Today, QSimulate, a biotech software company that develops quantum-informed simulation tools, announced a new round of financing that brings its total funding to over $11 million and simultaneously unveiled QUELO v2.3, the latest version of its quantum-enabled molecular simulation platform. The news signals that quantum mechanics is rapidly moving from a theoretical aspiration to a practical engine for pharmaceutical innovation.
Why quantum matters in drug discovery
Drug discovery is an inherently challenging process. Predicting the behavior of molecules in vivo with a high degree of accuracy is not only technically difficult but also fiscally expensive. Computers have become essential tools in this process, but the models scientists usually rely on — whether traditional chemistry simulations or even advanced artificial intelligence (AI) — can only capture part of the story.
Determining whether a drug works requires modelling interactions between thousands of atoms that are moving and changing shape in complex, aqueous environments inside the body. Even the most powerful computers struggle to fully represent all the factors that influence how strongly a drug binds to its target, how stable it is, and how it reacts with other molecules.
The rise of AI has provided powerful new tools for navigating chemical space, predicting properties, and generating candidate molecules. However, AI models inherit the limitations of their training data. They cannot reconstruct physical laws absent from that data, and they cannot extrapolate reliably into regions where experiments are sparse or classical simulations are untrustworthy.
As AI continues to evolve, quantum mechanics will be more important than ever in the next frontier of drug discovery where both quantum and AI technologies will act as complementary forces.
- Toru Shiozaki, QSimulate
It is this context that has made quantum computing so compelling. Unlike traditional computers, quantum computers can naturally capture the behavior of molecules at a fundamental level. They can represent and manipulate many possible states of a system at the same time, allowing them to explore molecular interactions in ways that classical computers cannot. In principle, this means quantum computers could simulate molecules with extremely high accuracy, predicting how they interact, how they fold, and how they respond to their environment without relying on simplified assumptions.
However, quantum computing is still extremely resource-intensive and expensive. Simulating even a single conformation of a protein at the quantum-mechanical level requires enormous computational resources. When you consider that proteins adopt millions of different conformations to capture their full dynamics, the scale of the problem quickly becomes astronomical. Current quantum hardware, while advancing rapidly, is still limited in the number of qubits, susceptible to noise and errors, and unable to perform the deep, precise calculations required for industrial-scale drug discovery.
Because fully quantum simulations are still out of reach, researchers are increasingly using hybrid approaches that combine classical computing, AI, and quantum-inspired algorithms to achieve results impossible with any one method alone.
Building a hybrid approach
In recent years, researchers have shown that hybrid techniques can tackle small, well-defined molecular problems more efficiently than either approach alone. These methods allow for high-accuracy simulations of molecular interactions while keeping computational costs manageable.
A case study recently published in JMIR Bioinformatics and Biotechnology used both AI and small-scale quantum simulations to predict the molecular stability, binding interactions, and potential toxicity of collagen fragments used in dermal fillers. The computational predictions closely matched laboratory results, demonstrating that hybrid approaches can reliably assess safety and reduce experimental workload.
Another study from 2022 demonstrated that quantum approaches could be used to identify 3D conformations of RNA. RNA folding is a computationally challenging problem because RNA molecules can form complex secondary and tertiary structures with many possible conformations. In this study, both classical and quantum methods were used to simulate small RNA sequences, showing that quantum algorithms can explore multiple folding pathways simultaneously. This enabled researchers to identify likely stable structures and folding dynamics more efficiently than traditional methods. While limited to small RNA molecules, the results suggest that quantum approaches could eventually tackle larger, biologically relevant RNAs, such as those in viral genomes or ribozymes.
In a 2024 Nature paper, researchers demonstrated that hybrid quantum computing can be applied to real-world drug discovery. Using a superconducting quantum device combined with classical computing, they modeled complex molecular interactions in two case studies: one exploring prodrug activation and another examining covalent inhibitors targeting the cancer-associated KRAS G12C mutation. The work showed that integrating quantum effects into simulations enabled more accurate predictions of molecular behavior, reaction pathways, and electronic properties, helping scientists explore chemical space in ways that classical methods cannot.
From proof-of-concept to industry
Building on these proof-of-concept experiments, a few companies and research groups are beginning to explore practical applications of these techniques. For example, Gero, a biotech startup focused on aging and longevity, has combined machine learning with a small quantum component to generate thousands of novel drug-like molecules. Published in Scientific Reports, Gero was able to produce molecules with promising chemical properties that were not present in the training datasets, demonstrating the potential of quantum-informed AI to explore new chemical space and generate molecules that classical methods might miss.
Qubit Pharmaceuticals is taking a different approach by using quantum-accurate data to train large-scale AI models. The company recently introduced FeNNix-Bio1, a foundation model built entirely on synthetic quantum chemistry simulations generated using exascale supercomputers in the US and Europe. FeNNix-Bio1 enables reactive molecular dynamics at a scale traditionally out of reach, supporting the formation and breaking of bonds, proton transfer, and quantum nuclear effects. It can simulate systems with up to a million atoms over nanosecond timescales while maintaining quantum accuracy. This allows researchers to explore dynamic protein-drug interactions, solvent effects, and molecular energetics in ways that complement static AI structure predictors like AlphaFold.
QSimulate is focused on using quantum mechanics to design faster, more accurate simulations of molecular interactions. Their approach combines advanced algorithms, insights from quantum chemistry, and modern high-performance computing to create tools that behave “as if” they were quantum-powered, even on today’s classical hardware. By integrating these simulations directly into the drug discovery pipeline, QSimulate enables predictive molecular modeling up to 1,000 times faster than traditional methods — reducing processes that once took months to just hours.
“Quantum mechanics is the often-overlooked key ingredient, and we’ve pioneered a quantum mechanics approach to unlock molecular insights in drug discovery that conventional AI methods cannot reach,” said Toru Shiozaki, cofounder and CEO of QSimulate, in the press release. “As AI continues to evolve, quantum mechanics will be more important than ever in the next frontier of drug discovery where both quantum and AI technologies will act as complementary forces.”
Today, the company unveiled QUELO v2.3, which is designed to tackle some of the most challenging problems in drug discovery, including modeling complex proteins, peptide drugs, and interactions involving metal ions.
“Peptide drug discovery is one of the new frontier in the drug discovery space, but the computational solutions for that are limited due to the fact that classical mechanics is not well suited to describe peptides with various constructs,” Shiozaki told DDN. “In collaborations, we will be providing QUELO v2.3 to optimize these drug molecules using quantum mechanics, something that could not be done accurately prior to our solution.”
As quantum simulation technology matures, it may eventually enable predictive modeling of entire biological systems, from metabolism to disease progression, fundamentally changing how drugs are discovered, tested, and optimized. Companies like QSimulate, Qubit Pharmaceuticals, and Gero are exploring this frontier today, but the broader scientific community is only beginning to scratch the surface of what quantum computing could mean for molecular medicine.
