Supplementary data and literature-mapping records for "AI-driven grass genomics: a user-centric and evidence-guided roadmap from bioinformatics resources to breeding decisions"

This record contains the verified literature-mapping and citation-curation package supporting the review "AI-driven grass genomics: a user-centric and evidence-guided roadmap from bioinformatics resources to breeding decisions". The final verified curation corpus comprises 709 unique records: 594 structured-search-only records, 59 targeted-only records, 15 records represented in both the structured-search and targeted/core sets, and 41 citation-completion records added transparently during final manuscript reconciliation. The structured-search workflow initially retrieved 2,432 records from OpenAlex and PubMed, retained 2,247 records after initial deduplication, and produced a 923-record structured-search screening pool. The package includes the reference crosswalk for the 123 scholarly references cited in the review, Supplementary Methods, Supplementary Tables S1-S4, machine-readable CSV files, metadata-quality-control logs, publication-version reconciliation records, author-decision records, validation scripts and documentation. The metadata-correction log contains 249 audit entries covering a record-level metadata-correction subset of 224 records. Evidence roles A-E were author-confirmed for the 123 scholarly references used to support manuscript claims. Evidence-role assignments for uncited records remain provisional where explicitly labelled. The review used structured evidence mapping and claim-level citation curation and was not designed as a formal systematic review or meta-analysis. Data tables and documentation are licensed under the Creative Commons Attribution 4.0 International License. Python scripts in the scripts directory are additionally licensed under the MIT License.