Elamur Research⌕
☰
home / papers / 2024 / CRISPR-GPT for Agentic Automation of Gene-editing Experiments

CRISPR-GPT for Agentic Automation of Gene-editing Experiments

Paper · 2024 · Yuanhao Qu, Kaixuan Huang, Ming Yin et al.

Summary

General-purpose LLMs such as ChatGPT-4 often perform poorly when asked to plan CRISPR experiments. They can produce guide RNA sequences that match no real genomic location, omit practical details like reagents, off-target checks and protocols, or bury useful advice in irrelevant text. The authors present CRISPR-GPT, an agent that pairs an LLM with curated domain expertise, recent literature and external bioinformatics tools. It guides non-specialists through the full design workflow: choosing an editing system, designing gRNAs, picking a delivery method, predicting off-target effects, drafting protocols and planning validation, including primer design. The architecture has an LLM planner, a tool provider, a task executor built as state machines, and a user-facing LLM agent that the researcher can supervise and correct. The authors tested the system in a real laboratory use case covering cloning, cell transduction, efficiency measurement and sequencing-based readout. The work matters because it shows that domain-grounded agents can make complex biology methods more accessible. It also builds in safeguards for ethics and genetic privacy.

Key points

  • Architecture: four core modules (LLM planner, tool provider, task executor, LLM agent). The paper implements 22 tasks as state machines that break each goal into sub-steps.
  • Modes: four predefined meta-tasks or LLM-planned task chains, plus a Freestyle Q&A Mode and an Off-target Prediction Mode for gRNAs the user has already designed.
  • gRNA design: draws on the Broad Institute's pre-designed guide libraries and the CRISPRPick toolkit. This contrasts with general LLMs, whose guides for genes like EMX1 or EGFR fail BLAST checks.
  • Heritable-editing safeguard: the agent always asks which organism is being edited. For human targets, it shows a warning, links to the international moratorium on heritable editing, and requires the user to confirm before continuing.
  • Privacy safeguard: following HIPAA principles, a filter blocks prompts containing nucleotide stretches of 20 bp or longer before they reach external LLMs. Long genome sequences are never stored. A stated limitation is that the agent cannot yet build full constructs or vectors from natural-language input.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🏢GoogleMountain View · United States533 works · 443,885 citations🎓Stanford UniversityStanford · United States567 works · 335,190 citations🎓University of California, BerkeleyBerkeley · United States243 works · 251,774 citations🎓Princeton UniversityPrinceton · United States137 works · 53,849 citations🏥Stanford MedicineStanford · United States6 works · 1,507 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Biochemistry, Genetics and Molecular Biology · 273 worksMolecular Biology · 270 worksMedicine · 288 worksHealth Informatics · 227 works

🧭 Topics 1Research subjects and languages the work is about.Language models & pretraining · 4,145 works

Cite

Cite as Yuanhao Qu, Kaixuan Huang, Ming Yin, Kanghong Zhan, Dyllan Liu, Di Yin, Henry C. Cousins, William A. Johnson, Xiaotong Wang, Mihir Shah, Russ B. Altman, Denny Zhou, Mengdi Wang and Le Cong. 2024. CRISPR-GPT for Agentic Automation of Gene-editing Experiments. arXiv:2404.18021.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2404.18021 · arXiv 2404.18021

@misc{qu2024crispr,
  title = {{CRISPR-GPT for Agentic Automation of Gene-editing Experiments}},
  author = {Yuanhao Qu, Kaixuan Huang, Ming Yin, Kanghong Zhan, Dyllan Liu, Di Yin, Henry C. Cousins, William A. Johnson, Xiaotong Wang, Mihir Shah, Russ B. Altman, Denny Zhou, Mengdi Wang and Le Cong},
  year = {2024},
  doi = {10.48550/arxiv.2404.18021},
  eprint = {2404.18021},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2404.18021}
}

References

54 references, 17 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Chatgpt plugins. https://openai.com/ blog/ chatgpt-plugins . ↗
  2. Rights (ocr), o. for c. summary of the hipaa privacy rule. https://www.hhs. gov/ hipaa/ for-professionals/ privacy/ laws-regulations/ index. html , 2008. ↗
  3. Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. Gpt-4 technical report. arXiv preprint arXiv:2303.08774, 2023.
  4. Rohan Anil, Andrew M Dai, Orhan Firat, Melvin Johnson, Dmitry Lepikhin, Alexandre Passos, Siamak Shakeri, Emanuel Taropa, Paige Bailey, Zhifeng Chen, et al. Palm 2 technical report. arXiv preprint arXiv:2305.10403, 2023.
  5. Anthropic. Claude 3 model card. https://www.anthropic. com/ claude-3-model-card . ↗
  6. Andrew V Anzalone, Luke W Koblan, and David R Liu. Genome editing with crispr–cas nucleases, base editors, transposases and prime editors. Nature biotechnology, 38(7):824–844, 2020. find
  7. Andrew V Anzalone, Peyton B Randolph, Jessie R Davis, Alexander A Sousa, Luke W Koblan, Jonathan M Levy, Peter J Chen, Christopher Wilson, Gregory A Newby, Aditya Raguram, et al. Search-and-replace genome editing without double-strand breaks or donor dna. Nature, 576(7785):149–157, 2019. find
  8. Daniil A Boiko, Robert MacKnight, Ben Kline, and Gabe Gomes. Autonomous chemical research with large language models. Nature, 624(7992):570–578, 2023.
  9. Andres M Bran, Sam Cox, Oliver Schilter, Carlo Baldassari, Andrew D White, and Philippe Schwaller. Chemcrow: Augmenting large-language models with chemistry tools. arXiv preprint arXiv:2304.05376, 2023.
  10. Tianle Cai, Xuezhi Wang, Tengyu Ma, Xinyun Chen, and Denny Zhou. Large language models as tool makers. arXiv preprint arXiv:2305.17126, 2023.
  11. Samuele Cancellieri, Matthew C Canver, Nicola Bombieri, Rosalba Giugno, and Luca Pinello. Crispritz: rapid, high-throughput and variant-aware in silico off-target site identification for crispr genome editing. Bioinformatics, 36(7):2001–2008, 2020. find
  12. Ryan D Chow, Jennifer S Chen, Johanna Shen, and Sidi Chen. A web tool for the design of prime-editing guide rnas. Nature biomedical engineering, 5(2):190–194, 2021. find
  13. Aakanksha Chowdhery, Sharan Narang, Jacob Devlin, Maarten Bosma, Gaurav Mishra, Adam Roberts, Paul Barham, Hyung Won Chung, Charles Sutton, Sebastian Gehrmann, et al. Palm: Scaling language modeling with pathways. Journal of Machine Learning Research, 24(240):1– 113, 2023.
  14. Peter C DeWeirdt, Kendall R Sanson, Annabel K Sangree, Mudra Hegde, Ruth E Hanna, Marissa N Feeley, Audrey L Griffith, Teng Teng, Samantha M Borys, Christine Strand, et al. Optimization of ascas12a for combinatorial genetic screens in human cells. Nature biotechnology, 39(1):94–104, 2021. find
  15. John G Doench, Nicolo Fusi, Meagan Sullender, Mudra Hegde, Emma W Vaimberg, Katherine F Donovan, Ian Smith, Zuzana Tothova, Craig Wilen, Robert Orchard, et al. Optimized sgrna design to maximize activity and minimize off-target effects of crispr-cas9. Nature biotechnology, 34(2):184–191, 2016. find
  16. Jordan L Doman, Alexander A Sousa, Peyton B Randolph, Peter J Chen, and David R Liu. Designing and executing prime editing experiments in mammalian cells. Nature protocols, 17(11):2431–2468, 2022. find
  17. Jennifer A Doudna and Emmanuelle Charpentier. The new frontier of genome engineering with crispr-cas9. Science, 346(6213):1258096, 2014. find
  18. Dan Du and Lei S Qi. Crispr technology for genome activation and repression in mammalian cells. Cold Spring Harbor Protocols, 2016(1):pdb–prot090175, 2016. find
  19. Nicole M Gaudelli, Alexis C Komor, Holly A Rees, Michael S Packer, Ahmed H Badran, David I Bryson, and David R Liu. Programmable base editing of a• t to g• c in genomic dna without dna cleavage. Nature, 551(7681):464–471, 2017. find
  20. Luke A Gilbert, Matthew H Larson, Leonardo Morsut, Zairan Liu, Gloria A Brar, Sandra E Torres, Noam Stern-Ginossar, Onn Brandman, Evan H Whitehead, Jennifer A Doudna, et al. Crispr-mediated modular rna-guided regulation of transcription in eukaryotes. Cell, 154(2):442–451, 2013. find
  21. Christopher J Giuliano, Ann Lin, Vishruth Girish, and Jason M Sheltzer. Generating single cell–derived knockout clones in mammalian cells with crispr/cas9. Current protocols in molecular biology, 128(1):e100, 2019. find
  22. Ruth E Hanna, Mudra Hegde, Christian R Fagre, Peter C DeWeirdt, Annabel K Sangree, Zsofia Szegletes, Audrey Griffith, Marissa N Feeley, Kendall R Sanson, Yossef Baidi, et al. Massively parallel assessment of human variants with base editor screens. Cell, 184(4):1064– 1080, 2021. find
  23. Shibo Hao, Tianyang Liu, Zhen Wang, and Zhiting Hu. Toolkengpt: Augmenting frozen language models with massive tools via tool embeddings. Advances in neural information processing systems, 36, 2024.
  24. Jordan Hoffmann, Sebastian Borgeaud, Arthur Mensch, Elena Buchatskaya, Trevor Cai, Eliza Rutherford, Diego de Las Casas, Lisa Anne Hendricks, Johannes Welbl, Aidan Clark, et al. Training compute-optimal large language models. arXiv preprint arXiv:2203.15556, 2022.
  25. Patrick D Hsu, Eric S Lander, and Feng Zhang. Development and applications of crispr-cas9 for genome engineering. Cell, 157(6):1262–1278, 2014. find
  26. Tony P Huang, Gregory A Newby, and David R Liu. Precision genome editing using cytosine and adenine base editors in mammalian cells. Nature protocols, 16(2):1089–1128, 2021. find
  27. Gue-Ho Hwang, You Kyeong Jeong, Omer Habib, Sung-Ah Hong, Kayeong Lim, Jin-Soo Kim, and Sangsu Bae. Pe-designer and pe-analyzer: web-based design and analysis tools for crispr prime editing. Nucleic Acids Research, 49(W1):W499–W504, 2021. find
  28. Hui Kwon Kim, Seonwoo Min, Myungjae Song, Soobin Jung, Jae Woo Choi, Younggwang Kim, Sangeun Lee, Sungroh Yoon, and Hyongbum Kim. Deep learning improves prediction of crispr–cpf1 guide rna activity. Nature biotechnology, 36(3):239–241, 2018. find
  29. Silvana Konermann, Mark D Brigham, Alexandro E Trevino, Patrick D Hsu, Matthias Heidenreich, Le Cong, Randall J Platt, David A Scott, George M Church, and Feng Zhang. Optical control of mammalian endogenous transcription and epigenetic states. Nature, 500(7463):472– 476, 2013. find
  30. Eric S Lander, Françoise Baylis, Feng Zhang, Emmanuelle Charpentier, Paul Berg, Catherine Bourgain, Bärbel Friedrich, J Keith Joung, Jinsong Li, David Liu, et al. Adopt a moratorium on heritable genome editing. Nature, 567(7747):165–168, 2019. find
  31. Tianhao Li, Sandesh Shetty, Advaith Kamath, Ajay Jaiswal, Xiaoqian Jiang, Ying Ding, and Yejin Kim. Cancergpt for few shot drug pair synergy prediction using large pretrained language models. npj Digital Medicine, 7(1):40, 2024. find
  32. Ruibo Liu, Jason Wei, Shixiang Shane Gu, Te-Yen Wu, Soroush Vosoughi, Claire Cui, Denny Zhou, and Andrew M Dai. Mind’s eye: Grounded language model reasoning through simulation. arXiv preprint arXiv:2210.05359, 2022. find
  33. Morgan L Maeder, Samantha J Linder, Vincent M Cascio, Yanfang Fu, Quan H Ho, and J Keith Joung. Crispr rna–guided activation of endogenous human genes. Nature methods, 10(10):977–979, 2013. find
  34. Prashant Mali, Kevin M Esvelt, and George M Church. Cas9 as a versatile tool for engineering biology. Nature methods, 10(10):957–963, 2013. find
  35. Abby V McGee, Yanjing V Liu, Audrey L Griffith, Zsofia M Szegletes, Bronte Wen, Carolyn Kraus, Nathan W Miller, Ryan J Steger, Berta Escude Velasco, Justin A Bosch, et al. Modular vector assembly enables rapid assessment of emerging crispr technologies. Cell Genomics, 4(3), 2024. find
  36. John A Morris, Jahan A Rahman, Xinyi Guo, and Neville E Sanjana. Automated design of crispr prime editors for 56,000 human pathogenic variants. Iscience, 24(11), 2021. find
  37. Shishir G Patil, Tianjun Zhang, Xin Wang, and Joseph E Gonzalez. Gorilla: Large language model connected with massive apis. arXiv preprint arXiv:2305.15334, 2023.
  38. Pablo Perez-Pinera, D Dewran Kocak, Christopher M Vockley, Andrew F Adler, Ami M Kabadi, Lauren R Polstein, Pratiksha I Thakore, Katherine A Glass, David G Ousterout, Kam W Leong, et al. Rna-guided gene activation by crispr-cas9–based transcription factors. Nature methods, 10(10):973–976, 2013. find
  39. Adrian Pickar-Oliver and Charles A Gersbach. The next generation of crispr–cas technologies and applications. Nature reviews Molecular cell biology, 20(8):490–507, 2019. find
  40. Lei S Qi, Matthew H Larson, Luke A Gilbert, Jennifer A Doudna, Jonathan S Weissman, Adam P Arkin, and Wendell A Lim. Repurposing crispr as an rna-guided platform for sequence-specific control of gene expression. Cell, 152(5):1173–1183, 2013. find
  41. FAFA Ran, Patrick D Hsu, Jason Wright, Vineeta Agarwala, David A Scott, and Feng Zhang. Genome engineering using the crispr-cas9 system. Nature protocols, 8(11):2281–2308, 2013. find
  42. Jeffry D Sander and J Keith Joung. Crispr-cas systems for editing, regulating and targeting genomes. Nature biotechnology, 32(4):347–355, 2014. find
  43. Kendall R Sanson, Ruth E Hanna, Mudra Hegde, Katherine F Donovan, Christine Strand, Meagan E Sullender, Emma W Vaimberg, Amy Goodale, David E Root, Federica Piccioni, et al. Optimized libraries for crispr-cas9 genetic screens with multiple modalities. Nature communications, 9(1):5416, 2018. find
  44. Timo Schick, Jane Dwivedi-Yu, Roberto Dessı̀, Roberta Raileanu, Maria Lomeli, Eric Hambro, Luke Zettlemoyer, Nicola Cancedda, and Thomas Scialom. Toolformer: Language models can teach themselves to use tools. Advances in Neural Information Processing Systems, 36, 2024.
  45. Ophir Shalem, Neville E Sanjana, and Feng Zhang. High-throughput functional genomics using crispr–cas9. Nature Reviews Genetics, 16(5):299–311, 2015. find
  46. Yongliang Shen, Kaitao Song, Xu Tan, Dongsheng Li, Weiming Lu, and Yueting Zhuang. Hugginggpt: Solving ai tasks with chatgpt and its friends in hugging face. Advances in Neural Information Processing Systems, 36, 2024.
  47. Gemini Team, Rohan Anil, Sebastian Borgeaud, Yonghui Wu, Jean-Baptiste Alayrac, Jiahui Yu, Radu Soricut, Johan Schalkwyk, Andrew M Dai, Anja Hauth, et al. Gemini: a family of highly capable multimodal models. arXiv preprint arXiv:2312.11805, 2023.
  48. Andreas Untergasser, Ioana Cutcutache, Triinu Koressaar, Jian Ye, Brant C Faircloth, Maido Remm, and Steven G Rozen. Primer3—new capabilities and interfaces. Nucleic acids research, 40(15):e115–e115, 2012. find
  49. Guanzhi Wang, Yuqi Xie, Yunfan Jiang, Ajay Mandlekar, Chaowei Xiao, Yuke Zhu, Linxi Fan, and Anima Anandkumar. Voyager: An open-ended embodied agent with large language models. arXiv preprint arXiv:2305.16291, 2023. find
  50. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al. Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems, 35:24824–24837, 2022.
  51. Shijie Wu, Ozan Irsoy, Steven Lu, Vadim Dabravolski, Mark Dredze, Sebastian Gehrmann, Prabhanjan Kambadur, David Rosenberg, and Gideon Mann. Bloomberggpt: A large language model for finance. arXiv preprint arXiv:2303.17564, 2023.
  52. Chengrun Yang, Xuezhi Wang, Yifeng Lu, Hanxiao Liu, Quoc V Le, Denny Zhou, and Xinyun Chen. Large language models as optimizers. arXiv preprint arXiv:2309.03409, 2023.
  53. Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao. React: Synergizing reasoning and acting in language models. arXiv preprint arXiv:2210.03629, 2022.
  54. Denny Zhou, Nathanael Schärli, Le Hou, Jason Wei, Nathan Scales, Xuezhi Wang, Dale Schuurmans, Claire Cui, Olivier Bousquet, Quoc Le, et al. Least-to-most prompting enables complex reasoning in large language models. arXiv preprint arXiv:2205.10625, 2022.

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/qu-2024-crispr-gpt-agentic-automation/