{
  "abstract": "High-dimensional mediators are routinely collapsed into low-dimensional scores before mediation analysis, but aggregation can change the mediated estimand rather than merely reduce noise. We characterize when a linear score preserves the full mediated effect. Exact preservation for every outcome model occurs precisely when the score contains the covariance-adjusted exposure-predictive mediator direction, the classical sufficient reduction for exposure under a Gaussian inverse-regression model. We introduce a distribution-free minimax adequacy loss equal to the fraction of exposure-relevant mediator information discarded by the score and exactly linked to overlap recovered by aggregation. A split-sample inverse-Wishart correction yields unbiased estimators of retained and full exposure information in moderate dimensions. Under spherically symmetric outcome-effect benchmarks, relative aggregation error has a Cauchy location-scale profile, and paired influence-function inference targets the realized gap for an observed outcome. In simulations, paired interval coverage ranged from 93.3% to 97.5%. In an outcome-free TCGA lung adenocarcinoma audit of 1,314 smoking-responsive genes, the pooled gene-mean loss was 0.811 (descriptive 95% gene-bootstrap interval 0.797 to 0.823), versus a dimension-only random-subspace benchmark of 0.913; retaining promoter and non-promoter means separately reduced pooled loss to 0.625. Independent half-sample screening was unstable, but top-100 and top-250 cross-screen pooled gene-mean losses ranged from 0.827 to 0.932. The framework separates structural adequacy, benchmark fragility, and outcome-specific discrepancy.",
  "abstract_shown": true,
  "affiliations": [
    "Michigan State University"
  ],
  "article_type": "New results",
  "arxiv": null,
  "author_total": 1,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "S. Hait"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/69gxfvcjkw2ve.bib",
  "cited_by": 0,
  "doi": "10.64898/2026.10.02.756279",
  "fulltext_url": "https://www.biorxiv.org/content/10.64898/2026.10.02.756279v1.full",
  "id": "69gxfvcjkw2ve",
  "issns": [],
  "landing_url": "https://www.biorxiv.org/content/10.64898/2026.10.02.756279v1",
  "language": null,
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "biorxiv",
      "bioinformatics"
    ],
    [
      "openalex",
      "2613|Advanced Causal Inference Techniques"
    ]
  ],
  "openalex": "W7221110065",
  "pdf_url": "https://www.biorxiv.org/content/10.64898/2026.10.02.756279v1.full.pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-02",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/69gxfvcjkw2ve.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:26:44Z",
      "native_id": "10.64898/2026.10.02.756279",
      "source": "biorxiv",
      "url": "https://www.biorxiv.org/content/10.64898/2026.10.02.756279v1"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "biorxiv:bioinformatics",
      "subfield": "bioinformatics"
    }
  ],
  "title": "When Does a Pathway Score Preserve a Mediated Effect? Aggregation Geometry, Minimax Adequacy, and Gap Inference for High-Dimensional Mediation",
  "title_text": "When Does a Pathway Score Preserve a Mediated Effect? Aggregation Geometry, Minimax Adequacy, and Gap Inference for High-Dimensional Mediation",
  "type": "preprint",
  "updated_at": "2026-10-30T12:26:44Z",
  "url": "https://research.draconitic.com/paper/69gxfvcjkw2ve",
  "venue": "bioRxiv",
  "versions": []
}