United States Swine Pathogen Database

United States Swine Pathogen Database integrates and curates nucleotide sequence data from clinical swine pathogen samples to support genomic surveillance and comparative analyses.

Key Features:

• Data Integration: Consolidates previously confidential clinical sequences with publicly available genomic data and integrates data mined from published literature, exemplified by porcine reproductive and respiratory syndrome virus (PRRSV).
• Technological Framework: Built on the Tripal toolkit leveraging the Drupal content management system and the Chado database schema.
• Cloud Hosting: Hosted on Amazon Web Services (AWS) EC2 cloud instances.
• Sequence Similarity Search: Supports sequence similarity searches using BLAST-based tools and exploration of annotated reference genomes.
• Detailed Annotation: Implements custom curation and annotation pipelines for PRRSV that determine genotype (Type 1 or 2), identify open reading frames and nonstructural proteins, generate amino acid sequences, detect putative frame shifts, and classify GP5 genes using restriction fragment length polymorphism (RFLP).
• Pathogen Coverage: Contains curated and annotated genomic data from seven major swine pathogens.
• Metadata Integration: Associates sequences with metadata including genomic information, collection date, location (state or province), and a unique identifier.

Scientific Applications:

• Vaccine Design: Public sequence data informs antigenic variation and supports design of field-relevant vaccines.
• Epidemiological Studies: Enables tracking of pathogen spread across geographical landscapes and identification of transmission hotspots within the United States and internationally.
• Pandemic Preparedness: Provides centralized genomic surveillance data to support preparedness and response efforts.

Methodology:

The database is implemented using the Tripal toolkit with Drupal and the Chado schema on AWS EC2, collects sequences from diagnostic laboratories with metadata (genomic information, collection date, location at state or province level, and a unique identifier), supports BLAST-based sequence similarity searches, and uses custom curation and annotation pipelines for PRRSV to determine genotype (Type 1 or 2), identify open reading frames and nonstructural proteins, generate amino acid sequences, detect putative frame shifts, and classify GP5 genes by RFLP.