Scientists are training AI models to recognize and write pieces of human DNA that control gene expression, in hopes that one day these synthetic sequences can improve genetic medicine.
Image Credit: Scientific Frontline / AI generated (Gemini)
Scientific Frontline: Extended "At a Glance" Summary
- The Core Concept: A generative AI model designed to create synthetic DNA sequences, specifically cis-regulatory elements (CREs), that can precisely control gene activity within targeted cell types.
- Key Distinction/Mechanism: Unlike traditional methods that modify existing DNA by removing or inserting segments, this model generates entirely new, functional sequences from scratch. It adapts diffusion model technology—similar to that used in image generators like DALL-E—to analyze chromatin accessibility data and write novel genetic "instructions."
- Origin/History: Developed by scientists at the Broad Institute and Mass General Brigham; the study was published in Nature Genetics in December 2025, with further details released in January 2026.
- Major Frameworks/Components:
- Diffusion Models: The generative AI architecture used to create the sequences.
- Cis-Regulatory Elements (CREs): The short DNA segments targeted for generation, responsible for tuning gene expression.
- Chromatin Accessibility Data: The training dataset used to teach the model which regulatory elements are active in specific cells.
- AXIN2: A protective gene used as a proof-of-concept target to demonstrate the model's ability to reactivate suppressed genes in leukemia cells.
- Branch of Science:
- Computational Biology / Bioinformatics
- Artificial Intelligence (Generative AI)
- Genetics and Genomics
- Future Application: The technology aims to enhance gene therapies by creating synthetic regulatory elements that ensure treatments are active only in the correct tissues. Future uses could involve pairing these sequences with delivery vectors like adeno-associated viruses (AAVs) or genome editors.
- Why It Matters: This advancement moves beyond merely editing the genome to "writing" it, enabling the design of highly specific, potent genetic switches. This could lead to more effective treatments for complex diseases like cancer by ensuring therapeutic genes are turned on more effectively than their natural counterparts would allow.
Scientists at the Broad Institute and Mass General Brigham have built a generative AI model that creates short DNA segments that can control gene activity in specific cells. These sequences, called cis-regulatory elements (CREs), make up a large part of the human genome, and synthetic versions of these bits of DNA could one day be part of gene therapies that tune gene activity to treat disease.
The model, called DNA-Diffusion, designed robust synthetic regulatory elements, including ones that reactivated a protective gene in leukemia cell lines. According to study senior author Luca Pinello, this technology could potentially lead to new therapeutic strategies that combine synthetic regulatory elements with existing gene therapy technologies to ensure that these therapies reach the right cell types in the body. The study is published in Nature Genetics.
“If you think about DNA as a language, you cannot master the language just by removing letters or inserting words into a sentence,” said Pinello, a Broad associate member and faculty at Mass General Brigham Cancer Institute. “To learn a language, we should be able to create whole new sentences.”
The team says their model outperformed other methods in generating synthetic CREs that were functional, cell type-specific, and had a large variety of sequences.
“In this paper, we demonstrate that this model not only can create sequences that appear to work in cells, but we can modulate the specificity, the activity, and the intensity of gene expression,” Pinello said.
Targeting a cancer gene
The core technology is based on diffusion models, an AI technology that powers image generators like DALL-E and Stable Diffusion. These models are trained to analyze images at the pixel level and then to generate new images using patterns they learned during training. Pinello and his team adapted this approach for DNA, training their model on chromatin accessibility data from regions of the genome that contain regulatory elements. Chromatin accessibility data reveal which regulatory elements are actively being used by the cell to control gene activity. After training the model on this data from three cell types, the team used it to generate more than 5,800 synthetic CREs. When they tested these sequences in the lab, they found that they maintained their gene regulatory functions in specific cell types.
To further demonstrate their technology, the researchers focused on the AXIN2 gene, which protects against chronic lymphocytic leukemia and is often turned off in B cells in patients with this disease. Using a cell line derived from individuals with chronic lymphocytic leukemia, Pinello and his collaborators analyzed the activity of 100 sequences, including 60 generated by their model. They found that many of the synthetic CREs were more effective at switching on AXIN2 than their natural counterparts. They also showed that sequences designed for B cells showed activity, whereas sequences designed for other cell types did not activate AXIN2.
Pinello and his team are now expanding the scope of their model and hope to combine their technology with genomic medicine approaches such as genome editors or gene therapies that use adeno-associated viruses (AAVs) to deliver therapeutic cargo to specific cells or tissues in the body.
“People at the Broad and all over the world are working on technologies to modify the genome in therapeutic ways, so gene therapies combined with this technology, we propose, can be a very powerful tool,” Pinello said.
Funding: This research was supported by the National Institutes of Health and the Rappaport MGH Research Scholar Award, as well as a Krantz Center Spark Award.
Published in journal: Nature Genetics
Title: Designing synthetic regulatory elements using the generative AI framework DNA-Diffusion
Authors: Lucas Ferreira DaSilva, Simon Senan, Judith F. Kribelbauer-Swietek, Zain Munir Patel, Lithin Karmel Louis, Aniketh Janardhan Reddy, Sameer Gabbita, Jonathan D. Rosen, Zach Nussbaum, César Miguel Valdez Córdova, Aaron Wenteler, Noah Weber, Tin M. Tunjic, Martino Mansoldo, Talha Ahmad Khan, Gue-Ho Hwang, Vincent Gardeux, David T. Humphreys, Cameron Smith, Matei Bejan, Peter Bromley, Will Connell, Bart Deplancke, Michael I. Love, Emily S. Wong, Wouter Meuleman, and Luca Pinello
Source/Credit: Broad Institute | Jessica Colarossi
Reference Number: gen012426_01
.jpg)