Viral competence data improves rodent reservoir predictions for American orthohantaviruses

Abstract

Identifying reservoir host species is crucial for understanding the risk of pathogen spillover from wildlife to people. Orthohantaviruses are zoonotic pathogens primarily carried by rodents that cause the diseases hemorrhagic fever with renal syndrome (HFRS) and hantavirus cardiopulmonary syndrome (HCPS) in humans. Given their diversity and abundance, many orthohantaviruses are expected to be undiscovered, and several host relationships remain unclear, particularly in the Americas. Despite the increasing use of predictive models for understanding zoonotic reservoirs, explicit comparisons between different evidence types for demonstrating host associations, and relevance to model performance in applied settings, have not been previously made. Using multiple machine learning methods, we identified phylogenetic patterns in and predicted unidentified reservoir hosts of New World orthohantaviruses based on evidence of infection (RT-PCR data) and competence (live virus isolation data). Infection data were driven by phylogeny, unlike competence data, and boosted regression tree (BRT) models using competence data displayed higher accuracy and a narrower list of predicted reservoirs than those using infection data. Eight species were identified by both BRT models as likely orthohantavirus hosts, with a total of 98 species identified by our infection models and 14 species identified by our competence models. Hosts predicted by competence models are concentrated in the northeastern United States (particularly Myodes gapperi and Reithrodontomys megalotis) and northern South America (several members of tribe Oryzomyini) and should be key targets for empirical monitoring. More broadly, these results demonstrate the value of infection competence data for predictive models of zoonotic pathogen hosts, which can be applied across a range of settings and host-pathogen systems.Author SummaryHuman diseases with wildlife origins constitute a significant risk for human health. Orthohantaviruses are viruses found primarily in rodents that cause disease with high rates of mortality and other complications in humans. An important step in disease prevention is to identify which rodent species carry and transmit orthohantaviruses. By incorporating species relatedness and evidence of different levels of host capacity to be infected and transmit virus, we used predictive modeling to determine unidentified rodent hosts of orthohantaviruses. Models using host competence data outperformed models using host infection data, highlighting the importance of stronger data in model optimization. Our results highlighted roughly a dozen key target species to be monitored that are concentrated in two geographic regions—northeastern United States and northern South America. More broadly, the approaches used in this study can be applied to a variety of other host-pathogen systems that threaten public health.

A Novel Dataset Towards Extracting Virus-Host Interactions

“…Distinguishing the confidence in host-pathogen data according to type of detection method has been shown to significantly improve models predicting zoonotic disease risk in rodents (Mull et al 2021). However, the information required to incorporate detection method as a variable in zoonotic disease risk models is rarely available from current host-pathogen databases or article metadata, with the important exception of (Olival et al, 2017), which we explore further here.…”