An antigen-data submission needs the assay roles, not just a variant name
Review the requested version and scientific comparison conditions before accepting a data bundle.
The label does not define the assay role
A familiar variant name does not make two laboratory results comparable. The same label may refer to an antigen in one field and an antiserum exposure in another. Truthound can check a prepared submission matrix against expert-approved pairs. It cannot judge vaccine effectiveness, experimental validity or which vaccination recommendation should be made.
The updated scientific request
WHO updated its antigen-composition data request on October 27, 2026 ahead of December 2026 TAG-CO-VAC deliberations. It describes several types of evidence and comparison conditions. This is a request to scientific contributors, not an endorsement of a particular result or a declaration that any two tables sharing variant labels can be combined.
Preserve the comparison conditions
For a bundle review, retain request version, antiserum, assay antigen, assay method, collection time and result definition. A qualified scientist must determine the permitted comparisons from the actual request and protocol. Keep the roles in separate columns. A shared name cannot resolve different exposures, timepoints or measurement definitions.
Check an expert-supplied pair
For the antiserum and antigen assay mapping prepared as a Polars DataFrame, Truthound Core Python SDK 3.3.4's ColumnPairInSetValidator can compare the antiserum-and-antigen fields with valid_pairs supplied by that expert. Define missing-value handling and the applicable request version before running the check. Passing the pair only establishes consistency with the supplied matrix. It does not establish that the assay was performed correctly or that the data answer the scientific question.
Keep implementation and science separate
Pandera 0.35.0 offers schema and custom predicates, including Polars, for an existing preparation workflow. Generic predicate documentation does not establish the identical scientific pairing and null policy. Retain a separately tested implementation where appropriate, while keeping the actual comparison matrix and scientific review explicit. Neither framework supplies the experimental judgment.
Test a role mismatch
Test a synthetic record where the name is valid but the roles are swapped, and another with a missing collection condition. The submission review should ask for the relevant clarification rather than silently normalize both into the same series. Do not generate or recommend experimental procedures from this data-management example.
Submit within the reviewed version
Accept a bundle for the authorized review only when the required comparison conditions have supporting records or an explicit expert disposition. Revisit the matrix after a request revision. The quality check helps preserve what the submitted numbers mean; the responsible scientific process determines what conclusions they support.
Evidence
- Event checked
- Subject checked
- Modified
