{
  "abstract": "Model-guided directed evolution seeks to identify high-fitness protein variants under limited oracle budgets. Protein language models (PLMs) provide rich representations for this task, but task-agnostic zero-shot scores can be misaligned with a target assay, while supervised search in high-dimensional embedding spaces can make surrogate modeling and uncertainty estimation sample-inefficient. We propose the Linear Fitness Subspace (LFS) hypothesis: within mutation-induced residue-level representation changes, a compact, assay-specific set of directions makes fitness variation linearly accessible from few labeled variants. This is a local, supervision-recoverable statement rather than a claim that protein fitness landscapes or global PLM geometry are universally linear. Building on this observation, we introduce Subspace-Guided Evolutionary Search (SGES), which estimates an LFS from a small initial sample and performs surrogate modeling, uncertainty estimation, and acquisition in the learned subspace. Across 10 core ProteinGym assays, 87 extended static-validation assays, and an 18-assay budgeted-search evaluation, SGES improves fitness prediction and search efficiency over zero-shot PLMs and recent ML-guided protein optimization baselines. Controlled comparisons with PCA, random projections, label-shuffled PLS, classical mutation features, and acquisition ablations further isolate the benefit of a fitness-aligned site-delta coordinate.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.07607",
  "author_total": 8,
  "authors": [
    {
      "name": "SiYuan Ma"
    },
    {
      "name": "Canran Xiao"
    },
    {
      "name": "Zikai Xiao"
    },
    {
      "name": "Albert Gao"
    },
    {
      "name": "Liang He"
    },
    {
      "name": "Xuan-Yu Wang"
    },
    {
      "name": "Shuying Cao"
    },
    {
      "name": "Xiaojun Jia"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/fpkd6x231rw00.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.07607",
  "fulltext_url": null,
  "id": "fpkd6x231rw00",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.07607",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.AI"
    ],
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "q-bio.QM"
    ],
    [
      "openalex",
      "1312|Protein Structure and Dynamics"
    ]
  ],
  "openalex": "W7220894630",
  "pdf_url": "https://arxiv.org/pdf/2610.07607",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-06",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/fpkd6x231rw00.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:27:27Z",
      "native_id": "2610.07607",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.07607"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "arxiv:q-bio.QM",
      "subfield": "bioinformatics"
    }
  ],
  "title": "Linear Fitness Subspace in Protein Language Models Enables Sample-Efficient Directed Evolution",
  "title_text": "Linear Fitness Subspace in Protein Language Models Enables Sample-Efficient Directed Evolution",
  "type": "preprint",
  "updated_at": "2026-10-30T12:27:27Z",
  "url": "https://research.draconitic.com/paper/fpkd6x231rw00",
  "venue": "arXiv",
  "versions": []
}