Supporting data and analysis code for “Human gastrointestinal representation conditions lineage-based source overlap and correction transfer in public non-typhoidal Salmonella genomics”

No genome sequences, read data or personally identifiable information are included.The deposited tables contain country-by-lineage and serovar-by-compartment counts, screening information for 156 countries, the nine-country analytical cohort, two alternative metadata adjudications, exclusion-flow counts and singleton sensitivity results. Unresolved, ambiguous or excluded records are documented rather than imputed. The code/ directory contains 12 Python scripts covering metadata aggregation, lineage-level analysis, supplementary-table generation and correction-transfer analysis.

Stage 1 used Python 3.10.12; downstream analyses used Python 3.11.15 with pinned NumPy, pandas, SciPy and openpyxl versions. Reconstructed scripts are explicitly identified, and all known small bootstrap-reproduction residuals are documented in code/RECONSTRUCTION_NOTES.md; these residuals do not change any reported statistical verdict. File integrity can be checked using CHECKSUMS.md5.Geographical coverageGlobal NCBI Pathogen Detection Salmonella metadata.

Country screening covered 156 countries; the primary analysis included the United States, Canada, China, Brazil, the United Kingdom, Japan, Germany, South Korea and South Africa.Temporal coverageA single rolling-database snapshot identified by checksum. The earliest verifiable timestamp is the source file modification time, 27 May 2026 at 14:11:15 UTC; this is not presented as the original download date. Collection dates were not used in the analysis.Spatial and analytical resolutionCountry × NCBI Pathogen Detection SNP cluster for lineage tables; serovar × sampling compartment for serovar tables.