A Schema for Digitized Surface Swab Site Metadata in Open-Source DNA Sequence Databases

Area of Science:

• Microbiology and Bioinformatics
• Genomic Epidemiology
• Public Health Surveillance

Background:

• Large-scale DNA sequence databases are crucial for public health surveillance of microbial pathogens.
• Current metadata collection for swab site locations is unstructured, leading to poor data quality and hindering automated analysis.
• Inconsistent free-text descriptions in the 'isolation source' field impede machine-actionability and data interoperability.

Purpose of the Study:

• To develop a standardized, hierarchical vocabulary for describing microbial pathogen swab site locations.
• To improve the machine-actionability and quality of metadata in public DNA sequence databases.
• To enhance the utility of genomic data for public health surveillance and food safety.

Main Methods:

• Content analysis of 1,498 free-text swab site descriptions from foodborne pathogen surveillance.
• Evaluation of metadata lexicon to identify informational facets and unique terms.
• Development of hierarchical vocabularies and logical relationships using Open Biological Ontologies (OBO) Foundry libraries.

Main Results:

• Identified 5 key informational facets and 338 unique terms within the free-text metadata.
• Developed a structured schema with hierarchical vocabularies and logical axioms for swab site locations.
• Integrated the developed schema into a publicly available pathogen metadata standard (One Health Enteric Package).

Conclusions:

• Standardized metadata, particularly for swab site locations, is essential for improving the quality and interoperability of DNA sequence databases.
• The developed vocabulary system facilitates large-scale data sharing, artificial intelligence, and big-data solutions for food safety.
• Implementation of metadata standards supports the future of open genomic epidemiology networks and efficient public health investigations.