ProkEvo

ProkEvo performs automated, scalable analyses of bacterial Whole Genome Sequence (WGS) data starting from raw Illumina paired-end reads to characterize population structure, genotype frequencies, antimicrobial resistance (AMR) genes, virulence factors, plasmids, and pan-genome content for ecological, diagnostic, and epidemiological investigations.

Key Features:

• Automation and Scalability: Automates complex combinations of computational analyses for thousands of bacterial genomes starting from raw Illumina paired-end sequence reads.
• Reproducibility and Modularity: Leverages the Pegasus Workflow Management System (WMS) to provide reproducibility, modularity, fault-tolerance, scalability, and robust file management.
• High-Performance Computing: Operates on high-performance and high-throughput computational platforms and has been validated with datasets of approximately 2,400 to 23,000 genomes.
• Hierarchical Population Structure Analysis: Employs hierarchical-based population structure analyses using multi-locus and Bayesian statistical approaches for detailed classification.
• Integration with Curated Databases: Associates antimicrobial resistance (AMR) genes, putative virulence factors, and plasmids from curated databases with genotypic classifications.
• Pan-Genome Annotations: Generates comprehensive pan-genome annotations and compiles data for downstream analyses, including identification of population-specific genomic signatures.

Scientific Applications:

• Basic Microbiological Research: Provides population-scale WGS analyses that inform microbial ecology and evolutionary studies.
• Clinical Diagnostics: Links genomic data to AMR genes and virulence factors to support clinical microbiology diagnostics.
• Epidemiological Surveillance: Enables tracking of pathogen spread and informs public health interventions via hierarchical population structure analyses and genotype frequency data.

Methodology:

Processes raw Illumina paired-end reads through automated workflows managed by Pegasus WMS; performs hierarchical population structure analyses using multi-locus and Bayesian statistical approaches; associates AMR genes, putative virulence factors, and plasmids via curated databases; generates pan-genome annotations; demonstrated on datasets of ~2,400–23,000 genomes on high-performance/high-throughput computing platforms.