{
  "abstract": "Inverse perturbation benchmarks rank interventions toward a desired cellular state, but high accuracy can answer a different question from biological prioritization. We present InverseBench, which makes candidate information and endpoint construction explicit, and FSTN, a linear pipeline evaluated under it. With identical inputs, candidates and validation, nested ridge reaches Spearman 0.959 for a transcriptional principal-component proxy but 0.445 for independently measured K562 growth fitness across 7,354 genes. Controlled response-entry availability across seven tasks gives heterogeneous gains, 0.442 for CD4 activation and 0.116 for external fitness. Yet predefined shortlist comparisons show no primary-cutoff penalty in K562 or a melanoma co-culture screen, and secondary cutoffs reverse the K562 ordering. FSTN's nonnegative set selector recovers 0.840 of measured two-gene constituents in three picks, against 0.771 for an unconstrained greedy. The framework separates proxy reconstruction, response-informed prioritization and unseen-intervention ranking: an endpoint-dependent accuracy gap does not by itself mean worse biological decisions.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": "New results",
  "arxiv": null,
  "author_total": 8,
  "authors": [
    {
      "name": "Jianfu Li",
      "orcid": "0000-0002-9949-7007"
    },
    {
      "name": "L. Frank Huang"
    },
    {
      "name": "Wenyu Song"
    },
    {
      "name": "Ying Li"
    },
    {
      "name": "James W. Jakub"
    },
    {
      "name": "Hong Qin"
    },
    {
      "name": "Yan Luo"
    },
    {
      "name": "Cui Tao",
      "orcid": "0000-0002-4267-1924"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/f6pdyr7qjyxq0.bib",
  "cited_by": 0,
  "doi": "10.64898/2026.10.19.760271",
  "fulltext_url": "https://www.biorxiv.org/content/10.64898/2026.10.19.760271v1.full",
  "id": "f6pdyr7qjyxq0",
  "issns": [
    "2692-8205"
  ],
  "landing_url": "https://www.biorxiv.org/content/10.64898/2026.10.19.760271v1",
  "language": "en",
  "licence": "CC-BY-NC",
  "licence_label": "CC BY-NC",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by-nc/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "biorxiv",
      "bioinformatics"
    ],
    [
      "heading",
      "Article"
    ],
    [
      "openalex",
      "1312|Gene Regulatory Network Analysis"
    ]
  ],
  "openalex": "W7223405545",
  "pdf_url": "https://www.biorxiv.org/content/10.64898/2026.10.19.760271v1.full.pdf",
  "pmcid": "PMC13738628",
  "pmid": "42995937",
  "published": "2026-10-19",
  "publisher": "Cold Spring Harbor Laboratory Preprints",
  "ris": "https://research.draconitic.com/paper/f6pdyr7qjyxq0.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:29:21Z",
      "native_id": "10.64898/2026.10.19.760271",
      "source": "biorxiv",
      "url": "https://www.biorxiv.org/content/10.64898/2026.10.19.760271v1"
    },
    {
      "first_seen": "2026-11-09T12:23:35Z",
      "native_id": "PMC13738628",
      "source": "pmc",
      "url": "https://doi.org/10.64898/2026.10.19.760271"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "biorxiv:bioinformatics",
      "subfield": "bioinformatics"
    }
  ],
  "title": "Information contracts determine what inverse perturbation benchmarks measure",
  "title_text": "Information contracts determine what inverse perturbation benchmarks measure",
  "type": "preprint",
  "updated_at": "2026-11-09T12:23:35Z",
  "url": "https://research.draconitic.com/paper/f6pdyr7qjyxq0",
  "venue": "bioRxiv",
  "versions": []
}