GIAB CYP2D6 Long-Read Validation Artifacts — Anukriti PGx Engine v0.6.0

StarPhase 1.4.2 CYP2D6 diplotype call outputs (JSON), phenotype concordance score, checksums, and the scoring script for three GIAB/GeT-RM reference samples (HG002, HG001/NA12878, HG01190) used to validate the Anukriti pharmacogenomics interpretation engine (anukriti-pgx-core v0.6.0). Samples span European (HG002, HG001/NA12878 CEU/Utah) and South Asian/Gujarati (HG01190) ancestry. StarPhase version 1.4.2-f11d33f, database v1.4.0-fe124d0. Overall phenotype concordance: 1.000 (3/3). NOTE: raw aligned read data (BAM/BAI) is intentionally NOT included here, as those derive from GIAB/1000 Genomes reference samples under their own data use terms; this deposit contains only the StarPhase-derived call outputs, scores, checksums, and analysis code produced by the authors. The checksums.sha256 / checksums.md5 files list hashes for the full local 18-file artifact set (including the excluded BAMs) as a provenance record. Produced in support of the Anukriti validation paper (in submission).