Benchmark datasets for phylogenomic pipeline validation, applications for foodborne pathogen surveillance

Area of Science:

• Microbiology
• Bioinformatics
• Epidemiology

Background:

• Advances in next-generation sequencing necessitate validated genome-scale analysis programs for public health.
• Current phylogenomic methods often bypass traditional ortholog determination and multi-gene alignment.
• Public health requires robust datasets to validate these methods for applications like outbreak surveillance.

Purpose of the Study:

• To propose and provide benchmark datasets for the comparison and validation of phylogenomic pipelines.
• To establish standardized datasets for assessing the performance of evolutionary and epidemiological analyses.

Main Methods:

• Identified four well-documented foodborne pathogen outbreaks with concordant epidemiological and phylogenomic data.
• Compiled sequence data, metadata, and known phylogenetic trees for these events.
• Developed a standardized format for dataset description and an automated script for easy downloading from public databases.

Main Results:

• Created five benchmark datasets covering major foodborne bacterial pathogens (Listeria, Salmonella, E. coli, Campylobacter) and one simulated dataset.
• Made sequence data, metadata, and phylogenetic trees publicly accessible.
• Provided a GitHub repository with a downloading script and dataset information.

Conclusions:

• The benchmark datasets will standardize the comparison of current and future phylogenomic pipelines.
• Facilitates cross-institutional collaboration in sequence data analysis.
• Represents a step towards global standardization of phylogenomic pipelines and infrastructure.