1. Given the public health, economic, and conservation implications of zoonotic diseases, their effective surveillance is of paramount importance. The traditional approach to estimating pathogen prevalence as the proportion of infected individuals in the population is biased because it fails to account for imperfect detection. A statistically robust way to reduce bias in prevalence estimates is to obtain repeated samples (or sample many tissues in multi-tissue disease systems) and to apply statistical methods that account for imperfect detection and permit the interdependence of the infection process across multiple tissues.

2. We developed a multi-state occupancy modeling framework which considers two scenarios about the infection process, one where no assumptions about the dependencies among the tissues are made (general), and another where dependence among tissues is not permitted (constrained).

3. We applied this model to pseudorabies virus (PrV) DNA detection data obtained from whole blood; and oral, nasal, and genital mucosa of 510 feral swine (Sus scrofa) during the years 2014-2016 in Florida, USA.

4. The constrained model was better supported by data. Estimated PrV prevalence varied among tissues, ranging from to 0.06 (CI: 0.02-0.14) in genital to 0.54 (CI: 0.14-0.82) in nasal tissue. Probability of PrV detection ranged from 0.11 (CI: 0.06-0.18) in nasal to 0.51 (CI: 0.21-0.81) in genital tissue. Estimates of PrV prevalence after accounting for imperfect detection were higher than the naïve estimates for all four tissues.

5. PrV prevalence was not affected by the age or sex of the animal or the year of sampling, but prevalence increased as drought severity increased.

6. The conditional probability of detecting PrV given infection in at least one tissue type within an individual was highest for nasal tissue, suggesting that nasal is the best tissue to sample for PrV surveillance if only one tissue can be sampled, at least for systems with tissue-specific prevalence and detection probabilities similar to ours.

7. We found that pathogen prevalence in multi-tissue disease systems can vary across tissues. Our results emphasize the importance of sampling multiple tissues, and the application of robust statistical models to account for imperfect detection in the surveillance of systemic diseases. The multi-state modeling framework is broadly applicable to the surveillance of pathogens that infect multiple tissues and where the infection status or detection of the pathogen in one tissue may depend on the infection status of the pathogen in other tissues). 29-Jul-2020

 TISSUE SAMPLE COLLECTION 

Between January 2014 and March 2016, we opportunistically sampled 549 wild pigs at 39 sites across Florida (Figure 1 in the main text) as part of national feral swine disease monitoring effort led by the United States Department of Agriculture, Animal Plant and Health Inspection Service, United States Department of Agriculture Wildlife Services National Wildlife Disease Program. Swine were either euthanized and sampled immediately during animal-control efforts or were hunted by hunters on federal and state wildlife management areas, military bases and private properties. No animals were killed for the express purpose of this study. The sampling protocol for this study was approved by University of Florida’s Institutional Animal Care and Use Committee. We collected whole blood and nasal, oral and genital swabs from the sampled swine. We collected the following information for each individual: age category (adult: > 1 year; sub-adult: 2 months to 1 year; and juvenile: < 2 months); sex (male or female) and GPS location. We determined age using tooth-eruption, body-size and characteristics of external reproductive organs (Matschke, 1967). Genital swabs were collected only from females. Data on age and sex were available for 510 pigs; therefore, all analyses testing for the influence of these covariates were conducted using only data from those individuals. Oral, blood, nasal and genital tissue samples were collected from 408, 439, 497, and 196 individuals, respectively. 

 MOLECULAR METHODS AND PrV DETECTION

Whole blood (0.5 mL), and nasal, oral and genital swabs were placed in 1 mL mammalian lysis buffer (Qiagnen, Valencia, California, USA) and were refrigerated at 4⁰ C after collection. These samples were transferred to University of Florida (Gainesville, Florida, USA) and stored at -80⁰ C until DNA was extracted. We extracted DNA from blood and swabs using Qiagen DNeasy Blood and Tissue Kit (Qiagen) following the manufacturer’s instructions. The concentration of recovered nucleic acid was quantified using Epoch Microplate Spectrophotometer running the Gen5 software, version 2.09 (BioTek Instruments, Winooski, Vermont, USA). We stored recovered DNA at -20⁰ C until further processing. We used primers and probes targeting the 5^’ coding region of the PrV glycoprotein B (gB) gene (UL27) in order to detect PrV DNA in various sample types (Sayler et al. 2017). We controlled for false negative diagnostic error caused by failure of individual PCR assay to run by using a commercially available nucleic acid internal control (VetMAX Xeno Internal Positive Control DNA, Applied Biosystems, Foster City, California, USA). Molecular grade water was used as a negative control and extraction control (i.e., no template controls) to detect false positives due to contamination. We performed PrV-gB qPCR assays on the ABI 7500 fast thermocycler by using Brilliant III Ultra-Fast qPCR Master Mix (Agilent, Santa Clara, California, USA) with 2 μL of template DNA, 0.4 μL of PrV-gB forward primer at 20 μM, 0.4 μL of PrV-gB reverse primer at 20 μM, and 0.4 μL of PrV-probe at 10 μM. We set the cycling conditions as: 95⁰ C for 3 minutes followed by 40 cycles of 95⁰ C for 15 seconds and 60⁰ C for 30 seconds. Quantification cycle (Cq) values >35 were considered a negative result (see Hernández et al. 2018 for additional details on qPCR techniques). Each DNA sample was amplified using qPCR assays a maximum of three times per tissue type per individual. 

This PrVreadme.txt file was generated on 2020-08-21 by Vratika Chaudhary

GENERAL INFORMATION

1. Title of Dataset: Pseudorabies virus (PrV) disease detection in feral swine (Sus scrofa) in Florida.

2. Author Information
     Corresponding author Contact Information
        Name: Vratika Chaudhary
        Institution: University of Florida
        Email: Chaudhary.vratika@gmail.com

3. Date of data collection: 2014-2016

4. Geographic location of data collection : Florida, USA

SHARING/ACCESS INFORMATION

1. Licenses/restrictions placed on the data: This work is licensed under a CC0 1.0 Universal (CC0 1.0) Public Domain Dedication license.

2. Links to publications that cite or use the data: Chaudhary, Vratika et al. (2020), A multi-state occupancy modeling framework for robust estimation of disease prevalence in multi-tissue disease systems, v2, Dryad, Dataset, https://doi.org/10.5061/dryad.z8w9ghx94

3. Links to other publicly accessible locations of the data: none

4. Links/relationships to ancillary data sets: none

5. Was data derived from another source? yes/no
    A. If yes, list source(s): no

6. Recommended citation for this dataset: Chaudhary, V., Wisely, S.M., Hernández, F.A., Hines, J.E., Nichols, J.D. and Oli, M.K. (2020), A multi‐state occupancy modelling framework for robust estimation of disease prevalence in multi‐tissue disease systems. J Appl Ecol. Accepted Author Manuscript. doi:10.1111/1365-2664.13744

DATA & FILE OVERVIEW

1. File List: PrV_detection_data.csv

2. Relationship between files, if important: none

3. Additional related data collected that was not included in the current data package: none

4. Are there multiple versions of the dataset? yes/no: no
    

METHODOLOGICAL INFORMATION

1. Description of methods used for collection/generation of data: Please see:
Chaudhary, V., Wisely, S.M., Hernández, F.A., Hines, J.E., Nichols, J.D. and Oli, M.K. (2020), A multi‐state occupancy modelling framework for robust estimation of disease prevalence in multi‐tissue disease systems. J Appl Ecol. Accepted Author Manuscript. doi:10.1111/1365-2664.13744

DATA-SPECIFIC INFORMATION FOR:  PrV_detection_data.csv

1. Number of variables: 16

2. Number of cases/rows: 510

3. Variable List: 
i. Sex: sex of the individual (Male (M) or Female (F)).
ii. Age: age of the animal (Adult (A), Juvenile (J), Subadult (SA)).
iii. Year: year of the sample collection (2014, 2015, 2016).
iv. pdsi: The Palmer Drought Severity Index (PDSI) uses readily available temperature and precipitation data to estimate relative dryness. It is a standardized index that generally spans -10 (dry) to +10 (wet) (https://climatedataguide.ucar.edu/climate-data/palmer-drought-severity-index-pdsi).
v. sampling1.oral1, sampling2.oral1, sampling3.oral1: virus was detected (1) or not detected (0) in the animal using qPCR in oral tissue.  
vi. sampling1.nasal1, sampling2.nasal1, sampling3.nasal1: virus was detected (1) or not detected (0) in the animal using qPCR in nasal tissue. 
vii. sampling1.genital1, sampling2.genital1, sampling3.nasal1: virus was detected (1) or not detected (0) in the animal using qPCR in genital tissue. 
viii. sampling1.blood1, sampling2.blood1, sampling3.blood1: virus was detected (1) or not detected (0) in the animal using qPCR in blood tissue
4. Missing data codes: 'NA'. The tissue was not tested.
5. Specialized formats or other abbreviations used: 0 (PrV virus was not detected), 1(PrV virus was detected), NA(tissue was not tested). 
 

Code to implement the modelling framework on the data can be accessed through: https://github.com/vratchaudhary/Multi-state-occupacy-model-for-systemic-disease