rewrite this title DTU’s Researchers Develop AI Platform To Design Custom Proteins For Targeted Cancer Immunotherapy

by
Alisa Davidson

Published: July 25, 2025 at 10:25 am Updated: July 25, 2025 at 9:52 am

by Ana

Edited and fact-checked:
July 25, 2025 at 10:25 am

To improve your local-language experience, sometimes we employ an auto-translation plugin. Please note auto-translation may not be accurate, so read original article for precise information.

Researchers at the Technical University of Denmark have introduced an AI-based platform designed to accelerate the development of custom proteins that support immune system responses against cancer. This system enables the creation of specialized proteins within weeks instead of years, facilitating immune T cells in identifying and attacking cancerous cells. 

The platform integrates three AI models to engineer “minibinder” proteins that attach to T cells and provide them with a molecular guidance mechanism for targeting cancers, such as melanoma. The approach has been applied to develop proteins tailored for both broadly occurring and patient-specific cancer markers, suggesting potential applications in personalized oncology. 

In order to construct these molecular tools, the researchers first employed a generative model known as RFdiffusion to examine the structure of a cancer-related protein called NY-ESO-1, which is commonly present on tumor cells. A second AI model generated amino acid sequences predicted to fold into precise structures capable of binding to this target. A third model refined the results, filtering tens of thousands of generated sequences down to 44 candidates for laboratory evaluation. Among these, one design demonstrated effective performance. 

Additionally, the platform includes a virtual safety screening process to anticipate and exclude protein designs that may interact with healthy tissues, enhancing safety before any physical testing occurs. The workflow incorporates AlphaFold2, the award-winning protein prediction tool developed by Google DeepMind, and compresses a traditionally multi-year development cycle into a matter of weeks.

AI-Enhanced Cell Therapy Projected To Enter Human Trials Within Five Years

Timothy Patrick Jenkins, one of the researchers in the team, anticipates that it may take approximately five years before the newly developed method progresses to the stage of initial clinical trials involving human participants. Once prepared for application, the treatment protocol is expected to align with existing cancer therapies that utilize genetically engineered T cells, commonly referred to as CAR-T cells, which are currently approved for treating conditions such as lymphoma and leukemia. 

The therapeutic process will begin with a blood draw conducted in a hospital setting, comparable to a standard blood test. Immune cells will then be isolated from the collected sample and subsequently modified in a laboratory environment to incorporate the AI-generated minibinder proteins. These altered immune cells will then be reintroduced into the patient’s body, where they are designed to operate with high precision, locating and neutralizing cancerous cells with targeted accuracy. 

The latest research is part of a broader trend of advances in computational biology. Earlier in the year, Timothy Patrick Jenkins’ team also applied AI to develop engineered proteins intended to improve the effectiveness of snakebite antivenoms.

About The Author

Alisa, a dedicated journalist at the MPost, specializes in cryptocurrency, zero-knowledge proofs, investments, and the expansive realm of Web3. With a keen eye for emerging trends and technologies, she delivers comprehensive coverage to inform and engage readers in the ever-evolving landscape of digital finance.

More articles

Alisa Davidson

Alisa, a dedicated journalist at the MPost, specializes in cryptocurrency, zero-knowledge proofs, investments, and the expansive realm of Web3. With a keen eye for emerging trends and technologies, she delivers comprehensive coverage to inform and engage readers in the ever-evolving landscape of digital finance.

More articles