Missing by design: why genomic metadata must be captured at submission, not recovered from the literature

A literature-recovery pipeline for plant genome assemblies misassigned ploidy once among 82 assemblies and found ploidy in only about half of 1,059 assemblies, often missing it entirely.

• Why it matters: Genome assemblies in INSDC frequently lack key biological context like ploidy, cultivar, and subgenome structure, and free-text optional fields make records inconsistent and hard to interpret.
• What they did: The authors analyzed gaps in ploidy, chromosome naming, and subgenome-linked inconsistencies, then built a pipeline to extract these attributes from publications linked to assemblies.
• The result: Because recovery can annotate already-deposited assemblies but cannot prevent new incomplete submissions, they recommend capturing ploidy, cultivar, and subgenome structure at submission as mandatory controlled-vocabulary fields.

AI summary of the abstract · may contain errors · not medical advice · read the paper.