Skip to main content
ARS Home » Plains Area » Manhattan, Kansas » Center for Grain and Animal Health Research » ABADRU » Research » Research Project #446117

Research Project: Using House Flies as Sentinels to Monitor Existing, Emerging and Cryptic Microbial Threats to U.S. Military Personnel

Location: Arthropod-borne Animal Diseases Research

Project Number: 3020-32000-018-021-I
Project Type: Interagency Reimbursable Agreement

Start Date: Apr 1, 2024
End Date: Aug 31, 2029

Objective 1: Identify existing, emerging, or hidden microbial threats to US Service members by exploring house fly microbial communities via the metagenome. Subobjective 1A: Determine the environmental prevalence, incidence, and frequency of priority gastrointestinal and ESKAPE pathogens through sentinel fly surveillance Subobjective 1B: Characterize and analyze microbe-associated genes of interest carried in house flies including known, emerging, and novel virulence and AMR genes. Objective 2: Generate risk maps indicating potential radius within which microbial threats are located Objective 2A: Mine house fly metagenomic data for nucleic acid signatures of microbial indicator species, hosts and environmental DNA Objective 2B: Correlate geographic location of house fly collection point to points of interest (potential sources of findings in #2A) using GIS and other mapping data Objective 2C: Determine potential dispersal routes from source of microbial threat(s) to collection points to create risk maps.

Study site is Fort Riley in Kansas near sources of flies (commissary, agriculture, barracks). This multi-year project will expand to OCONUS installations to construct predictive/forecasting and risk models that relate house fly surveillance data to outbreaks of disease in later years. The study assesses microbial threats to US military servicemembers and associated personnel by leveraging house fly biology (visiting microberich environmental substrates) as a tool for surveillance of existing, hidden and emerging microbial threats. At least 10 fly pools (n=5) will be collected per date/location to provide a comprehensive snapshot of the microbial threats in the environment. Flies will be collected using baits or by sweep netting, knocked down, pooled (n=5) processed in preservative (DNA/RNA shield) on site. Fly gut contents (FGC) will be expressed using pestle and carcass will be removed to reduce the amount of house fly nucleic acid. Samples will be stored at -80C until shipment to NECE for total nucleic acid (TNA) extraction (Kingfisher IndiMag kit) and sequencing (long read: Oxford Nanopore; short read: Illumina). Libraries will be prepared using a Ligation Kit (ONT R10.4.1) or Flex Kit (Illumina). ONT libraries will be run on FLO-MIN114. Illumina libraries will be run on v3 flow cells at a 2×250 bp read length. Sequences will be aligned to reference databases to identify pathogenic taxa and genes of interest (virulence, ARGs, etc.). QC will be conducted using trimmomatic for illumina data and NanoFilt for nanopore data. House fly reads will be removed in silico. De novo assembly of quality checked, de-hosted reads will be performed using FastQC, followed by extensive bioinformatics characterization (AMR Finder, FuncScan, etc.). Contigs will be subjected to homology searches via BLASTn and the nr nucleotide database. Contigs with hits that contain an E-score <1e-3 will be binned and examined for similarity to known microorganisms. The remaining contigs will be subjected to homology searches using BLASTx and Kraken2 on the nr protein database and SILVA respectively. Contigs with E-scores <1e-3 will be manually inspected for genetic similarity to known or closely related microorganisms. Other tools may be used to ID unassigned sequences. A report will be created containing the disease agents or genes of interest (e.g., AMR) found, their “criticality” (a measure of the prevalence of the corresponding indicator sequences found in collected samples, the distance of the closest sample containing the disease to a defined point, and the R0 of the disease agent, if known), whether the presence of relevant genes of interest were found in the sample, the distance of the closest sample with the disease agent, and likely environments of agent origin (determined by indicator species). This report will be generated from custom Python code and will help construct a RISK MAP. The prevalence, incidence and frequency of these microbial threats will be reported in the Year 1 study as well as risk maps around Ft Riley. In future years of the project, we will incorporate other biotic and abiotic variables to construct risk and forecasting models.