Paper detail

Predicting Novel Tick Vectors of Zoonotic Disease

With the resurgence of tick-borne diseases such as Lyme disease and the emergence of new pathogens such as Powassan virus, understanding what distinguishes vector from non-vector species, and predicting undiscovered tick vectors is an important step towards mitigating human disease risk. We apply generalized boosted regression to interrogate over 90 features for over 240 species of Ixodes ticks. Our model predicted vector status with ~97% accuracy and implicated 14 tick species whose intrinsic trait profiles confer high probabilities (~80%) that they are capable of transmitting infections from animal hosts to humans. Distinguishing characteristics of zoonotic tick vectors include several anatomical structures that facilitate efficient host seeking and blood-feeding from a wide variety of host species. Boosted regression analysis produced both actionable predictions to guide ongoing surveillance as well as testable hypotheses about the biological underpinnings of vectorial capacity across tick species.