2026-10-15T04:09:06Z https://www.scidb.cn/oai

10.57760/sciencedb.012qb 2026-10-08T11:02:08Z

2026-10-08 Co-occurrence and permutation-based analysis pipeline for syndrome-element network reorganization across gastric precancerous lesion stages (software and simulated data) doi:10.57760/sciencedb.012qb en This dataset provides the reproducible analysis code and simulated data (v2.0.0, updated 2026-10) for a methodological study. The pipeline quantifies how pairwise syndrome-element co-occurrence changes across gastric precancerous lesion stages (non-atrophic, atrophic, intestinal metaplasia/dysplasia) and reports the margin-induced and relational components separately: the lift is normalized by the marginal frequencies but not invariant to them, and its margin sensitivity is quantified; confirmatory inference comes from a margin-preserving parametric-bootstrap test (each stage keeps its own margins while the latent association structure is pooled), reported as a margin-adjusted residual with an interval under Benjamini-Yekutieli control, which is valid under arbitrary dependence. A patient-level stage-label permutation test of no distributional difference is reported alongside as a descriptive layer. The pipeline also includes weighted network metrics, a descriptive drift statistic D (the "qi progressing into blood" reading, whose margin-induced component has a known sign), single-mechanism counterfactual cohorts, null calibrations (equal-lift, common-dependence, mixed design), independent and whole-pattern omission stress tests, a pattern-level mapping null, and a label-permutation-calibrated TreeSHAP interaction diagnostic (descriptive, not a superadditivity test). All analyses use simulated data only; the stage contrasts are cross-sectional; every number regenerates from global seed 20260913. The package contains the analysis code, the simulated cohorts, the DGP parameters and seeds, the syndrome-element mapping table, the keyword rule table, and all edge-level results. traditional Chinese medicine; syndrome element; co-occurrence network; lift; permutation test; gastric precancerous lesions; chronic gastritis; reproducible research li xuan ye summer Li Zhuohao Wang Shaoli Liu Zhen PUBLIC https://mit-license.org dataset Science Data Bank