{
  "abstract": "Reliable generalization is essential for deploying drug–drug interaction (DDI) and drug–target interaction (DTI) models in drug discovery, pharmacovigilance, and clinical decision support. However, strong performance on random splits or weakly constrained evaluations may not reflect how well a model generalizes to new drugs, targets, or deployment conditions. In pairwise prediction, test instances may still resemble training examples in molecular structure, sequence, or network context. Such overlap can inflate performance estimates and make interpolation appear to be extrapolation. In this Perspective, we argue that cold-start evaluation should be treated as an auditable deployment claim. We organize common evaluation settings by their source and scope of novelty, ranging from random splits and entity-disjoint protocols to structure-level, temporal, domain-disjoint, and many-domain evaluations. The L0–L5 framework clarifies which deployment claim each setting is designed to test. Coverage, overlap, and distance quantify the resulting train–test discrepancy. We further provide a practical checklist for leakage control, validation alignment, candidate-pair construction, and domain-level risk reporting. This Perspective clarifies the boundaries and evidential scope of cold-start evaluation, providing a common basis for interpreting and comparing results across DDI and DTI studies.",
  "abstract_shown": true,
  "affiliations": [
    "The Quzhou Affiliated Hospital of Wenzhou Medical University, Quzhou People's Hospital, Quzhou, China.",
    "Institute of Fundamental and Frontier Sciences, University of Electronic Science and Technology of China, Chengdu, China.",
    "Department of Integrative Biotechnology, College of Biotechnology and Bioengineering, Sungkyunkwan University, Suwon, Republic of Korea.",
    "School of Electronic and Communication Engineering, Shenzhen Polytechnic University, Shenzhen, China."
  ],
  "article_type": "Perspective",
  "arxiv": null,
  "author_total": 4,
  "authors": [
    {
      "affs": [
        0,
        1,
        2
      ],
      "name": "Xiaoqing Ru",
      "orcid": "0000-0002-2968-6435"
    },
    {
      "affs": [
        1
      ],
      "name": "Yansu Wang"
    },
    {
      "affs": [
        1
      ],
      "name": "Quan Zou"
    },
    {
      "affs": [
        1,
        3
      ],
      "name": "Mengting Niu"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/dv8ejevz2n5m7.bib",
  "cited_by": 0,
  "doi": "10.1371/journal.pcbi.1014927",
  "fulltext_url": "https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014927",
  "id": "dv8ejevz2n5m7",
  "issns": [
    "1553-7358",
    "1553-734X"
  ],
  "landing_url": "https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014927",
  "language": "eng",
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "heading",
      "Biochemistry"
    ],
    [
      "heading",
      "Biological Databases"
    ],
    [
      "heading",
      "Biology and Life Sciences"
    ],
    [
      "heading",
      "Body Weight"
    ],
    [
      "heading",
      "Database and Informatics Methods"
    ],
    [
      "heading",
      "Drug Discovery"
    ],
    [
      "heading",
      "Drug Interactions"
    ],
    [
      "heading",
      "Drug Research and Development"
    ],
    [
      "heading",
      "Extrapolation"
    ],
    [
      "heading",
      "Interpolation"
    ],
    [
      "heading",
      "Macromolecular Structure Analysis"
    ],
    [
      "heading",
      "Mathematics"
    ],
    [
      "heading",
      "Medicine and Health Sciences"
    ],
    [
      "heading",
      "Molecular Biology"
    ],
    [
      "heading",
      "Numerical Analysis"
    ],
    [
      "heading",
      "Perspective"
    ],
    [
      "heading",
      "Pharmacology"
    ],
    [
      "heading",
      "Physical Sciences"
    ],
    [
      "heading",
      "Physiological Parameters"
    ],
    [
      "heading",
      "Physiology"
    ],
    [
      "heading",
      "Protein Structure"
    ],
    [
      "heading",
      "Protein Structure Databases"
    ],
    [
      "heading",
      "Proteins"
    ],
    [
      "heading",
      "Research and Analysis Methods"
    ],
    [
      "heading",
      "Structural Proteins"
    ],
    [
      "plos",
      "/Biology and life sciences/Biochemistry/Proteins/Protein structure/Protein structure databases"
    ],
    [
      "plos",
      "/Biology and life sciences/Biochemistry/Proteins/Structural proteins"
    ],
    [
      "plos",
      "/Biology and life sciences/Molecular biology/Macromolecular structure analysis/Protein structure/Protein structure databases"
    ],
    [
      "plos",
      "/Biology and life sciences/Physiology/Physiological parameters/Body weight"
    ],
    [
      "plos",
      "/Medicine and health sciences/Pharmacology/Drug interactions"
    ],
    [
      "plos",
      "/Medicine and health sciences/Pharmacology/Drug research and development"
    ],
    [
      "plos",
      "/Medicine and health sciences/Pharmacology/Drug research and development/Drug discovery"
    ],
    [
      "plos",
      "/Physical sciences/Mathematics/Numerical analysis/Extrapolation"
    ],
    [
      "plos",
      "/Physical sciences/Mathematics/Numerical analysis/Interpolation"
    ],
    [
      "plos",
      "/Research and analysis methods/Database and informatics methods/Biological databases/Protein structure databases"
    ]
  ],
  "openalex": null,
  "pdf_url": "https://journals.plos.org/ploscompbiol/article/file?id=10.1371/journal.pcbi.1014927&type=printable",
  "pmcid": "PMC13750547",
  "pmid": "43008011",
  "published": "2026-11-10",
  "publisher": "Public Library of Science",
  "ris": "https://research.draconitic.com/paper/dv8ejevz2n5m7.ris",
  "sources": [
    {
      "first_seen": "2026-11-11T12:21:00Z",
      "native_id": "10.1371/journal.pcbi.1014927",
      "source": "plos",
      "url": "https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014927"
    },
    {
      "first_seen": "2026-11-12T15:21:47Z",
      "native_id": "MED:43008011",
      "source": "europepmc",
      "url": "https://europepmc.org/article/MED/43008011"
    },
    {
      "first_seen": "2026-11-14T12:23:06Z",
      "native_id": "PMC13750547",
      "source": "pmc",
      "url": "https://doi.org/10.1371/journal.pcbi.1014927"
    }
  ],
  "subjects": [
    {
      "field": "biochemistry",
      "origin": "plos:molecular biology",
      "subfield": "molecular-biology"
    },
    {
      "field": "biology",
      "origin": "heading:Biology and Life Sciences",
      "subfield": "general"
    },
    {
      "field": "chemistry",
      "origin": "heading:Drug Discovery",
      "subfield": "medicinal"
    },
    {
      "field": "mathematics",
      "origin": "plos:numerical analysis",
      "subfield": "numerical-analysis"
    },
    {
      "field": "medicine",
      "origin": "plos:pharmacology",
      "subfield": "pharmacology-toxicology"
    }
  ],
  "title": "Beyond unseen entities: Auditable cold-start evaluation for drug interaction prediction",
  "title_text": "Beyond unseen entities: Auditable cold-start evaluation for drug interaction prediction",
  "type": "peer-reviewed",
  "updated_at": "2026-11-14T12:23:06Z",
  "url": "https://research.draconitic.com/paper/dv8ejevz2n5m7",
  "venue": "PLOS Computational Biology",
  "versions": []
}