This specification comprises an international minimal metadata standard for public health and One Health genomic surveillance, reusable across pathogen and project initiatives globally, for use in single isolate as well as metagenomic sequencing.
This work has been developed by a group of participants at the 9th Microbes and Food Safety Bioinformatics Hackathon held in Cambridge, UK in May 2023.
Metadata is crucial for effective genomic surveillance in public health as it provides valuable information about the samples being analyzed, and it allows for data sharing and interoperability between different organizations and jurisdictions. However, there are a few challenges to overcome, such as combining data from different sources, establishing trust in data sharing, and dealing with data standards which may not be designed for public health. Moreover, data integration needs to be timely, as public health emergencies happen in a quicker timescale than academic research work, and metadata is collected and held by different organizations with varying levels of granularity.
The aim of this work is to present a framework that can be applied across public health surveillance, regardless of the pathogen or sequence type. The framework emphasizes the importance of well-documented and stored sequences and their associated metadata, and seeks to make it easier for decision makers and data management personnel to understand the benefits of such documentation.
Table 1: Minimal contextual data requirements for public health and One Health surveillance
| Field | Ontology ID | Definition | Guidance | Expected Value Type | Benefits | Considerations and Privacy |
|---|---|---|---|---|---|---|
| specimen collector sample ID | GENEPIO:0001123 | The user-defined name for the sample. | Original ID given to the sample at the primary laboratory | Alpha-numeric value | Linkage of sample with genomic, lab and epi data; Identifier tracking and establishing chain of custody; Can be considered Personal Health Identifiable (PHI) information | Public facing identifier can be used as replacement when sharing (such as a hash ID) |
| specimen source ID | GENEPIO:0100696 | The identifier assigned to the host or environment from which the specimen was derived (e.g. host subject ID, case ID, environmental site ID) | Provide the identifier for the source of the sample such as the case ID or site ID. | Alpha-numeric value | Allow for deduplication of cases; Traceability | Can be considered Personal Health Identifiable (PHI) information; Public facing identifier can be used as replacement when sharing (such as a hash ID) |
| geo_loc name (country) | GENEPIO:0001181 | The country where the sample was collected. | Use an ontology such as the Gazateer (GAZ). Equivalent to ISO3166-1 | Permitted values | Spatio-temporal analysis; International Health Regulations reporting requirements | If event counts in a country are low, perturbation may be required |
| geo_loc name (state/province/region) | GENEPIO:0001185 | The state/province/region where the sample was collected. | Use an ontology such as the Gazateer ontology (GAZ). Equivalent to ISO3166-2 | Permitted values | Spatio-temporal analysis | If event counts in a state/province/region are low, perturbation may be required |
| sample collected by | GENEPIO:0001153 | The name of the agency that collected the original sample. | The official organization name should be used (avoid abbreviations, alternate names as much as possible). | Auto complete controlled list | Recognition; Provenance; Contact for follow-up | Care should be taken with regards to potential geographical identification associated with organizational name. |
| sequenced by | GENEPIO:0100416 | The name of the agency that generated the sequence. | The official organization name should be used (avoid abbreviations, alternate names as much as possible). | Auto complete controlled list | Recognition; Provenance; Contact for follow-up | Negligible privacy concerns |
| sequence submitted by | GENEPIO:0001159 | The name of the agency that submitted the sequence to a database. | The official organization name should be used (avoid abbreviations, alternate names as much as possible). | Auto complete controlled list | Recognition; Provenance; Contact for follow-up | Negligible privacy concerns |
| data custodian organization name | GENEPIO:0100695 | The name of the organization with authority over how the sample and its associated information can be used. | The official organization name should be used (avoid abbreviations, alternate names as much as possible). | Auto complete controlled list | Jurisdictional authority over data use; Decision-making; Contact for follow-up | Care should be taken with regards to potential geographical identification associated with organizational name. |
| sample collection date | GENEPIO:0001174 | Date on which the sample was collected | For public health analyses this should be defined to the day. | ISO 8601 format (YYYY-MM-DD) | Spatio-temporal analysis | If event counts over a specific time period are low, perturbation may be required. |
| organism | GENEPIO:0001191 | Taxonomic name of the organism. | Use an ontology such as the NCBI Taxonomy Ontology (NCBITaxon). The name of the species should be provided. More granular information (e.g. serotype, sequence type, lineage) should be included in other fields. | Permitted values | Benefits all analyses | Taxonomic determinations may change depending on the method (PCR, culture, WGS etc) and reference database. The “organism” field in the sample record should reflect the taxonomic designation based on sequencing. |
| purpose of sampling | GENEPIO:0001198 | The reason that the sample was collected. | Use an ontology such as the Genomic Epidemiology Ontology (GenEpiO). | Permitted values | Informs potential sampling strategy bias | May have identifiability concerns when used in conjunction with other fields (i.e. geographic and temporal information) |
| purpose of sequencing | GENEPIO:0001445 | The reason that the sample was sequenced. | Use an ontology such as the Genomic Epidemiology Ontology (GenEpiO). | Permitted values | Informs potential sequencing strategy bias | May have identifiability concerns when used in conjunction with other fields (i.e. geographic and temporal information). |