A team at Standford have created CRISPR-GPT, a large language model (LLM) agent system to automate and enhance CRISPR-based gene-editing design and data analysis.  

CRISPR-GPT leverages the reasoning capabilities of LLMs for complex task decomposition, decision-making and interactive human–artificial intelligence (AI) collaboration. 

The system incorporates domain expertise, retrieval techniques, external tools and a specialised LLM fine-tuned with open-forum discussions among scientists.  

It assists users in selecting CRISPR systems, experiment planning, designing guide RNAs, choosing delivery methods, drafting protocols, designing assays and analysing data.  

The paper by lead scientist Le Cong and colleagues has been published in Nature Biomedical Engineering. 

In the study, the team used CRISPR-GPT to knock out four genes with CRISPR-Cas12a in a human lung adenocarcinoma cell line and epigenetically activate two genes using CRISPR-dCas9 in a human melanoma cell line.  

CRISPR-GPT enabled fully AI-guided gene-editing experiment design and analysis across different modalities, validating its effectiveness as an AI co-pilot in genome engineering.  

They report that an undergraduate with no experience achieved 90%+ editing efficiency on their first attempt using the program. 

Le Cong posted on LinkedIn: “CRISPR is revolutionary but requires PhD-level expertise, it can take weeks to learn, adopt, and design, analyse a CRISPR experiment for R&D or making life-saving medicine. Even pro scientists can make small mistakes (e.g. typos in guideRNA or cloning design) that cost months to find out, slowing us down. 

“We’re not just automating gene editing – we’re using AI to power scientists to cure diseases.” 

The researchers are encouraging others to try it out themselves – beta access is available at genomics.stanford.edu.

Diana Turner, Senior Digital Content Editor, DDW