Expanding the Frontiers of Drug Discovery
An elderly man sits at his doctor’s office reporting that his shortness of breath, muscle cramps and persistent itching has gotten worse. His doctor has the unfortunate job of telling him that his chronic kidney disease has progressed further and available medications are not helping. A teenage girl fights for her life in an intensive care unit. The treating oncologist has to report to the family that her tumor has mutated and become unresponsive to all available medications.
Scenes like these play out millions of times every day around the world. Despite a century of progress, modern medicine has a long way to go. It turns out that so far, we can treat only about 500 of the roughly 10,000 known human diseases. And even the ones we treat, we often do not treat very well. Fixing the process of finding new medicine has a direct impact on everyone’s health.
The Narrow Lens of Modern Drug Discovery
This shortfall in medicine is not due to a lack of effort or money but to a physical constraint: each compound that might serve as a drug must first be made before it can be tested.
The central challenge of modern medicine has been deceptively simple: find small-molecule drugs that can precisely bind to proteins in the human body and correct the malfunctions that cause disease. These proteins—shaped by the code of our genes—are responsible for nearly every process of life. When under certain conditions functions of one or more proteins go awry, disease follows.
Following the Human Genome Project, despite the explosion of knowledge about disease pathways, i.e. the proteins implicated in specific diseases, progress in developing new drugs has slowed. The annual number of new FDA-approved small-molecule drugs has remained largely stagnant for more than a decade. One major reason: the compound collections we are testing to find new drugs are too limited.
Humanity has managed to make fewer than 10 million distinct drug-like compounds so far. But the rules of organic chemistry could produce more than 10^33, or a billion trillion trillion, distinct families of compounds that represent potential small-molecule drugs. Somewhere within that chemical universe lie the cures for thousands of diseases. Even with automated robotic screening systems, the total number of tested compound families so far—less than ten million—represents a vanishingly small fraction of what chemistry allows. This slow and expensive process of having to make and test potential drug like compounds has remained a chokepoint for the discovery of new drugs.
The reality is stark—physical synthesis of compounds followed by trial-and-error testing—will never get us far enough. If we want to find the medicines of tomorrow, we need a new way to explore the possible chemical universe without first having to build it.
Computation: The Next Frontier
Over the years researchers have hoped that computational methods will break this impasse, The vision is powerful: use computers to predict how potential drug like compounds will bind to disease-associated proteins—without needing to first make those compounds. The hope is that computational approaches could open vast new regions of the chemical universe to exploration.
In recent years, many in the industry have turned to artificial intelligence to accelerate discovery. The concept sounds revolutionary: train algorithms on vast datasets of known compounds, then let the machine design new ones. But in practice, AI has run into a fundamental obstacle—data scarcity.
AI systems can only learn from existing experimental data, which come exclusively from the small number of molecules that humanity has already synthesized. Because of this, AI tends to generate minor variations of known chemical structures instead of finding new ones. Its predictions are bounded by prior human knowledge. In short, AI cannot imagine what it has never seen. It can optimize the known, but it cannot chart the unknown chemical universe that holds the keys to the diseases we still cannot treat.
Breaking through this century-old barrier requires a fundamentally different approach—one that starts with physics, not guesswork or patchy data. Verseon has developed precisely that approach through its Deep Quantum Modeling (DQM) platform. Nobel laureate Hartmut Michel says, “The fundamental advances Verseon has made in quantum mechanical modeling of protein-drug interactions are extremely impressive”.
DQM tackles the exquisite complexities of quantum physics at the at the sub-nanometer or Angstrom scale and below to systematically design novel drug molecules that bind strongly and selectively to their target protein. Verseon’s scientists then make and test the most promising candidates in the lab, feeding those real-world results into VersAI, a proprietary artificial intelligence system specifically designed to operate in data-sparse environments typical of early-stage drug discovery. VersAI in turn generates variants of the drug-like molecules DQM finds and the best ones are further tested in the lab and advanced to clinical trials.
Because of their breakthroughs in physics driven design, Verseon can explore entirely new regions of the chemical universe. The platform is capable of designing and validating drug-like molecules that have never existed, breaking free from the constraints that have held back conventional discovery.
From the Convergence of Physics, Biology and AI to Real Drugs
Verseon’s technology is producing tangible results. The company’s precision oral anticoagulants (PROACs) are designed to prevent dangerous blood clots without the bleeding risks common to current therapies. Peer-reviewed studies show these molecules can block clot formation while preserving the body’s natural ability to stop bleeding—a feat no existing anticoagulant achieves. A key opinion leader in the field of cardiology, UCL Professor John Deanfield says, “Verseon’s anticoagulants with their unique mode of action and low bleeding risk represent an exciting ‘precision medicine’ opportunity for the treatment of a huge population of cardiovascular disease patients.”
Another program targets diabetic retinopathy, one of the leading causes of blindness worldwide. The current standard of care is to wait until vision has already deteriorated substantially. Only at that point does the severity of disease justify the risks of regularly injecting repurposed cancer drugs into the eye. And the symptoms only improve in about 50% of patients. In stark contrast, Verseon has developed orally administered, first-in-class drug candidates to prevent or even reverse retinal damage leading to vision loss before it occurs.
Verseon is developing drugs that will prevent all diabetic end organ damage including progressive kidney degeneration that leads many diabetics to kidney failure and dialysis. And the company’s novel chemotherapy drug candidates continue to be affective even when tumors mutate to become resistant to other known chemotherapy agents.
These programs demonstrate how the ability to explore previously inaccessible drug like chemicals through physics driven predictive design can deliver safer, more effective medicines for diseases long considered untreatable.
The Future of Medicine
Verseon’s CEO Adityo Prakash envisions a future in which drugs can target disease processes with exquisite precision. “The ability to control each protein’s function has implications far beyond treating diseases once they’ve started. It opens the door to preventing disease before it starts and slowing the processes that cause the ravages of aging.”
Verseon’s breakthroughs in quantum-physics modeling is redefining what is possible in drug discovery. For the first time, computation can truly explore the chemical universe—not by extrapolating from what we already know, but by simulating the underlying laws of nature that govern molecular interactions. This ability to model biology at the quantum level would ultimately allow medicine to reach far beyond its current limits.
As this approach progresses, the world will finally move from treating a few hundred diseases to treating thousands. The future of medicine will be built not on trial and error, but on understanding and design. Verseon’s breakthroughs show that the path to curing more diseases lies not in guessing better, but in knowing more about the physics of life itself.
The implications are profound. Fine-tuned control of disease-causing protein activity would allow medicine to intervene before diseases emerge and extend healthy human lifespan far beyond what’s now possible. The convergence of physics, biology, and AI is reshaping the foundations of drug discovery. Verseon’s platform opens the door to exploring the vast and untapped chemical universe for remarkable new medicines—and forever change billions of lives.
Author:
Arnold Kristoff
Content Producer and Writer
