TLDR: While everyone isf ocusing on the latest advancements in ChatGPT and how quick it can send emails for you or making songs about ninjas having a snowball fight with Winnie The Pooh, it’s crucial to highlight how these AI-driven tools are not just theoretical concepts but are paving the way for practical applications in everytihng. Medical therapies and diagnostics are immediately relevant as they affect our lives directly. The integration of AI with CRISPR technology promises to enhance the precision of genetic edits, which could lead to breakthroughs in treating genetic disorders. In fact, they appear to show potential to do it better than even nature does.
Let’s get into it…
Researchers are harnessing the power of generative artificial intelligence (AI) to design novel CRISPR gene-editing systems, potentially revolutionizing the field of precision medicine. Two recent studies have demonstrated the feasibility of using AI models to create custom CRISPR proteins and guide RNAs.
In the first study, a team from Profluent, led by Ali Madani, used a protein language model called ProGen to design CRISPR proteins. They successfully edited the human genome with these AI-designed proteins, marking a significant milestone in the field.
The second study, conducted by researchers from Stanford University and the Arc Institute, involved the development of an AI model called EVO. Trained on a vast dataset of microbial genomes, EVO was used to design new CRISPR systems, although these have not yet been tested in the laboratory.
The AI models in these studies were trained on extensive biological data, enabling them to learn the intricacies of naturally occurring genetic sequences. This knowledge was then applied to design millions of novel CRISPR protein sequences and guide RNAs.
The most promising AI-designed CRISPR protein, OpenCRISPR-1, demonstrated efficiency comparable to widely used bacterial CRISPR-Cas9 enzymes while minimizing off-target cuts. The researchers also used OpenCRISPR-1 to create a base editor, which proved to be both efficient and less error-prone.
The potential impact of AI-designed CRISPR systems on precision medicine is significant. These tools could be tailored to specific applications, overcoming the limitations of naturally occurring gene-editing systems. Profluent aims to collaborate with companies developing gene-editing therapies to test AI-generated CRISPRs in a clinical setting.
The accessibility of these AI-designed tools is another crucial aspect. The OpenCRISPR-1 molecule and the ProGen2 model are freely available to researchers, promoting further advancements in the field.
As AI continues to play a more significant role in designing gene-editing tools, the possibilities for precision medicine expand. With the ability to create bespoke CRISPR systems, researchers can unlock new avenues for treating genetic disorders and developing targeted therapies. The intersection of AI and gene editing holds immense promise for the future of healthcare.
