@inproceedings{song-etal-2025-virtual,
    title = "Virtual {CRISPR}: Can {LLM}s Predict {CRISPR} Screen Results?",
    author = "Song, Steven  and
      Abdrabou, Abdalla  and
      Dabholkar, Asmita  and
      Day, Kastan  and
      Dharmoju, Pavan  and
      Perera, Jason  and
      Kindratenko, Volodymyr  and
      Khan, Aly",
    editor = "Demner-Fushman, Dina  and
      Ananiadou, Sophia  and
      Miwa, Makoto  and
      Tsujii, Junichi",
    booktitle = "ACL 2025",
    month = aug,
    year = "2025",
    address = "Viena, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/acl25-workshop-ingestion/2025.bionlp-1.30/",
    pages = "354--364",
    ISBN = "979-8-89176-275-6",
    abstract = "CRISPR-Cas systems enable systematic investigation of gene function, but experimental CRISPR screens are resource-intensive. Here, we investigate the potential of Large Language Models (LLMs) to predict the outcomes of CRISPR screens in silico, thereby prioritizing experiments and accelerating biological discovery. We introduce a benchmark dataset derived from BioGRID-ORCS and manually curated sources, and evaluate the performance of several LLMs across various prompting strategies, including chain-of-thought and few-shot learning. Furthermore, we develop a novel, efficient prediction framework using LLM-derived embeddings, achieving significantly improved performance and scalability compared to direct prompting. Our results demonstrate the feasibility of using LLMs to guide CRISPR screen experiments."
}