Modeling Tick Populations: An Ecological Test Case for Gradient Boosted Trees

General linear models have been the foundational statistical framework used to discover the ecological processes that explain the distribution and abundance of natural populations. Analyses of the rapidly expanding cache of environmental and ecological data, however, require advanced statistical methods to contend with complexities inherent to extremely large natural data sets. Modern machine learning frameworks such as gradient boosted trees efficiently identify complex ecological relationships in massive data sets, which are expected to result in accurate predictions of the distribution and abundance of organisms in nature. However, rigorous assessments of the theoretical advantages of these methodologies on natural data sets are rare. Here we compare the abilities of gradient boosted and linear models to identify environmental features that explain observed variations in the distribution and abundance of blacklegged tick (Ixodes scapularis) populations in a data set collected across New York State over a ten-year period. The gradient boosted and linear models use similar environmental features to explain tick demography, although the gradient boosted models found non-linear relationships and interactions that are difficult to anticipate and often impractical to identify with a linear modeling framework. Further, the gradient boosted models predicted the distribution and abundance of ticks in years and areas beyond the training data with much greater accuracy than their linear model counterparts. The flexible gradient boosting framework also permitted additional model types that provide practical advantages for tick surveillance and public health. The results highlight the potential of gradient boosted models to discover novel ecological phenomena affecting pathogen demography and as a powerful public health tool to mitigate disease risks.

Associations of Anaplasma phagocytophilum Bacteria Variants in Ixodes scapularis Ticks and Humans, New York, USA

Anaplasmosis, caused by the tickborne bacterium Anaplasma phagocytophilum, is an emerging public health threat in the United States. In the northeastern United States, the blacklegged tick (Ixodes scapularis) transmits the human pathogenic genetic variant of A. phagocytophilum (Ap-ha) and a nonpathogenic variant (Ap-V1). New York has recently experienced a rapid and geographically focused increase in cases of anaplasmosis. We analyzed A. phagocytophilum-infected I. scapularis ticks collected across New York during 2008-2020 to differentiate between variants and calculate an entomological risk index (ERI) for each. Ap-ha ERI varied between regions and increased in all regions during the final years of the study. Space-time scan analyses detected expanding clusters of Ap-ha located within documented anaplasmosis hotspots. Ap-ha ERI was more positively correlated with anaplasmosis incidence than non-genotyped A. phagocytophilum ERI. Our findings help elucidate the relationship between the spatial ecology of A. phagocytophilum variants and anaplasmosis.

Phylogeographic dynamics of the arthropod vector, the blacklegged tick (Ixodes scapularis)

BACKGROUND

The emergence of vector-borne pathogens in novel geographic areas is regulated by the migration of their arthropod vectors. Blacklegged ticks (Ixodes scapularis) and the pathogens they vector, including the causative agents of Lyme disease, babesiosis and anaplasmosis, continue to grow in their population sizes and to expand in geographic range. Migration of this vector over the previous decades has been implicated as the cause of the re-emergence of the most prevalent infectious diseases in North America.

METHODS

We systematically collected ticks from across New York State (hereafter referred to as New York) from 2004 to 2017 as part of routine tick-borne pathogen surveillance in the state. This time frame corresponds with an increase in range and incidence of tick-borne diseases within New York. We randomly sampled ticks from this collection to explore the evolutionary history and population dynamics of I. scapularis. We sequenced the mitochondrial genomes of each tick to characterize their current and historical spatial genetic structure and population growth using phylogeographic methods.

RESULTS

We sequenced whole mitochondrial genomes from 277 ticks collected across New York between 2004 and 2017. We found evidence of population genetic structure at a broad geographic scale due to differences in the relative abundance, but not the composition, of haplotypes among sampled ticks. Ticks were often most closely related to ticks from the same and nearby collection sites. The data indicate that both short- and long-range migration events shape the population dynamics of blacklegged ticks in New York.

CONCLUSIONS

We detailed the population dynamics of the blacklegged tick (Ixodes scapularis) in New York during a time frame in which tick-borne diseases were increasing in range and incidence. Migration of ticks occurred at both coarse and fine scales in the recent past despite evidence of limits to gene flow. Past and current tick population dynamics have implications for further range expansion as habitat suitability for ticks changes due to global climate change. Analyses of mitochondrial genome sequencing data will expound upon previously identified drivers of tick presence and abundance as well as identify additional drivers. These data provide a foundation on which to generate testable hypotheses on the drivers of tick population dynamics occurring at finer scales.

Epidemiology and Spatial Emergence of Anaplasmosis, New York, USA, 2010‒2018

Human granulocytic anaplasmosis, a tickborne disease caused by the bacterium Anaplasma phagocytophilum, was first identified during 1994 and is now an emerging public health threat in the United States. New York state (NYS) has experienced a recent increase in the incidence of anaplasmosis. We analyzed human case surveillance and tick surveillance data collected by the NYS Department of Health for spatiotemporal patterns of disease emergence. We describe the epidemiology and growing incidence of anaplasmosis cases reported during 2010–2018. Spatial analysis showed an expanding hot spot of anaplasmosis in the Capital Region, where incidence increased >8-fold. The prevalence of A. phagocytophilum increased greatly within tick populations in the Capital Region over the same period, and entomologic risk factors were correlated with disease incidence at a local level. These results indicate that anaplasmosis is rapidly emerging in a geographically focused area of NYS, likely driven by localized changes in exposure risk.

Potential role of deer tick virus in Powassan encephalitis cases in Lyme disease-endemic areas of New York, U.S.A

TOC Summary: The epidemiologic pattern and limited laboratory testing indicate that this virus lineage might account for most of these illnesses.

Powassan virus, a member of the tick-borne encephalitis group of flaviviruses, encompasses 2 lineages with separate enzootic cycles. The prototype lineage of Powassan virus (POWV) is principally maintained between Ixodes cookei ticks and the groundhog (Marmota momax) or striped skunk (Mephitis mephitis), whereas the deer tick virus (DTV) lineage is believed to be maintained between Ixodes scapularis ticks and the white-footed mouse (Peromyscus leucopus). We report 14 cases of Powassan encephalitis from New York during 2004–2012. Ten (72%) of the patients were residents of the Lower Hudson Valley, a Lyme disease–endemic area in which I. scapularis ticks account for most human tick bites. This finding suggests that many of these cases were caused by DTV rather than POWV. In 2 patients, DTV infection was confirmed by genetic sequencing. As molecular testing becomes increasingly available, more cases of Powassan encephalitis may be determined to be attributable to the DTV lineage.

Geographical and environmental factors driving the increase in the Lyme disease vector Ixodes scapularis

The population densities of many organisms have changed dramatically in recent history. Increases in the population density of medically relevant organisms are of particular importance to public health as they are often correlated with the emergence of infectious diseases in human populations. Our aim is to delineate increases in density of a common disease vector in North America, the blacklegged tick, and to identify the environmental factors correlated with these population dynamics. Empirical data that capture the growth of a population are often necessary to identify environmental factors associated with these dynamics. We analyzed temporally- and spatially-structured field collected data in a geographical information systems framework to describe the population growth of blacklegged ticks (Ixodes scapularis) and to identify environmental and climatic factors correlated with these dynamics. The density of the ticks increased throughout the study's temporal and spatial ranges. Tick density increases were positively correlated with mild temperatures, low precipitation, low forest cover, and high urbanization. Importantly, models that accounted for these environmental factors accurately forecast future tick densities across the region. Tick density increased annually along the south-to-north gradient. These trends parallel the increases in human incidences of diseases commonly vectored by I. scapularis. For example, I. scapularis densities are correlated with human Lyme disease incidence, albeit in a non-linear manner that disappears at low tick densities, potentially indicating that a threshold tick density is needed to support epidemiologically-relevant levels of the Lyme disease bacterium. Our results demonstrate a connection between the biogeography of this species and public health.

Prevalence of Borrelia burgdorferi in small mammals in New York state

Intensive small mammal trapping was conducted in 12 counties in New York state during 1998-2000 to investigate the prevalence and site specificity of the Lyme disease spirochete, Borrelia burgdorferi in, and presence of the blacklegged tick, Ixodes scapularis Say on, the wild mice Peromyscus leucopus Rafinesque and Peromyscus maniculatus Wagner and other small mammal species. Previously captured mice (1992-1997) from throughout New York state also were recruited into the study, providing a total of 3,664 Peromyscus from 107 sites in 31 counties. Infection with B. burgdorferi was determined by polymerase chain reaction testing of ear tissue, and rates were determined by species, counties, and regions of the state. B. burgdorferi was detected in 10 small mammal species captured during 1998-2000. Peromyscus captured from Dutchess County in the lower Hudson Valley had the highest infection rate of 21%. The next highest infection rates were in counties within the Capital District: Albany (18%), Rensselaer (17%), and Columbia (13%). From 4,792 small animals examined, we recovered 2,073 ticks representing six species from 414 individuals of 12 mammal species, including 1,839 I. scapularis collected from 315 Peromyscus trapped in five counties. I. scapularis were most often collected from animals trapped in Albany, Rensselear, and Dutchess counties. We used protein electrophoresis of salivary amylase to distinguish between P. leucopus and P. maniculatus species. I. scapularis burdens were 5.7 ticks per P. leucopus and 14.3 ticks per P. maniculatus.