{
  "abstract": "BACKGROUND: Genomic sequence-to-expression deep learning models, which are trained to predict gene expression and other molecular phenotypes across the reference genome, have recently been shown to have poor out-of-the-box performance in predicting gene expression variation across individuals based on their personal genome sequences. RESULTS: Here, we explore whether additional training (fine-tuning) on paired personal genome and transcriptome data improves the performance of such sequence-to-expression models. Using Enformer as a representative pre-trained model, we explore various fine-tuning strategies. Our results show that fine-tuning improves expression predictions on held-out individuals, including from held-out populations, for genes seen during fine-tuning, with comparable performance to variant-based linear models commonly used in transcriptome-wide association studies. However, fine-tuning does not improve model generalizability to held-out genes, which contain sequences and variants unseen during fine-tuning. CONCLUSIONS: Including individual-level genetic variation and paired expression data during the training of sequence-to-expression models improves their understanding of seen variants, enabling their application to held-out individuals. However, this strategy does not improve generalizability to unseen genes, highlighting a remaining open challenge in the field.",
  "abstract_shown": true,
  "affiliations": [
    "Department of Electrical Engineering and Computer Sciences, University of California Berkeley, Berkeley, CA, USA",
    "Center for Computational Biology, University of California Berkeley, Berkeley, CA, USA",
    "Chan Zuckerberg Biohub, San Francisco, CA, USA"
  ],
  "article_type": null,
  "arxiv": null,
  "author_total": 4,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "Ruchir Rastogi",
      "orcid": "0000-0001-7288-1714"
    },
    {
      "affs": [
        0
      ],
      "name": "Aniketh Janardhan Reddy",
      "orcid": "0000-0002-9782-5361"
    },
    {
      "affs": [
        1
      ],
      "name": "Ryan Chung",
      "orcid": "0000-0002-6799-3866"
    },
    {
      "affs": [
        1,
        2,
        0
      ],
      "name": "Nilah M. Ioannidis",
      "orcid": "0000-0001-9628-8229"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/d1jqs5ygwjxef.bib",
  "cited_by": 21,
  "doi": "10.1186/s13059-026-04091-1",
  "fulltext_url": null,
  "id": "d1jqs5ygwjxef",
  "issns": [
    "1465-6906",
    "1465-6914",
    "1474-7596",
    "1474-760X"
  ],
  "landing_url": "https://doi.org/10.1186/s13059-026-04091-1",
  "language": "en",
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "journal",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "openalex",
      "1311|Genetic Associations and Epidemiology"
    ]
  ],
  "openalex": "W4402826099",
  "pdf_url": null,
  "pmcid": "PMC13386774",
  "pmid": "42186093",
  "published": "2026-05-25",
  "publisher": "BioMed Central",
  "ris": "https://research.draconitic.com/paper/d1jqs5ygwjxef.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T13:26:46Z",
      "native_id": "W4402826099",
      "source": "openalex",
      "url": "https://doi.org/10.1186/s13059-026-04091-1"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "openalex:1311|Genetic Associations and Epidemiology",
      "subfield": "genetics-genomics"
    }
  ],
  "title": "Fine-tuning sequence-to-expression models on personal genome and transcriptome data",
  "title_text": "Fine-tuning sequence-to-expression models on personal genome and transcriptome data",
  "type": "peer-reviewed",
  "updated_at": "2026-11-09T13:21:05Z",
  "url": "https://research.draconitic.com/paper/d1jqs5ygwjxef",
  "venue": "Genome biology",
  "versions": []
}