In this post, Qixin He, an assistant professor in the College of Science and member of the Purdue Institute for a Sustainable Future, discusses her recently published research “Predicting host range expansion in parasitic mites using a global mammalian-acarine dataset” which appears in Nature Communications.
What did you want to know?
Mites associated with mammals, including humans, are ubiquitous and megadiverse. Their relationships with hosts range from being extremely virulent parasites to relatively benign commensals. Among them, more virulent parasites usually infect multiple host species. In contrast, single-host parasites are less virulent, but a small fraction of them has the potential to host-shift and become virulent. The mite’s ability to infect one or several hosts may depend on a set of factors related to mite or host traits, mite-host and host-host interactions, climate, or habitat disturbance by humans.
Here, we aimed to build a statistical model that can predict and forecast the parasites’ host range given our data from a comprehensive worldwide survey and a set of variables related to both biotic (mite or host traits) and abiotic factors (climate). From the model output, we focused on predicting whether single-host parasites have the potential to expand their host ranges and become multi-host parasites. This category of parasites is particularly dangerous because they could, in the future, shift to immunologically naïve hosts and cause epidemics. This category is also notoriously difficult to predict due to technical constraints in machine learning. Furthermore, some of the parasite species could have been labeled as single-host species because they have received little research attention, posing additional challenges for model building.
What did you achieve?
Our global survey yielded a comprehensive database of all mite-mammal associations known to date, 1,984 mite and 1,432 mammal species. This significantly expands our knowledge in the field. For example, our database reports 77% more species parasitizing primates compared to previous efforts.
Using this large dataset, we developed a model to predict whether single-host parasites can become multi-host by analyzing 13 variables. Our model identified several key factors influencing host utilization by mites. Notably, we found that multi-host parasites tend to have feeding styles that minimize contact with the host’s immune system. Additionally, closely related hosts, especially those in geographic proximity, significantly increase the likelihood of mite parasite sharing.
When applied for forecasting, our model also identified risk-group species currently reported as single-host by researchers but with the potential to become multi-host. Most risk-group parasites are associated with rodents, bats, and carnivores, which serve as key parasite reservoirs, particularly in tropical regions. For example, five sarcoptid skin mites of the genus Notoedres, parasitizing tropical bats, are currently listed in the literature as single-host parasites. However, our model predicted, with a high probability, that these mites could become multihost. Interestingly, macroevolutionary evidence suggests that mites of this genus experienced ancient host shifts to unrelated hosts, such as rodents and cats, long before humans appeared. This indicates that similar host shifts could still occur today.
What is the impact of this research?
Our model helps to identify single-host parasites that can become multi-host and potentially dangerous parasites. One notable example of existing dangerous multi-host parasites is the human itch mite, which burrows in the skin, causing sarcoptic mange and, in severe cases, a condition known as crusted scabies. At least 200 million people worldwide suffer from scabies, mostly in developing countries. This mite primarily attacks humans but it can also shift to wild or domesticated animals causing much greater damage since their immune system is “naïve” to the parasite.
Our model also predicts the largest reservoirs of species among mammals that are likely to source new host shifts or spillover to human and domesticated animals. We estimated the potential danger of host shifts in little-known parasites of bats, rodents, and other mammalian groups in the tropics. Such a host-shift scenario is likely given that tropical mammals often serve as pathogen reservoirs, and this risk is also compounded by the increasing rate of anthropogenic habitat disturbance in these areas, which facilitates pathogen spillover.
Our model uses a relatively small set of predictors to identify risk-group parasites, which makes it useful in practice.