- Australian company OmnigeniQ has revealed the first computer model of a human protein as it exists in the body, confirming that native protein topology can be calculated directly from physics.
- The breakthrough was achieved using the company's physics-based Deterministic Intelligence model that shows proteins in their native, hydrated, dynamic form – something existing tools cannot do.
- This milestone supports OmnigeniQ's mission to build the world's first holographic twin of the human body, enabling more preventative, predictive and precise medicine.
OmnigeniQ has unveiled a world-first scientific milestone at Biotech Showcase in San Francisco, demonstrating the first deterministic computation of Cyclin-dependent kinase 5 (CDK5), a human enzyme implicated in neurological development and disease. This method confirms OmnigeniQ's physics-based approach to better understand protein behaviours and takes OmnigeniQ a step closer to modelling full biological systems.
Using its proprietary physics-based computational framework, OmnigeniQ successfully computed CDK5's 3D structure, hydration shell, and surface topology directly from first principles of physics, without relying on structural templates, experimental or statistical approximations, or AI pattern-matching.
The resulting structure is consistent with known CDK5 features observed through experimental techniques such as X-ray crystallography, while also revealing the protein in a native, hydrated, and dynamically moving state, which cannot be captured by existing tools. This marks the first time that human protein topology and behaviour have been calculated and visualised directly from physics, rather than inferred from experimental or data-driven models.
This Australian breakthrough is a significant discovery for the future of modern medicine development, as the shape of proteins in the human body dictate which drugs can bind to a protein and where, what activates when it binds and what fails and leads to disease.
Instead of visualising proteins as static or dehydrated structures, the system computes them as living, hydrated entities in motion – the way they exist inside the human body. This approach reveals molecular behaviours that existing experimental and traditional AI-based tools cannot capture.
Implications for drug discovery and development
Proteins are the living nanomachines that drive every biological process, constantly folding, flexing and responding to their environment. Almost every modern medicine works by targeting a protein, and a medicine's success in clinical trials depends on the exact 3D geometry of the molecule it engages.
CDK5 is a particularly important enzyme, playing a central role in neuronal signalling and regulation. Misregulation of CDK5 activity has been associated with a range of neurological and neurodegenerative conditions, making it a target of significant scientific and therapeutic interest. Understanding the true physical structure and behaviour of CDK5 in the human body is therefore critical for designing therapies that interact with it safely and effectively.
Tiffanwy (Tiff) Klippel-Cooper, OmnigeniQ's co-founder and Chief Science Officer, developed the underlying physics-first approach that powers Deterministic Intelligence and says the breakthrough reflects what the technology was always designed to achieve.
"Proteins have always been treated as objects to be imaged or inferred, rather than physical systems to be computed. In reality, a protein's structure emerges from interacting physical constraints – charge distribution, hydration, field effects, and continuous motion. I designed the computational model to let those constraints resolve the structure deterministically. Computing CDK5 from first principles shows that native protein conformations are not something we have to guess or approximate – they are a direct consequence of physics."
Jordana Blackman, co-founder and Chief Executive Officer of OmnigeniQ, agrees that the result marks a defining moment for the company.
"What this unlocks for modern medicine is profound. If you know the true, dynamic structure of a protein, you can design a drug that engages it with far higher specificity. That means fewer off target effects, fewer failed candidates, and a faster path to viable therapies. The industry spends billions each year on molecules that fail because their target wasn't fully understood. Physics accurate protein computation changes that equation and gives us the ability to completely overturn the process of drug development."
This milestone marks a major step in OmnigeniQ's mission to create the world's first holographic twin of the human body – a physics-accurate in silico replica to make medicine preventative, predictive and precise.
