DenovAI, an AI-driven biotechnology company developing a de novo protein design platform for biologics discovery, today announced the publication of a study in the peer-reviewed journal Molecular Systems Biology reporting accurate design of functionally active synthetic proteins

pexels-pixabay-373543

.

The study, titled “AlphaDesign: a de novo protein design framework based on AlphaFold,” introduces a new class of experimentally validated AI-driven protein design tools that have the potential to accelerate drug development. 

Dr. Kashif Sadiq, Founder and CEO of DenovAI and co-senior author of the paper, said: “Our research shows what happens when for design we stop only tweaking nature’s proteins and start building our own from scratch. We’ve demonstrated that AlphaDesign can generate proteins that fold as intended and function in living systems, including by targeting complex bacterial immune mechanisms.”

READ MORE: DenovAI uses artificial intelligence to discover therapeutic antibodies

READ MORE: Synthetic biology reveals the secrets of life without oxygen

Co-invented by Michael Jendrusch, Prof. Dr. Jan Korbel and Dr. Kashif Sadiq at the European Molecular Biology Laboratory (EMBL)AlphaDesign combines generative models with optimization approaches to design the structure and sequence of synthetic proteins. The result is a programmable, general-purpose framework that designs entirely novel proteins completely from scratch (de novo) with specific functions, without the need for additional model training, starting template molecules, or evolutionary data.

Powerful platform

Sadiq continued: ”At DenovAI, we have taken this technology much further, building a powerful proprietary platform for accurately designing therapeutic antibodies, miniproteins, and programmable biologics. This paper is both a scientific validation of our approach and a foundation for our broader mission to transform and accelerate drug discovery and precision medicine by designing biology itself.

”Accuracy is important because it reduces required design iteration, accelerating the path to clinical trials and enhancing the likelihood of improved efficacy, safety profiles and clinical success. Ultimately this translates to delivering better outcomes for patients. Our vision is that for a given disease we’ll one day be able to design the right therapeutic molecule, right from the start.”

The study goes beyond only reporting computational “hits” and demonstrates real-world biological function of proteins created with the tool.

Synthetic protein inhibitors

Using AlphaDesign, the researchers designed synthetic protein inhibitors targeting retrons, bacterial defense systems that protect against phage infection and impede therapeutic phage applications. Impressively, experimental validation in vivo showed that 17 out of their top-selected 88 designs were functional inhibitors, making this one of the first large-scale demonstrations of AI-designed proteins with biological activity to reach such levels of accuracy – not to mention against complex, previously uncharacterized bacterial immune systems.

AlphaDesign is licensed to DenovAI by EMBL’s technology transfer partner and commercial arm, EMBLEM. Since licensing the platform 18 months ago, DenovAI has integrated it into and further developed a proprietary platform for designing antibodies, miniprotein binders, and novel therapeutics for difficult-to-drug targets. The company has already demonstrated high computational and experimental success rates across a range of applications, positioning it at the forefront of AI-driven drug discovery.

New level of control

Prof. Dr. Jan Korbel, Head of Data Science and Interim Head of Site at EMBL Heidelberg, said: “AlphaDesign represents a new level of control in computational biology: the ability to generate new proteins with measurable, targeted function. This work demonstrates that AI can go beyond predicting biological structure to actively creating new molecular capabilities. The implications are broad: from therapeutics and diagnostics to synthetic biology, we are now seeing how machine learning can become a true engine of biological innovation.”

DenovAI already raised seed funding to expand R&D and platform development. Backed by AION Labs – a venture studio led by AstraZeneca, Merck KGaA, Pfizer, Teva, Amazon Web Services (AWS),  the Israel Biotech Fund and Amiti Ventures, supported by the Israel Innovation Authority (IIA) and BioMed X Institute – the company is now preparing for a Series A funding round to accelerate its therapeutic programs and advance existing and new collaborations with global pharmaceutical partners.