{
  "abstract": "Here we show that large language models (LLMs) can be transformed via supervised fine-tuning of engineered prompts into SmileyLlama for exploring the chemical space of drug molecules. We benchmark SmileyLlama against pretrained LLMs and chemical language models trained from scratch for generating valid and novel drug-like molecules, and use direct preference optimization to both improve SmileyLlama's adherence to a prompt and as part of the iMiner reinforcement learning framework to predict molecules with optimized three-dimensional conformations and high binding affinity to drug targets. By training an LLM to speak directly as a chemical language model, while retaining most of its natural language capabilities, we show that SmileyLlama can reliably generate molecules with user-specified properties rather than acting only as a chatbot with knowledge of chemistry or as a virtual assistant. While SmileyLlama is geared toward drug discovery, the supervised fine-tuning/direct preference optimization/LLM framework can be extended to other chemical, biological and materials applications.",
  "abstract_shown": true,
  "affiliations": [
    "Kenneth S. Pitzer Theory Center and Department of Chemistry, University of California, Berkeley, Berkeley, CA, USA",
    "Promontory Labs, San Francisco, CA, USA",
    "Department of Molecular Medicine, The Herbert Wertheim UF Scripps Institute for Biomedical Innovation and Technology, Jupiter, FL, USA",
    "Chemical Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA. thg@berkeley.edu",
    "Departments of Bioengineering and Chemical and Biomolecular Engineering, University of California, Berkeley, Berkeley, CA, USA. thg@berkeley.edu",
    "Kenneth S. Pitzer Theory Center and Department of Chemistry, University of California, Berkeley, Berkeley, CA, USA. thg@berkeley.edu"
  ],
  "article_type": null,
  "arxiv": null,
  "author_total": 7,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "Joseph M. Cavanagh"
    },
    {
      "affs": [
        0
      ],
      "name": "Kunyang Sun",
      "orcid": "0000-0001-6472-1665"
    },
    {
      "affs": [
        1
      ],
      "name": "Andrew Gritsevskiy",
      "orcid": "0000-0001-8138-8796"
    },
    {
      "affs": [
        0
      ],
      "name": "Dorian Bagni"
    },
    {
      "affs": [
        0
      ],
      "name": "Yingze Wang",
      "orcid": "0000-0002-1706-3791"
    },
    {
      "affs": [
        2
      ],
      "name": "Thomas D. Bannister",
      "orcid": "0000-0003-0683-8886"
    },
    {
      "affs": [
        3,
        4,
        5
      ],
      "name": "Teresa L. Head-Gordon",
      "orcid": "0000-0003-0025-8987"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/6rbqcyrzbd595.bib",
  "cited_by": 30,
  "doi": "10.1038/s43588-026-00986-y",
  "fulltext_url": null,
  "id": "6rbqcyrzbd595",
  "issns": [
    "2662-8457"
  ],
  "landing_url": "https://doi.org/10.1038/s43588-026-00986-y",
  "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",
      "3002|Computational Drug Discovery Methods"
    ]
  ],
  "openalex": "W4403619927",
  "pdf_url": "https://www.nature.com/articles/s43588-026-00986-y.pdf",
  "pmcid": "PMC13489950",
  "pmid": "42115404",
  "published": "2026-05-11",
  "publisher": "Nature Portfolio",
  "ris": "https://research.draconitic.com/paper/6rbqcyrzbd595.ris",
  "sources": [
    {
      "first_seen": "2026-11-06T13:20:40Z",
      "native_id": "W4403619927",
      "source": "openalex",
      "url": "https://doi.org/10.1038/s43588-026-00986-y"
    }
  ],
  "subjects": [
    {
      "field": "chemistry",
      "origin": "openalex:3002|Computational Drug Discovery Methods",
      "subfield": "medicinal"
    }
  ],
  "title": "SmileyLlama: modifying large language models for directed chemical space exploration",
  "title_text": "SmileyLlama: modifying large language models for directed chemical space exploration",
  "type": "peer-reviewed",
  "updated_at": "2026-11-07T13:21:00Z",
  "url": "https://research.draconitic.com/paper/6rbqcyrzbd595",
  "venue": "Nature Computational Science",
  "versions": []
}