Refer to the NIAID Guidelines below or in PDF format (updated August 12, 2024). All reference materials and metadata templates are available on the CEIRR Data Standards webpage.

Background and Justification

In March 2024, the first cases of highly pathogenic avian influenza (HPAI) H5N1 virus were reported in dairy cattle in the United States, with virus being detected in the milk of affected animals. To address potential concerns around metadata sensitivity as well as the complexity of characterizing the source of milk samples, NIAID’s Centers of Excellence for Influenza Research and Response, in coordination with US government partners and academic influenza researchers, have prepared a set of best practice guidelines for the appropriate collection and sharing of data from viral isolate samples related to the outbreak.

Application of Guidelines

• These guidelines will apply to all research conducted using CEIRR Network funding from NIH/NIAID.
• We recommend that all influenza researchers utilize the following guidelines where possible for community dissemination of data pertaining to the characterization of retail milk and other associated HPAI H5 samples.

Guideline 1: Use of a Centralized Data Repository

• Recommend sharing sampling data within a centralized data repository.
  • Recommended repository- BV-BRC Surveillance Data
  • While data may also be shared elsewhere, the intention is for the repository to become a ‘standard’ resource for retail milk results.
  • The advantage and incentive of the shared data repository is the ability to quickly compare data with that submitted by other researchers.
• In addition to a centralized data repository, it is recommended that a shared data reporting template, format, and minimum metadata standards are utilized so that data can be easily aggregated, compared, and analyzed.
  • Suggested templates are available on the CEIRR Network Data Standards webpage and outlined in the provided Naming Conventions Schematic.

Guideline 2: Data Reporting Practices

There are separate best practices for single animal or single farm samples, and for commingled or retail milk samples. Refer to the appropriate section for your sample type.

• Single Animal or Single Farm Sampling
  • Metadata Template: Use the schematic to determine the appropriate CEIRR iDPCC template to use for SRA data or metadata collection and reporting.
    • For animal swabs or farm samples including milk tank and environmental samples (such as surface or milking equipment swabs), metadata should indicate source of sample (exp. animal or environmental swab vs milk).
  • CEIRR Investigator Data Submission: Data generated using CEIRR funding (genomic, transcriptomic, immunologic, serology, etc.) should be submitted to the CEIRR iDPCC before submission to the relevant repository.
    • CEIRR investigators will submit data via the appropriate template to the CEIRR iDPCC. 
    • iDPCC will conduct QA/QC processing and standardization of metadata, and then will submit to NCBI, GenBank, and/or NIAID’s BV-BRC according to data type.
    • Raw sequence read data: all raw sequence reads should be submitted directly to SRA.
  • Sequence nomenclature: Refer to the schematic above.
    • Single animal or single farm samples will be treated as standard host-associated sampling with the associated strain designation (ex: A/host/state/isolate/year).

• Multiple-Farm Pooled, Commingled, Commercial, or Retail Milk Samples
  • Metadata Template: iDPCC Milk Sequence Submission Metadata Template v1.0
    • For environmental samples (such as surface or processing equipment swabs), metadata should indicate source of sample (exp. animal or environmental swab vs milk).
    • For commingled or commercial milk, clearly identify if sequence generated is from lab virus isolation or bulk sequencing.
  • CEIRR Investigator Data Submission: Data generated using CEIRR funding (genomic, transcriptomic, immunologic, serology, etc.) should be submitted to the CEIRR iDPCC before submission to the relevant repository.
    • CEIRR investigators will submit data via the appropriate template to the CEIRR iDPCC.
    • iDPCC will conduct QA/QC processing and standardization of metadata, and then will submit to NCBI, GenBank, and/or NIAID’s BV-BRC according to data type. 
    • Raw sequence read data: all raw sequence reads should be submitted directly to SRA.
  • PCR positive retail milk samples: confirm and report location of milk processing plant, not the dairy farm or original producer.
    • Retail milk is defined here as purchased from grocery store or similar outlet following distribution from a processing plant.
    • For commingled samples obtained from processing plants or similar operations, prior to commercial distribution, report the state where the operation is located.
    • To determine where retail milk was processed, the National Uniform Coding System for Packaging Identification of Milk and Milk Product Processing Plants can be found here.
  • Sequence nomenclature: Refer to the schematic above.
    • For commingled or commercial milk, strain identifier should begin ‘MILK-’. 
    • For commingled milk from multiple farms or commercial milk, strain_location should default to ‘USA’ in the metadata table and subsequently the strain name. 
    • More specific location data (two-letter state abbreviation), related to the location of the processing plant or commingled source, will be captured in the metadata table; however, this information will not contribute to the strain name. 
      • Due to the sensitive nature surrounding the release of the metadata, the location data will be access controlled by BV-BRC and the iDPCC and will only be available upon request. Please email support@ceirr-network.org for more information.

Guideline 3: Reporting of Sensitive Results

• CEIRR Investigators are required to inform NIAID and the CEIRR Program Officers PRIOR to the preprint, publication, or otherwise release of any research findings.
• It is recommended that all researchers, whether CEIRR affiliated or not, notify CEIRR, NIH/NIAID, and other USG partners of the presence of mammalian adaptations or other mutations of concern detected in variants 10 days prior to posting the data, so that the information may be verified, and a unified public health response can be organized.

Points of Contact

• Please reach out to the NIAID CEIRR Program Team at NIAID_CEIRR@nih.gov with any questions regarding these guidelines or if you have any results you would like to report.
• If you have any specific questions about the implementation of the metadata templates, please reach out directly to the CEIRR Influenza Data Processing and Communication Center (iDPCC)’s support team at support@ceirr-network.org.