive species infections: Alien mammals of European Union concern as a case study. Sci. Total Environ. 846 , (2022). Roy, H. E. et al. The role of invasive alien species in the emergence and spread of zoonoses. Biol. Invasions 25 , 1249–1264 (2023). Nugent, G., Buddle, B. & Knowles, G. Epidemiology and control of Mycobacterium bovis infection in brushtail possums ( Trichosurus vulpecula ), the primary wildlife host of bovine tuberculosis in New Zealand. N. Z. Vet. J. 63 , 28–41 (2015). Hulme, P. E. Invasive species challenge the global response to emerging diseases. Trends Parasitol. 30 , 267–270 (2014). Albery, G. F. et al. The science of the host–virus network. Nat. Microbiol. 6 , (2021). Davies, T. J. & Pedersen, A. B. Phylogeny and geography predict pathogen community similarity in wild primates and humans. Proc. R. Soc. B Biol. Sci. 275 , (2008). Huang, Z. Y. X. et al. Species’ Life-History Traits Explain Interspecific Variation in Reservoir Competence: A Possible Mechanism Underlying the Dilution Effect. PLoS ONE 8 , (2013). Cooper, N., Griffin, R., Franz, M., Omotayo, M. & Nunn, C. L. Phylogenetic host specificity and understanding parasite sharing in primates. Ecol. Lett. 15 , 1370–1377 (2012). Albery, G. F., Eskew, E. A., Ross, N. & Olival, K. J. Predicting the global mammalian viral sharing network using phylogeography. Nat. Commun. 11 , (2020). Nuñez, M. A., Pauchard, A. & Ricciardi, A. Invasion Science and the Global Spread of SARS-CoV-2. Trends Ecol. Evol. 35 , 642–645 (2020). Carlson, C. J. et al. The Global Virome in One Network (VIRION): an Atlas of Vertebrate-Virus Associations. mBio 13 , (2022). Rozemberczki, B. et al. The Shapley Value in Machine Learning. Preprint at https://doi.org/10.48550/arXiv.2202.05594 (2022). Huang, S., Bininda‐Emonds, O. R. P., Stephens, P. R., Gittleman, J. L. & Altizer, S. Phylogenetically related and ecologically similar carnivores harbour similar parasite assemblages. J. Anim. Ecol. 83 , 671–680 (2014). Stephens, P. R. et al. Parasite sharing in wild ungulates and their predators: Effects of phylogeny, range overlap, and trophic links. J. Anim. Ecol. 88 , 1017–1028 (2019). Schatz, A. M. & Park, A. W. Host and parasite traits predict cross-species parasite acquisition by introduced mammals. Proc. R. Soc. B Biol. Sci. 288 , rspb.2021.0341, 20210341 (2021). Longdon, B., Hadfield, J. D., Webster, C. L., Obbard, D. J. & Jiggins, F. M. Host Phylogeny Determines Viral Persistence and Replication in Novel Hosts. PLoS Pathog. 7 , e1002260 (2011). Tingley, R., Vallinoto, M., Sequeira, F. & Kearney, M. R. Realized niche shift during a global biological invasion. Proc. Natl. Acad. Sci. 111 , 10233–10238 (2014). Tedeschi, L. et al. Patterns and drivers of range filling of alien mammals in Europe. Oikos e11388 (2025) doi:10.1002/oik.11388. Gibb, R. et al. Mammal virus diversity estimates are unstable due to accelerating discovery effort. Biol. Lett. 18 , (2022). Hughes, A. C. et al. Sampling biases shape our view of the natural world. Ecography 44 , 1259–1269 (2021). García-Rodríguez, A. et al. The global distribution patterns of alien vertebrate richness in mountains. Nat. Commun. 16 , 1977 (2025). Capizzi, D., Monaco, A., Genovesi, P., Scalera, R. & Carnevali, L. Impact of alien mammals on human health. in Invasive species and human health 130–150 (2018). doi:10.1079/9781786390981.0130. Lefkowitz, E. J. et al. Virus taxonomy: the database of the International Committee on Taxonomy of Viruses (ICTV). Nucleic Acids Res. 46 , D708–D717 (2018). Lumbierres, M. et al. Area of Habitat maps for the world’s terrestrial birds and mammals. Sci. Data 9 , (2022). Biancolini, D. et al. DAMA: the global Distribution of Alien Mammals database. Ecology 102 , (2021). Gower, J. C. A General Coefficient of Similarity and Some of Its Properties. Biometrics 27 , 857–871 (1971). Laliberté, E. & Legendre, P. A distance-based framework for measuring functional diversity from multiple traits. Ecology 91 , 299–305 (2010). Soria, C. D., Pacifici, M., Di Marco, M., Stephen, S. M. & Rondinini, C. COMBINE: a coalesced mammal database of intrinsic and extrinsic traits. Ecology 102 , (2021). Faurby, S. et al. PHYLACINE 1.2: The Phylogenetic Atlas of Mammal Macroecology. Ecology 99 , (2018). Web of Science Platform | Clarivate. https://clarivate.com/academia-government/scientific-and-academic-research/research-discovery-and-referencing/web-of-science/. Wickham, H. & Wickham, M. H. Package ‘httr’. (2022). Ooms, J. The jsonlite package: A practical and consistent mapping between json data and r objects. ArXiv Prepr. ArXiv14032805 (2014). Chen, T. & Guestrin, C. XGBoost: A Scalable Tree Boosting System. in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 785–794 (Association for Computing Machinery, New York, NY, USA, 2016). doi:10.1145/2939672.2939785. Tonelli, A., Blagrove, M. S. C., Wardeh, M. & Di Marco, M. A framework to predict zoonotic hosts under data uncertainty: A case study on betacoronaviruses. Methods Ecol. Evol. (2025). Glidden, C. K. et al. Phylogenetic and biogeographical traits predict unrecognized hosts of zoonotic leishmaniasis. PLoS Negl. Trop. Dis. 17 , e0010879–e0010879 (2023). Kuhn, M. Package ‘tidymodels’: Easily Install and Load the ‘Tidymodels’ Packages. Cran (2020). Lundberg, S. & Lee, S.-I. A Unified Approach to Interpreting Model Predictions. Preprint at https://doi.org/10.48550/arXiv.1705.07874 (2017). Additional Declarations There is NO Competing Interest. Supplementary Files SIAliensharing.docx Supplementary Information for Alien mammal introductions can reshape global viral sharing networks Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7622438","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":522086067,"identity":"33a22001-b9dc-415e-900c-2976de5d0752","order_by":0,"name":"Andrea Tonelli","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBElEQVRIiWNgGAWjYHACAxiD8cADAxsGBmYwR4KBjYAWCRBxIMEgDabFgoENtx5kLQyHYaIVDLis4Z/dvO3DhxqGOn7p5gMHEgrOJ25n53346EaNBAOffANWLRJ3jhXPnHGMQUJyzrEEoMNuJ+5sZjc2zjkmgdthN3KMmXnYGCQMbuQYgLVsOMzGJp3DhluLPEjLn38gLfkfgFrOQbX8w63FAKSFsQ1sCyjEDkC05Lbh1mJ4I62YsbdPQnLmjDSQw5KNgVqYjXP7JHjY2BKwapG7kbyZ4cc3G35+ieSHDz78sZPdcP4Y4+Ocb3Vy8s0HcPgfEnCYQjz41I+CUTAKRsEowA8A+k1WtrTcx68AAAAASUVORK5CYII=","orcid":"","institution":"Sapienza University of Rome","correspondingAuthor":true,"prefix":"","firstName":"Andrea","middleName":"","lastName":"Tonelli","suffix":""},{"id":522086068,"identity":"b901c9fc-7993-4b96-b954-d8bb30fa455f","order_by":1,"name":"Gregory Albery","email":"","orcid":"https://orcid.org/0000-0001-6260-2662","institution":"Trinity College Dublin","correspondingAuthor":false,"prefix":"","firstName":"Gregory","middleName":"","lastName":"Albery","suffix":""},{"id":522086069,"identity":"3b769702-99f5-4d68-92a0-0cfa92a73fba","order_by":2,"name":"Moreno Di Marco","email":"","orcid":"https://orcid.org/0000-0002-8902-4193","institution":"Sapienza University of Rome","correspondingAuthor":false,"prefix":"","firstName":"Moreno","middleName":"Di","lastName":"Marco","suffix":""}],"badges":[],"createdAt":"2025-09-15 15:45:36","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7622438/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7622438/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92566136,"identity":"a922310d-bb5d-4327-ad74-e40d14fdd6dc","added_by":"auto","created_at":"2025-10-01 06:32:15","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7709593,"visible":true,"origin":"","legend":"","description":"","filename":"MainAliensharing.docx","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/e4f1a548568f0f80619bb08c.docx"},{"id":92565925,"identity":"1b67779c-d7f1-4dec-8ba8-aaab58bab710","added_by":"auto","created_at":"2025-10-01 06:24:15","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":342,"visible":true,"origin":"","legend":"","description":"","filename":"rs7622438.json","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/1cba16eb5b56f21780f73f26.json"},{"id":92566135,"identity":"2a505322-5f45-400f-9399-c86d7719ced6","added_by":"auto","created_at":"2025-10-01 06:32:15","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":97534,"visible":true,"origin":"","legend":"","description":"","filename":"rs76224382enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/84ba14c8f8770bcea0b3ae9b.xml"},{"id":92565930,"identity":"426c614d-4ce6-4fd6-9433-2158cade7cfb","added_by":"auto","created_at":"2025-10-01 06:24:15","extension":"jpeg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":215780,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/0fcdc04993647063db1c5734.jpeg"},{"id":92565933,"identity":"f836a8d9-2802-46b7-b275-da0e55954067","added_by":"auto","created_at":"2025-10-01 06:24:15","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3326785,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/08618abc6de3ea7e832f509a.png"},{"id":92565931,"identity":"4be50d0d-3f0a-467b-883c-6526be5ed12c","added_by":"auto","created_at":"2025-10-01 06:24:15","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":399604,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/4cc5f985e58bf1f6ce79502c.png"},{"id":92566962,"identity":"9559f457-c813-425a-adf2-1352d18755bf","added_by":"auto","created_at":"2025-10-01 06:40:15","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":168923,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/ce1f93d63ff9e87ea474d421.jpeg"},{"id":92566138,"identity":"7bd24e15-ab57-4faf-ad2b-ed728137e9d9","added_by":"auto","created_at":"2025-10-01 06:32:16","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":162174,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/7d286dd1b7e30fb0ff55731c.png"},{"id":92565941,"identity":"81568701-457b-4a34-988c-47b0b99d7097","added_by":"auto","created_at":"2025-10-01 06:24:16","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":151213,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/285c7deb383c7c3f9d82a6be.png"},{"id":92565937,"identity":"a55459c4-c4bf-4691-9e57-ae7b76c5d262","added_by":"auto","created_at":"2025-10-01 06:24:16","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":90776,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/53e9ae983e590490edd4bec9.png"},{"id":92565940,"identity":"4e636509-c499-4e0f-b280-c836074e54c8","added_by":"auto","created_at":"2025-10-01 06:24:16","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":127768,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/7fd7a0af3c4853468afa3368.png"},{"id":92566137,"identity":"5408970a-ebf5-41d1-b282-bed518ebd4b9","added_by":"auto","created_at":"2025-10-01 06:32:16","extension":"xml","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":96248,"visible":true,"origin":"","legend":"","description":"","filename":"rs76224382structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/1a45cb7c8f398f8c7bf1da95.xml"},{"id":92565936,"identity":"87c9a426-fc15-40d4-b99c-d0727d2580b2","added_by":"auto","created_at":"2025-10-01 06:24:15","extension":"html","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":104245,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/ddae8e76deb9d7437b11716f.html"},{"id":92565924,"identity":"9b2bddd0-c99d-408b-bc76-5f67202a71e5","added_by":"auto","created_at":"2025-10-01 06:24:15","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":682951,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSpatial and taxonomic distribution of 67 alien mammals introduced in the last 50 years. \u003c/strong\u003eThe maps show the locations of introduction events for the different mammalian orders. The lines connect the centroids of native (crosses) and alien (circles) ranges. Each circle denotes one introduction event (a). The barplot shows the number of alien species per mammalian family, colored by taxonomic order.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/3b6917a9f3c6a40b7309b701.png"},{"id":92565927,"identity":"91b242c7-89c9-4e50-8273-7698d0c0e9a7","added_by":"auto","created_at":"2025-10-01 06:24:15","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":471165,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePredicted surfaces of viral sharing probability for different combinations of predictor variables. \u003c/strong\u003eThe 3D surfaces show the probability of viral sharing in response to the interaction of predictor variables: phylogenetic similarity (a, c, e), geographic overlap (a, b, d), foraging similarity (d, e, f), trait similarity (b, c, f). To obtain these surfaces, we used the fully trained sharing model to predict viral sharing within a homogeneous surface of data points given by all combinations of x and y variables values while all other predictors were held at their median value. Brighter colors of the probability surface indicate increasing sharing probability. The floor of each plot displays data density across all combinations of predicted values, highlighting areas where model extrapolation is higher in lighter colors. An interactive version of the 3D plots is available as Supplementary files 1–6.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/22732713259de7a75547b5e8.png"},{"id":92565929,"identity":"19a07d52-a58b-4cd8-9a7d-52cb3e88ad51","added_by":"auto","created_at":"2025-10-01 06:24:15","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":307492,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePredicted sharing events driven by 67 alien species across the different biogeographic realms. \u003c/strong\u003eWe used a viral sharing model to predict potential novel alien-native viral sharing events across 3,153 combinations of sympatric alien and native mammals. The barplot shows the median number of novel viral sharing events for each alien species within its alien range across different biogeographic realms. Error bars indicate the 95% CI of the number of predicted sharing events obtained using a ≥0.5 threshold on the predicted sharing probability. The observed and predicted viral sharing links in alien ranges is shown in the viral sharing network. The alien species and their relative links are highlighted in red, while existing viral sharing links between native species are displayed in grey. The nodes that represent the ten species with the highest number of sharing events are circled in white and associated with a silhouette. All other nodes represent native species, with different colors based on the respective geographic realms to which they belong.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/ba5548aaf830d28ea4547a81.png"},{"id":92565926,"identity":"509fd98c-d160-4fa8-bb60-bdafef3f626e","added_by":"auto","created_at":"2025-10-01 06:24:15","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":496137,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGlobal distribution of predicted sharing events associated with individual alien mammal introductions. \u003c/strong\u003eThe map shows the distribution of viral sharing events between alien and native mammal species per introduction event. Lines connect the centroids of native (crosses) and alien (circles) ranges of 67 species that established alien populations outside of their native ranges. Circle colour indicates the number of predicted sharing events associated with each introduction event, with darker red denoting higher counts.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/ff1dad27cb3b7f5bb146fe92.png"},{"id":92567116,"identity":"ddb12ef5-239d-44e9-8668-0014cc94d807","added_by":"auto","created_at":"2025-10-01 06:48:17","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2734990,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/98b2163a-8d17-4b7e-9c49-e3cf82583681.pdf"},{"id":92565934,"identity":"cdb8d699-2fea-4e14-9960-cdd07bb85a27","added_by":"auto","created_at":"2025-10-01 06:24:15","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":4730363,"visible":true,"origin":"","legend":"Supplementary Information for Alien mammal introductions can reshape global viral sharing networks","description":"","filename":"SIAliensharing.docx","url":"https://assets-eu.researchsquare.com/files/rs-7622438/v1/8f4ec5ac413c50207fb582f2.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Alien mammal introductions can reshape global viral sharing networks","fulltext":[{"header":"Main","content":"\u003cp\u003eThe rate of alien species introduction has increased steadily over the past decades\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e, representing a threat to ecosystem stability and human well-being\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. Alien species introduced into new regions \u0026ndash; either deliberately or accidentally \u0026ndash; may carry novel pathogens into colonised environments\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e, potentially leading to spillover into na\u0026iuml;ve populations of native species, including humans. Approximately half of the vertebrates included in the list of the world\u0026rsquo;s worst invasive alien species\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e have been linked to wildlife diseases\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e, and pathogen-mediated competition is one of the ways by which alien species can displace native fauna\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. For instance, in the UK, native red squirrel populations (\u003cem\u003eSciurus vulgaris\u003c/em\u003e) declined up to 25 times faster when in contact with squirrelpox virus (SQPV)-seropositive alien grey squirrels (\u003cem\u003eSciurus carolinensis\u003c/em\u003e), compared to those exposed to seronegative populations\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Even when introduced species do not carry novel pathogens into the colonised environments, they may become infected with local pathogens and affect existing transmission networks, thereby altering pathogen dynamics within the local community\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. Through such direct and indirect pathways, biological invasions are increasingly recognised as drivers of emerging infectious diseases globally, including zoonoses of concern for human health\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. This evidence underscores the importance of assessing the disease-related impacts of alien species introductions for both biodiversity conservation and public health.\u003c/p\u003e\u003cp\u003eThe contribution of alien species to pathogen transmission dynamics, including those of public and animal health concern, is frequently underreported\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e and projections of emerging zoonoses often neglect the role of alien species as potential drivers of disease emergence and spread\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. This is in part due to incomplete data on pathogen distributions in wildlife and the inherent difficulty of predicting cascading effects following species introductions. Assessments of how alien species alter the prevalence and distribution of pathogens remain rare and typically limited to well-studied examples (e.g., brushtail possums as reservoirs of bovine tuberculosis in New Zealand\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e). As a result, the risk of pathogen spillover is seldom integrated into formal risk assessment frameworks for invasive alien species\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e, and quantitative tools to estimate the potential for pathogen transmission following alien species introductions are still lacking. Addressing this gap requires predictive approaches that can assess disease hazards and inform international policy on biological invasions\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eBuilding on existing theory on pathogen sharing networks\u003csup\u003e\u003cspan additionalcitationids=\"CR18 CR19\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e, we aim to identify quantifiable patterns that drive cross-species viral sharing between alien and native mammals. We used a viral sharing model\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e to predict the probability of viral sharing between alien and native mammal species following the establishment of alien populations introduced globally over the past 50 years (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). This temporal window allowed us to assess alien species that are already established in their non-native ranges, while remaining early enough to explore potential but still unaddressed viral transmission in local communities\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. We modelled viral sharing probability as a function of predictors that characterise species\u0026rsquo; pairwise similarity and eco-geographical opportunity for pathogen sharing, approximating the immunological and ecological barriers to cross-species transmission: phylogenetic distance, geographic overlap of suitable habitat, life-history trait similarity, and foraging similarity. We also accounted for levels of research efforts, to account for knowledge biases across different areas and different species. We predicted the probability of viral sharing across unique pairs of alien and native mammals that overlap in geographic areas where the alien species got established. We highlight that introduction of alien mammalian species can lead to hundreds of novel opportunities for cross-species viral transmission in alien ranges, especially between closely related host species with similar traits and dietary habits that share the same habitat. At the same time, we found research efforts to have an important effect on our predictions, highlighting potential knowledge gaps affecting wildlife disease research in the context of alien species introductions.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eWe used an extreme gradient boosting model to uncover viral sharing patterns within a sample of 1,116 mammal species which share at least one virus with co-occurring species. The model was trained on a sharing network made of 78,241 pairs of spatially overlapping mammals, derived from 1,920 mammal\u0026ndash;virus associations gathered from a published database of host virus associations\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e (see Methods). We then applied the viral sharing model to 67 species that established alien populations outside their native ranges in the last 50 years, in order to predict potential viral sharing between alien and native mammals following introduction.\u003c/p\u003e\u003cp\u003eOur viral sharing model achieved moderate performance during nested cross-validation (Supplementary Table\u0026nbsp;1), with a mean AUC of 0.83 (s.d. = 0.01) and a mean recall of 0.71 (s.d. = 0.01). When tested on an independent set of 343 native-alien pairs known to share at least one virus (see Methods) the model showed a mean out-of-sample recall of 0.71 (s.d. = 0.04), being able to correctly predict viral sharing for 244 of those pairs (median, 95% CI\u0026thinsp;=\u0026thinsp;235\u0026ndash;253).\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eGlobal drivers of cross-species viral sharing\u003c/h2\u003e\u003cp\u003eWe obtained variable importance scores during model training by estimating the average gain (improvement in accuracy) contributed by each predictor variable (Supplementary Fig.\u0026nbsp;1). Phylogenetic similarity between species emerged as the strongest determinant of viral sharing (median gain\u0026thinsp;=\u0026thinsp;0.37), followed by research effort, quantified as the sum of species-level citations of the pair. Foraging and trait similarities contributed moderately and at comparable levels, while the extent of geographic overlap had a relatively small influence on model performance.\u003c/p\u003e\u003cp\u003eWe explored patterns of viral sharing as a function of the interaction between pairs of predictor variables, holding all other variables constant (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; Supplementary Fig.\u0026nbsp;2). As expected, higher phylogenetic similarity (i.e., smaller patristic distance) strongly increased the probability of viral sharing across all combinations with the other predictors. Phylogenetic similarity and the extent of geographic overlap showed an additive positive effect, with sharing probability peaking at the highest phylogenetic similarity values (above 0.90, approximately the average similarity between congeneric mammals) and geographic overlap above 10\u0026nbsp;million Km\u003csup\u003e2\u003c/sup\u003e. Sharing probability showed a sharp increase around a phylogenetic similarity of ~\u0026thinsp;0.45, which roughly corresponds to the similarity between different bat families, beyond which sharing probability remained high, with geographic overlap contributing only a minor additive effect. Interestingly, viral sharing probability showed a spike at the lowest values of phylogenetic similarity (~\u0026thinsp;0.02, which roughly corresponds to the divergence of Xenarthra and Afrotheria from other mammalian lineages), where it increased evenly along increasing geographic overlap.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe interaction between foraging similarity and geographic overlap showed a mild additive trend, with sharing probability peaking at both high and low foraging similarity and around 10\u0026nbsp;million Km2 of geographic overlap, before declining in regions where the availability of training data was scarce. Foraging similarity had a limited contribution to sharing probability in combination with phylogenetic similarity, only showing a noticeable effect at low values of phylogenetic similarity, producing a spike in sharing probability (p\u0026thinsp;\u0026asymp;\u0026thinsp;0.6) at the lowest phylogenetic similarity values and high foraging similarity. Sharing probability increased with increasing trait similarity, and plateaued for lower values, generally showing minor effects in combination with other predictors.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eNovel viral sharing driven by the introduction of alien mammals\u003c/h3\u003e\n\u003cp\u003eWe built an alien-native network that included 3,153 unique pairs of alien and native mammals that overlapped geographically in areas where the alien species had been introduced within the past 50 years. Among the 67 alien species considered, the majority were even-toed ungulates (37.3%), followed by rodents (25.4%), carnivores (20.9%), insectivores (6.0%), lagomorphs (4.5%), pangolins, armadillos and primates (collectively 6.0%). Most species (58.2%) had multiple introduction events, both across different locations and at different points in time. The ten species with the highest number of introduction events accounted for approximately half of all introductions: \u003cem\u003eDama dama\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;65), \u003cem\u003eCervus nippon\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;45), \u003cem\u003eMyocastor coypus\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;29), \u003cem\u003eEutamias sibiricus\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;19), \u003cem\u003eNeovison vison\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;13), \u003cem\u003eOryctolagus cuniculus\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;11), \u003cem\u003eProcyon lotor\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;9), \u003cem\u003eCallosciurus erythraeus\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;7), \u003cem\u003eFunambulus pennantii\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;7), \u003cem\u003eOndatra zibethicus\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;7). Reported introduction events spanned all continents, with the highest concentrations in continental Europe (28.3%) and New Zealand (22.9%).\u003c/p\u003e\u003cp\u003eWe applied the viral sharing model to identify potential novel viral sharing events between 67 alien mammals and the 1,090 native mammals found in their alien ranges. The model predicted a median of 616 novel viral sharing events (95% CI\u0026thinsp;=\u0026thinsp;582\u0026ndash;650). The ten species with the highest predicted sharing accounted for approximately 62% of all novel viral sharing events (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). About 54% of novel viral sharing events were associated with alien species whose native ranges encompass multiple biogeographic realms, and about 55% involved species from the palearctic region.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWe used SHapley Additive exPlanations (SHAP) to assess the contribution of individual predictors to the probability of viral sharing between alien and native mammal species. SHAP values quantify the marginal contribution of each feature to the model\u0026rsquo;s output for each observation, offering an interpretable framework for understanding complex machine learning predictions\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. SHAP values largely confirmed the trends found for the patterns that drive viral sharing in native ranges, where phylogenetic similarity between species was the most important explainer of viral sharing (sum of |SHAP| = 253.72) once research effort was accounted for (Supplementary Fig.\u0026nbsp;3). Importantly, increasing values of geographic overlap, foraging and trait similarity gave important contributions to sharing probability even for distantly related species, sometimes showing non-linear effects with phylogenetic similarity (Supplementary Fig.\u0026nbsp;4).\u003c/p\u003e\u003cp\u003eWe predicted that each introduction event would generate, on average, approximately six sharing events with native mammals (95% CI\u0026thinsp;=\u0026thinsp;5.03\u0026ndash;6.98). As a comparison, mammals in our training dataset had, on average, approximately 27 observed sharing links (95% CI\u0026thinsp;=\u0026thinsp;24.51\u0026ndash;29.04) in their native ranges. Novel predicted sharing events cluster strongly in continental Europe and Japan, two regions with a high density of reported alien introductions, and include long-distance connections across multiple biogeographic realms, especially between Europe, Asia, and the Americas (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Consistently with the general trends of alien species introduction and reporting, the main hotspot of alien-driven viral sharing was continental Europe, with 201 novel sharing events between unique alien-native pairs and an average of approximately 11 sharing events per introduction. Other areas that were predicted as important hotspots of novel alien-native viral sharing were Japan with 72 events, and Africa with 51 events (42 of which were linked to the introduction of the Asian house rat, \u003cem\u003eRattus tanezumi\u003c/em\u003e). New Zealand was the landmass with the highest proportion of predicted sharing events on the number of alien-native encounters due to its low mammalian species richness. Other areas with a high proportion of realized viral sharing (i.e., the proportion of predicted alien-native viral sharing events on all alien-native encounters) included Australia and Japan, together with several smaller islands such as Bornholm, Green Island, La Palma, Lolland, Saint Martin, and Shetland (Supplementary Fig.\u0026nbsp;5).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003eThe effect of sampling bias\u003c/h3\u003e\n\u003cp\u003eWhen we repeated the predictions while holding the sum of citations for each alien\u0026ndash;native mammal pair at the median value, the predicted number of potential novel viral sharing events was 651 (median, 95% CI\u0026thinsp;=\u0026thinsp;606\u0026ndash;696). Under these conditions, the spatial distribution of predicted viral sharing events showed a general reduction of both the number (Supplementary Fig.\u0026nbsp;6) and the proportion of sharing events (Supplementary Fig.\u0026nbsp;7). These patterns were especially evident across Japan, New Zealand and North America. The analysis of SHAP values highlighted research effort as one of the strongest drivers of predicted viral sharing (Supplementary Fig.\u0026nbsp;8). Particularly, virological research effort was a very strong predictor for several well-studied species associated to adverse impact on conservation and/or human health, such as the raccoon (\u003cem\u003eProcyon lotor\u003c/em\u003e), the European rabbit (\u003cem\u003eOryctolagus cuniculus\u003c/em\u003e) and the European fallow deer (\u003cem\u003eDama dama\u003c/em\u003e).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe showed that the introduction of alien mammal populations beyond their native ranges can create novel opportunities for viral sharing with native species. By applying a trait- and phylogeny-informed viral sharing model, we demonstrated that the probability of cross-species viral transmission is shaped by evolutionary and ecological compatibility between alien and native hosts, extending the existing theory of pathogen sharing in the context of biological invasions. Consistently with previous studies on pathogen sharing\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e and parasite acquisition by alien species\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e, we found phylogenetic similarity to be the main driver of viral sharing. This pattern likely reflects conserved molecular and immunological traits between closely related species, which reduce cross-species barriers to viral transmission\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. For example, the Asian house rat (\u003cem\u003eRattus tanezumi\u003c/em\u003e), introduced to South Africa as a ship stowaway, was among the alien species with the highest number of potential viral sharing events. According to our model outputs, this was mostly driven by its close phylogenetic proximity to native African rodents and its strong ecological overlap with other small mammals in the region.\u003c/p\u003e\u003cp\u003eOur findings also underscore the importance of ecological similarity as a complementary predictor of mammal-mammal viral sharing. In cases where phylogenetic distance was high, similarities in diet and habitat use increased predictive accuracy, serving as important proxies for potential interspecific contact. Indeed, some viral sharing links were better explained by the extent of geographic overlap, foraging and trait similarity than by phylogenetic relatedness alone. This was the case for the North American raccoon (\u003cem\u003eProcyon lotor\u003c/em\u003e), a habitat and dietary generalist whose novel sharing events in Japanese islands were mostly driven by high geographic and ecological overlap with native mammals even when phylogenetic distance was high (such as with rodents and bats), once the effect of research effort was excluded. This shows the importance of considering ecological proxies of species interactions, such as similar habitat use and diet preferences, to gather a more comprehensive picture of mammal-mammal viral sharing potential.\u003c/p\u003e\u003cp\u003eOur results imply that novel alien-driven viral sharing can substantially reshape the structure of viral sharing networks in colonised areas worldwide. While our analysis focussed on predicting novel direct links between overlapping species in alien ranges, the ecological consequences are likely to extend far beyond individual species pairs. As host species are embedded within broader ecological networks, transmission can potentially go further than the single alien-native interaction, affecting a broader part of the local community. Furthermore, introduced species may undergo shifts in their realised or fundamental niches\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e, potentially leading to further adaptation and range expansion\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. As alien species expand their distributions, our analysis suggests that the probability of viral sharing with co-occurring species will increase, further affecting the cross-species transmission network within colonised ecosystems.\u003c/p\u003e\u003cp\u003eWhile phylogeography and ecological compatibility were key predictors of viral sharing, research effort (representing knowledge bias) gave an important contribution to model performance. Species with more known viral associations were more likely to be involved in predicted sharing events, consistently with biases in the known virome of mammals\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e that influenced both individual predictions and broader spatial patterns of viral sharing\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Most of the alien viral sharing events predicted by our model were localised in the Northern Hemisphere. This pattern likely reflects a dual bias: one in viral discovery efforts, and another in the underlying distribution of introduction records. The latter is consistent with broader patterns observed for terrestrial vertebrate records\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e and invasive species reporting\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. A previous assessment of invasive alien mammals impacting human health also detected a potential overrepresentation of wealthier countries, identifying European countries, Japan and Australia as the main global hotspots\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. By contrast, tropical regions \u0026ndash; where alien species detection capacity and management are limited \u0026ndash; may represent under-recognised hotspots of alien species introductions and cross-species transmission risk. Addressing these knowledge gaps across taxa and geographic areas could potentially reshape the perceived hazard of cross-species transmission driven by alien mammals at the global scale.\u003c/p\u003e\u003cp\u003eBiological invasions are inherently stochastic, and cross-species viral spillover events are even more so \u0026ndash; driven by a complex interplay of ecological, evolutionary, and environmental factors. Yet, analytical tools that integrate ecological and evolutionary insights into predictive frameworks may offer valuable information for assessing and anticipating disease risks associated with alien species introductions. Given the threats posed by emerging infectious diseases and the ongoing biodiversity crisis, these tools can support evidence-based decision-making at the interface of wildlife conservation and public health and facilitate the integration of disease hazard evaluations into international policy frameworks for invasive alien species.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eThe viral sharing network\u003c/h2\u003e\u003cp\u003eWe obtained host-virus associations for 1,249 mammalian species from the Vertebrate Virome in One Network (VIRION\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e). We discarded viral entities that were not ratified by the International Committee on the Taxonomy of Viruses (ICTV\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e), to avoid introducing potentially incorrect information and pseudoreplications. We followed Albery et al.\u003csup\u003e21\u003c/sup\u003e to obtain a unipartite mammal\u0026ndash;mammal network from the bipartite network of mammal\u0026ndash;virus associations. In this unipartite network, edges indicate whether a pair of mammalian species share at least one virus. Unobserved edges that involved species with at least one link were treated as pseudo-negatives and encoded as pseudo-absences (0s) during model training. This filtering step minimised bias from potentially undersampled species lacking observed viral sharing due to limited surveillance, thereby reducing overrepresentation of false negatives. Self connections and duplicate links were filtered out, and 49 marsupials and 1 monotremes were also excluded as they were extreme phylogenetic outliers. As we were interested in the patterns that drive viral sharing in the context of biological invasions, when the introduction of an alien species leads to novel mammal co-occurrences, we restricted the network to only include sympatric species pairs (i.e., those with overlapping geographic ranges). This left us with 1,116 mammal species for which complete data on biological traits, geographic ranges, and phylogeny were available, resulting in 78,241 unique mammal-mammal pairs. Of these, 23.8% shared at least one virus and 6.2% shared more than one virus.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003ePredictors of viral sharing\u003c/h3\u003e\n\u003cp\u003eWe retrieved the distribution of the 1,116 mammal species using Area of Habitat (AOH) maps\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. These maps provide more refined information on species presence than range polygons, by filtering out areas of the range that are not suitable for the species, thus reducing the risk of spurious geographic overlap due to overestimated range extents and suitability. We resampled AOH rasters (100m resolution) to a resolution of 5km with the \u003cem\u003eterra\u003c/em\u003e R package, using a sum of cell values aggregating function. Presence cells were retained using a\u0026thinsp;\u0026ge;\u0026thinsp;25% threshold of cell coverage, or the maximum coverage for species without any cell above the threshold. This value was selected as a compromise to avoid representing marginal species co-occurrence in a cell while avoiding penalising habitat specialists which might rarely have high coverage of suitable habitat in any cell. As a sensitivity analysis, we applied a\u0026thinsp;\u0026gt;\u0026thinsp;5% cell coverage threshold to a subset of 100 randomly sampled species, resulting in 32,697 overlaps with other mammals. Geographical overlap under this threshold was highly collinear with that obtained using the 25% threshold (Pearson\u0026rsquo;s r\u0026thinsp;\u0026gt;\u0026thinsp;0.99); this indicates that changing the threshold is unlikely to significantly impact our results. We then calculated the spatial overlap between ranges, quantified as the number of overlapping cells within the intersection of the AOHs. For alien species, the alien range polygons were obtained from the global Distribution of Alien Mammals database (DAMA\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e). The ranges of alien species that had multiple distinct introduction ranges in the selected time period were merged to obtain a single disjointed range. We converted the alien ranges into AOH maps, following the same procedure used in Lumbierres et al.\u003csup\u003e36\u003c/sup\u003e for full comparability with native AOH maps, and followed the same steps described above to obtain alien-native geographic overlaps.\u003c/p\u003e\u003cp\u003eFor each unique pair of mammals, we calculated biological traits and foraging dissimilarities between species using the Gower\u0026rsquo;s distance\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e, a metric used to measure dissimilarity in multidimensional trait space\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. Gower\u0026rsquo;s distance can handle continuous, categorical and ordinal variables (e.g., foraging classes), making it appropriate for our mixed dataset. Biological and foraging traits of the species were retrieved from the Coalesced Mammal Database of Intrinsic and Extrinsic traits (COMBINE\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e). Only traits with \u0026gt;\u0026thinsp;30% original data completeness and low imputation error (as reported in COMBINE) were selected. Biological trait dissimilarities were computed using body mass, longevity, age at first reproduction, gestation length, litter size, litters per year, and weaning age. Foraging dissimilarities were computed using trophic level, foraging stratum, and the percentages of the species\u0026rsquo; diet composed of vertebrates, invertebrates and plants. We also obtained pairwise patristic distances for all mammals from PHYLACINE\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. We then applied min\u0026ndash;max scaling to each distance metric, rescaling them to range from 0 to 1. Finally, we obtained biological, foraging, and phylogenetic similarity by subtracting 1 from the respective scaled distances. Data sources, description and rationale for each predictor is provided in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eDescription, rationale and data sources for the predictor variables used to model viral sharing.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariable\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDescription\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eRationale\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eData source\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGeographic overlap\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNumber of cells (resolution\u0026thinsp;=\u0026thinsp;25 Km\u003csup\u003e2\u003c/sup\u003e) in overlap within the intersection of ranges.\u003c/p\u003e\u003cp\u003e(Log\u003csub\u003e10\u003c/sub\u003e transformed)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eGreater geographic overlap provides more opportunities for viral sharing\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNative ranges:\u003c/p\u003e\u003cp\u003eAOH from \u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e;\u003c/p\u003e\u003cp\u003eInvasive ranges: AOH calculated from \u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePhylogenetic similarity\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eRelative phylogenetic similarity quantified as:\u003c/p\u003e\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:1-\\frac{patristic\\:distance}{max\\left(patristic\\:distance\\right)}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eClosely related species have similar physiology and immune systems which may make them susceptible to the same viruses\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003csup\u003e41\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eForaging similarity\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:1-\\frac{forag\\:dist\\:-\\:min\\left(forag\\:dist\\right)}{max\\left(forag\\:sit\\right)\\:-\\:min\\left(forag\\:dist\\right)}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003cp\u003eWhere \u003cem\u003eforag dist\u003c/em\u003e is the Gower distance between two species based on trophic level, foraging stratum, and the percentages of vertebrates, invertebrates and plants in the diet\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eHaving similar dietary niches may increase the ecological opportunities for viral sharing\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003csup\u003e40\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTrait\u003c/p\u003e\u003cp\u003esimilarity\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:1-\\frac{trait\\:dist\\:-\\:min\\left(trait\\:dist\\right)}{max\\left(trait\\:dist\\right)\\:-\\:min\\left(trait\\:dist\\right)}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003cp\u003eWhere \u003cem\u003etrait dist\u003c/em\u003e is the Gower distance between two species based on body mass, longevity, age at first reproduction, gestation length, litter size, litters per year, weaning age\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eLife-history traits correlate with determinants of pathogen susceptibility and exposure\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003csup\u003e40\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSum of citations\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSum of the number of virus-related papers for the two species. (Log\u003csub\u003e10\u003c/sub\u003e transformed)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAccounting for research effort\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003csup\u003e42\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eWe also controlled for bias in research efforts across species by collecting publication counts for each species. We obtained species-level virus-related paper counts to account for virological sampling effort in mammals, through an automated screening of Web of Science papers using the R packages \u003cem\u003ehttr\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e and \u003cem\u003ejsonlite\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. To account for variation in research effort, we summed paper counts for each mammal pair and applied a Log\u003csub\u003e10\u003c/sub\u003e transformation.\u003c/p\u003e\n\u003ch3\u003eModelling viral sharing probability\u003c/h3\u003e\n\u003cp\u003eWe trained an extreme gradient boosting model using the \u003cem\u003exgboost\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e algorithm in R, with binary viral sharing as the response variable. Predictor variables included geographic overlap (Log\u003csub\u003e10\u003c/sub\u003e transformed), phylogenetic similarity, trait similarity, foraging similarity and sum of citations (Log\u003csub\u003e10\u003c/sub\u003e transformed). Model training, tuning, and validation were conducted using a repeated nested cross-validation\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e,\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e with 10 inner folds for hyperparameter tuning and 10 outer folds for model evaluation. At each model iteration, the best combination of hyperparameters was selected based on the average recall calculated over the assessment inner folds. The best model was then evaluated against the outer test sets to estimate performance metrics. The whole process was repeated 10 times. The range of hyperparameter values selected during model tuning is provided in Supplementary Table\u0026nbsp;2. We extracted permutation-based variable importance from the trained model to identify the main determinants of viral sharing between mammals and visualised the pairwise interaction effect of predictors on sharing probability when all other predictors were held at their median values. The modeling workflow was implemented using the \u003cem\u003etidymodels\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e R package.\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003ePredicting viral sharing for alien-native pairs of mammals\u003c/h2\u003e\u003cp\u003eWe applied the fully trained viral sharing model to a set of selected alien species introduced in areas outside their native range in the last 50 years for which we had complete information on biological and foraging traits, geographic range and phylogeny (n\u0026thinsp;=\u0026thinsp;67). These 67 species were involved in a total of 315 introduction events globally, 79.4% of which resulted in the alien population of the species becoming invasive according to DAMA\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. The potential alien-native viral sharing comprised 3,153 unique pairs of 67 alien mammals and 1,090 native mammals that overlap in geographic areas where the alien species has been introduced. Predictions of novel sharing events were repeated for each final formulation of the models obtained during the nested cross validation routine to obtain estimates of uncertainty around predicted edges. We also obtained model predictions for alien-native pairs in which the sum of citations of each alien-native mammal pair was held at the median value to evaluate the effect of research effort on our predictions and the spatial patterns of predicted viral sharing events. Of all alien-native mammal pairs, 317 pairs (9.8%) already overlapped in their native ranges and shared at least one virus, while 343 pairs (10.9%) shared at least one virus without overlapping in their native ranges. Since the latter group was excluded from our training set (not having overlap), we selected it as an independent validation dataset for our viral sharing model.\u003c/p\u003e\u003cp\u003eWe then analysed SHapley Additive exPlanations (SHAP) using the \u003cem\u003ekernelshap\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e package in R to unveil the contribution of each variable to the predicted viral sharing probability between any pair of overlapping alien and native mammals. SHAP values provide a tool for interpreting machine learning predictions by quantifying the marginal contribution of each feature to the predicted outcome for individual observations. We also used SHAP values to separately assess the average contribution of sampling effort to the predicted sharing probabilities associated with the different alien species and highlight potential knowledge gaps, both taxonomic and geographic, in the available host\u0026ndash;virus association data.\u003c/p\u003e\u003c/div\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSeebens, H. \u003cem\u003eet al.\u003c/em\u003e No saturation in the accumulation of alien species worldwide. \u003cem\u003eNat. Commun. \u003c/em\u003e\u003cstrong\u003e8\u003c/strong\u003e, 14435 (2017). \u003c/li\u003e\n\u003cli\u003eRoy, H. E., Pauchard, A., Stoett, P. \u0026amp; Renard Truong, T. \u003cem\u003eIPBES Invasive Alien Species Assessment: Full Report\u003c/em\u003e. https://zenodo.org/doi/10.5281/zenodo.7430682 (2024) doi:10.5281/ZENODO.7430682. \u003c/li\u003e\n\u003cli\u003ePy\u0026scaron;ek, P. \u0026amp; Richardson, D. M. Invasive Species, Environmental Change and Management, and Health. \u003cem\u003eAnnu. Rev. Environ. Resour. \u003c/em\u003e\u003cstrong\u003e35\u003c/strong\u003e, 25\u0026ndash;55 (2010). \u003c/li\u003e\n\u003cli\u003eTorchin, M. E., Lafferty, K. D., Dobson, A. P., McKenzie, V. J. \u0026amp; Kuris, A. M. Introduced species and their missing parasites. \u003cem\u003eNature \u003c/em\u003e\u003cstrong\u003e421\u003c/strong\u003e, 628\u0026ndash;630 (2003). \u003c/li\u003e\n\u003cli\u003eSchatz, A. M. \u0026amp; Park, A. W. Patterns of host\u0026ndash;parasite coinvasion promote enemy release and specialist parasite spillover. \u003cem\u003eJ. Anim. Ecol. \u003c/em\u003e\u003cstrong\u003e92\u003c/strong\u003e, 1029\u0026ndash;1041 (2023). \u003c/li\u003e\n\u003cli\u003eof the World\u0026rsquo;s Worst Invasive Alien Species: A Selection From The Global Invasive Species Database. in \u003cem\u003eEncyclopedia of Biological Invasions\u003c/em\u003e (eds Simberloff, D. \u0026amp; Rejmanek, M.) 715\u0026ndash;716 (University of California Press, 2019). doi:10.1525/9780520948433-159. \u003c/li\u003e\n\u003cli\u003eHatcher, M. J., Dick, J. T. A. \u0026amp; Dunn, A. M. Disease emergence and invasions. \u003cem\u003eFunct. Ecol. \u003c/em\u003e\u003cstrong\u003e26\u003c/strong\u003e, 1275\u0026ndash;1287 (2012). \u003c/li\u003e\n\u003cli\u003eStrauss, A., White, A. \u0026amp; Boots, M. Invading with biological weapons: The importance of disease-mediated invasions. \u003cem\u003eFunct. Ecol. \u003c/em\u003e\u003cstrong\u003e26\u003c/strong\u003e, (2012). \u003c/li\u003e\n\u003cli\u003eRushton, S. P. \u003cem\u003eet al.\u003c/em\u003e Disease threats posed by alien species: the role of a poxvirus in the decline of the native red squirrel in Britain. \u003cem\u003eEpidemiol. Infect. \u003c/em\u003e\u003cstrong\u003e134\u003c/strong\u003e, 521\u0026ndash;533 (2006). \u003c/li\u003e\n\u003cli\u003eChinchio, E. \u003cem\u003eet al.\u003c/em\u003e Invasive alien species and disease risk: An open challenge in public and animal health. \u003cem\u003ePLOS Pathog. \u003c/em\u003e\u003cstrong\u003e16\u003c/strong\u003e, e1008922 (2020). \u003c/li\u003e\n\u003cli\u003eZhang, L. \u003cem\u003eet al.\u003c/em\u003e Biological invasions facilitate zoonotic disease emergences. \u003cem\u003eNat. Commun. \u003c/em\u003e\u003cstrong\u003e13\u003c/strong\u003e, (2022). \u003c/li\u003e\n\u003cli\u003eRoy, H. E. \u003cem\u003eet al.\u003c/em\u003e Alien Pathogens on the Horizon: Opportunities for Predicting their Threat to Wildlife. \u003cem\u003eConserv. Lett. \u003c/em\u003e\u003cstrong\u003e10\u003c/strong\u003e, 477\u0026ndash;484 (2017). \u003c/li\u003e\n\u003cli\u003eChinchio, E., Romeo, C., Crotta, M. \u0026amp; Ferrari, N. Knowledge gaps in invasive species infections: Alien mammals of European Union concern as a case study. \u003cem\u003eSci. Total Environ. \u003c/em\u003e\u003cstrong\u003e846\u003c/strong\u003e, (2022). \u003c/li\u003e\n\u003cli\u003eRoy, H. E. \u003cem\u003eet al.\u003c/em\u003e The role of invasive alien species in the emergence and spread of zoonoses. \u003cem\u003eBiol. Invasions \u003c/em\u003e\u003cstrong\u003e25\u003c/strong\u003e, 1249\u0026ndash;1264 (2023). \u003c/li\u003e\n\u003cli\u003eNugent, G., Buddle, B. \u0026amp; Knowles, G. Epidemiology and control of \u003cem\u003eMycobacterium bovis\u003c/em\u003e infection in brushtail possums ( \u003cem\u003eTrichosurus vulpecula\u003c/em\u003e ), the primary wildlife host of bovine tuberculosis in New Zealand. \u003cem\u003eN. Z. Vet. J. \u003c/em\u003e\u003cstrong\u003e63\u003c/strong\u003e, 28\u0026ndash;41 (2015). \u003c/li\u003e\n\u003cli\u003eHulme, P. E. Invasive species challenge the global response to emerging diseases. \u003cem\u003eTrends Parasitol. \u003c/em\u003e\u003cstrong\u003e30\u003c/strong\u003e, 267\u0026ndash;270 (2014). \u003c/li\u003e\n\u003cli\u003eAlbery, G. F. \u003cem\u003eet al.\u003c/em\u003e The science of the host\u0026ndash;virus network. \u003cem\u003eNat. Microbiol. \u003c/em\u003e\u003cstrong\u003e6\u003c/strong\u003e, (2021). \u003c/li\u003e\n\u003cli\u003eDavies, T. J. \u0026amp; Pedersen, A. B. Phylogeny and geography predict pathogen community similarity in wild primates and humans. \u003cem\u003eProc. R. Soc. B Biol. Sci. \u003c/em\u003e\u003cstrong\u003e275\u003c/strong\u003e, (2008). \u003c/li\u003e\n\u003cli\u003eHuang, Z. Y. X. \u003cem\u003eet al.\u003c/em\u003e Species\u0026rsquo; Life-History Traits Explain Interspecific Variation in Reservoir Competence: A Possible Mechanism Underlying the Dilution Effect. \u003cem\u003ePLoS ONE \u003c/em\u003e\u003cstrong\u003e8\u003c/strong\u003e, (2013). \u003c/li\u003e\n\u003cli\u003eCooper, N., Griffin, R., Franz, M., Omotayo, M. \u0026amp; Nunn, C. L. Phylogenetic host specificity and understanding parasite sharing in primates. \u003cem\u003eEcol. Lett. \u003c/em\u003e\u003cstrong\u003e15\u003c/strong\u003e, 1370\u0026ndash;1377 (2012). \u003c/li\u003e\n\u003cli\u003eAlbery, G. F., Eskew, E. A., Ross, N. \u0026amp; Olival, K. J. Predicting the global mammalian viral sharing network using phylogeography. \u003cem\u003eNat. Commun. \u003c/em\u003e\u003cstrong\u003e11\u003c/strong\u003e, (2020). \u003c/li\u003e\n\u003cli\u003eNu\u0026ntilde;ez, M. A., Pauchard, A. \u0026amp; Ricciardi, A. Invasion Science and the Global Spread of SARS-CoV-2. \u003cem\u003eTrends Ecol. Evol. \u003c/em\u003e\u003cstrong\u003e35\u003c/strong\u003e, 642\u0026ndash;645 (2020). \u003c/li\u003e\n\u003cli\u003eCarlson, C. J. \u003cem\u003eet al.\u003c/em\u003e The Global Virome in One Network (VIRION): an Atlas of Vertebrate-Virus Associations. \u003cem\u003emBio \u003c/em\u003e\u003cstrong\u003e13\u003c/strong\u003e, (2022). \u003c/li\u003e\n\u003cli\u003eRozemberczki, B. \u003cem\u003eet al.\u003c/em\u003e The Shapley Value in Machine Learning. Preprint at https://doi.org/10.48550/arXiv.2202.05594 (2022). \u003c/li\u003e\n\u003cli\u003eHuang, S., Bininda‐Emonds, O. R. P., Stephens, P. R., Gittleman, J. L. \u0026amp; Altizer, S. Phylogenetically related and ecologically similar carnivores harbour similar parasite assemblages. \u003cem\u003eJ. Anim. Ecol. \u003c/em\u003e\u003cstrong\u003e83\u003c/strong\u003e, 671\u0026ndash;680 (2014). \u003c/li\u003e\n\u003cli\u003eStephens, P. R. \u003cem\u003eet al.\u003c/em\u003e Parasite sharing in wild ungulates and their predators: Effects of phylogeny, range overlap, and trophic links. \u003cem\u003eJ. Anim. Ecol. \u003c/em\u003e\u003cstrong\u003e88\u003c/strong\u003e, 1017\u0026ndash;1028 (2019). \u003c/li\u003e\n\u003cli\u003eSchatz, A. M. \u0026amp; Park, A. W. Host and parasite traits predict cross-species parasite acquisition by introduced mammals. \u003cem\u003eProc. R. Soc. B Biol. Sci. \u003c/em\u003e\u003cstrong\u003e288\u003c/strong\u003e, rspb.2021.0341, 20210341 (2021). \u003c/li\u003e\n\u003cli\u003eLongdon, B., Hadfield, J. D., Webster, C. L., Obbard, D. J. \u0026amp; Jiggins, F. M. Host Phylogeny Determines Viral Persistence and Replication in Novel Hosts. \u003cem\u003ePLoS Pathog. \u003c/em\u003e\u003cstrong\u003e7\u003c/strong\u003e, e1002260 (2011). \u003c/li\u003e\n\u003cli\u003eTingley, R., Vallinoto, M., Sequeira, F. \u0026amp; Kearney, M. R. Realized niche shift during a global biological invasion. \u003cem\u003eProc. Natl. Acad. Sci. \u003c/em\u003e\u003cstrong\u003e111\u003c/strong\u003e, 10233\u0026ndash;10238 (2014). \u003c/li\u003e\n\u003cli\u003eTedeschi, L. \u003cem\u003eet al.\u003c/em\u003e Patterns and drivers of range filling of alien mammals in Europe. \u003cem\u003eOikos\u003c/em\u003e e11388 (2025) doi:10.1002/oik.11388. \u003c/li\u003e\n\u003cli\u003eGibb, R. \u003cem\u003eet al.\u003c/em\u003e Mammal virus diversity estimates are unstable due to accelerating discovery effort. \u003cem\u003eBiol. Lett. \u003c/em\u003e\u003cstrong\u003e18\u003c/strong\u003e, (2022). \u003c/li\u003e\n\u003cli\u003eHughes, A. C. \u003cem\u003eet al.\u003c/em\u003e Sampling biases shape our view of the natural world. \u003cem\u003eEcography \u003c/em\u003e\u003cstrong\u003e44\u003c/strong\u003e, 1259\u0026ndash;1269 (2021). \u003c/li\u003e\n\u003cli\u003eGarc\u0026iacute;a-Rodr\u0026iacute;guez, A. \u003cem\u003eet al.\u003c/em\u003e The global distribution patterns of alien vertebrate richness in mountains. \u003cem\u003eNat. Commun. \u003c/em\u003e\u003cstrong\u003e16\u003c/strong\u003e, 1977 (2025). \u003c/li\u003e\n\u003cli\u003eCapizzi, D., Monaco, A., Genovesi, P., Scalera, R. \u0026amp; Carnevali, L. Impact of alien mammals on human health. in \u003cem\u003eInvasive species and human health\u003c/em\u003e 130\u0026ndash;150 (2018). doi:10.1079/9781786390981.0130. \u003c/li\u003e\n\u003cli\u003eLefkowitz, E. J. \u003cem\u003eet al.\u003c/em\u003e Virus taxonomy: the database of the International Committee on Taxonomy of Viruses (ICTV). \u003cem\u003eNucleic Acids Res. \u003c/em\u003e\u003cstrong\u003e46\u003c/strong\u003e, D708\u0026ndash;D717 (2018). \u003c/li\u003e\n\u003cli\u003eLumbierres, M. \u003cem\u003eet al.\u003c/em\u003e Area of Habitat maps for the world\u0026rsquo;s terrestrial birds and mammals. \u003cem\u003eSci. Data \u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, (2022). \u003c/li\u003e\n\u003cli\u003eBiancolini, D. \u003cem\u003eet al.\u003c/em\u003e DAMA: the global Distribution of Alien Mammals database. \u003cem\u003eEcology \u003c/em\u003e\u003cstrong\u003e102\u003c/strong\u003e, (2021). \u003c/li\u003e\n\u003cli\u003eGower, J. C. A General Coefficient of Similarity and Some of Its Properties. \u003cem\u003eBiometrics \u003c/em\u003e\u003cstrong\u003e27\u003c/strong\u003e, 857\u0026ndash;871 (1971). \u003c/li\u003e\n\u003cli\u003eLalibert\u0026eacute;, E. \u0026amp; Legendre, P. A distance-based framework for measuring functional diversity from multiple traits. \u003cem\u003eEcology \u003c/em\u003e\u003cstrong\u003e91\u003c/strong\u003e, 299\u0026ndash;305 (2010). \u003c/li\u003e\n\u003cli\u003eSoria, C. D., Pacifici, M., Di Marco, M., Stephen, S. M. \u0026amp; Rondinini, C. COMBINE: a coalesced mammal database of intrinsic and extrinsic traits. \u003cem\u003eEcology \u003c/em\u003e\u003cstrong\u003e102\u003c/strong\u003e, (2021). \u003c/li\u003e\n\u003cli\u003eFaurby, S. \u003cem\u003eet al.\u003c/em\u003e PHYLACINE 1.2: The Phylogenetic Atlas of Mammal Macroecology. \u003cem\u003eEcology \u003c/em\u003e\u003cstrong\u003e99\u003c/strong\u003e, (2018). \u003c/li\u003e\n\u003cli\u003eWeb of Science Platform | Clarivate. https://clarivate.com/academia-government/scientific-and-academic-research/research-discovery-and-referencing/web-of-science/. \u003c/li\u003e\n\u003cli\u003eWickham, H. \u0026amp; Wickham, M. H. Package \u0026lsquo;httr\u0026rsquo;. (2022). \u003c/li\u003e\n\u003cli\u003eOoms, J. The jsonlite package: A practical and consistent mapping between json data and r objects. \u003cem\u003eArXiv Prepr. ArXiv14032805\u003c/em\u003e (2014). \u003c/li\u003e\n\u003cli\u003eChen, T. \u0026amp; Guestrin, C. XGBoost: A Scalable Tree Boosting System. in \u003cem\u003eProceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining\u003c/em\u003e 785\u0026ndash;794 (Association for Computing Machinery, New York, NY, USA, 2016). doi:10.1145/2939672.2939785. \u003c/li\u003e\n\u003cli\u003eTonelli, A., Blagrove, M. S. C., Wardeh, M. \u0026amp; Di Marco, M. A framework to predict zoonotic hosts under data uncertainty: A case study on betacoronaviruses. \u003cem\u003eMethods Ecol. Evol.\u003c/em\u003e (2025). \u003c/li\u003e\n\u003cli\u003eGlidden, C. K. \u003cem\u003eet al.\u003c/em\u003e Phylogenetic and biogeographical traits predict unrecognized hosts of zoonotic leishmaniasis. \u003cem\u003ePLoS Negl. Trop. Dis. \u003c/em\u003e\u003cstrong\u003e17\u003c/strong\u003e, e0010879\u0026ndash;e0010879 (2023). \u003c/li\u003e\n\u003cli\u003eKuhn, M. Package \u0026lsquo;tidymodels\u0026rsquo;: Easily Install and Load the \u0026lsquo;Tidymodels\u0026rsquo; Packages. \u003cem\u003eCran\u003c/em\u003e (2020). \u003c/li\u003e\n\u003cli\u003eLundberg, S. \u0026amp; Lee, S.-I. A Unified Approach to Interpreting Model Predictions. Preprint at https://doi.org/10.48550/arXiv.1705.07874 (2017). \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7622438/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7622438/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAlien mammals introduced beyond their native distribution ranges can bring novel pathogens into the colonised communities and alter pathogen transmission dynamics. Past studies identified immunological and ecological drivers of cross-species viral transmission, but this knowledge has rarely been applied to alien species, leading to an underestimation of their role in disease emergence. Here we predict the viral sharing network resulting from the establishment of 67 alien mammals introduced globally in the last 50 years, using a trait- and phylogeny-informed viral sharing model. We show that the introduction of alien mammals can result, on average, in six novel viral sharing events per introduction (95% CI\u0026thinsp;=\u0026thinsp;5.03\u0026ndash;6.98), potentially reshaping the viral sharing networks of local communities. Phylogenetic relatedness emerged as the strongest predictor of viral sharing between alien and native species, with additional contributions from trait-based, dietary, and habitat similarities. Predicted viral sharing was concentrated in the Global North, reflecting potential geographic biases in both introduction records and viral surveillance. Our approach provides a quantitative tool to estimate viral hazards driven by established alien species that can be used to support risk assessment frameworks and international policy on biological invasions.\u003c/p\u003e","manuscriptTitle":"Alien mammal introductions can reshape global viral sharing networks","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-01 06:24:10","doi":"10.21203/rs.3.rs-7622438/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3f749c5c-29ff-4374-a508-72810ad334c4","owner":[],"postedDate":"October 1st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":55486837,"name":"Biological sciences/Ecology/Ecological epidemiology"},{"id":55486838,"name":"Biological sciences/Ecology/Ecological networks"},{"id":55486839,"name":"Biological sciences/Ecology/Macroecology"}],"tags":[],"updatedAt":"2025-10-01T06:24:11+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-01 06:24:10","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7622438","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7622438","identity":"rs-7622438","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}