The convergence of quantum computing and early drug discovery

Nils Anton Berglund
Stig Elkjær Rasmussen
Drug development is an increasingly costly and time-consuming endeavour for biopharma companies, but the emergence of quantum computing is expected to transform the early drug discovery process for companies looking to make the process more cost-efficient and effective.

Computationally Aided Drug Design on Quantum Computers

The potential practical value of quantum computing solutions in early drug discovery stems from how quantum computing differs from its classical cousin.

Traditional or classical computing uses bits that exist in two states called 0 or 1. Quantum computers, on the other hand, use the qubit (quantum bit), which probabilistically can be in both states at the same time, a quantum property called superposition. Another important quantum property that can be used in quantum computations is entanglement. Entanglement describes the linking of two quantum particles irrespective of the distance between them, e.g., when one qubit is in state 1, then qubit two will also be in state 1.

Together, superposition and entanglement open up computational possibilities that may prove beyond what is possible with classical computing systems and provide increased precision and insight for certain algorithms and use cases.

This holds material promise in the life sciences and, in particular, for the field of biopharma, since quantum computing could significantly improve the modelling of biomolecular interactions, crucial to today’s drug discovery. Further, the basic building blocks of all pharmacological compounds—molecules, atoms, electrons—themselves operate on quantum principles, making them difficult subjects for even high-performance computing systems to model. As the size and complexity of biomolecules increase, so do the multivariate problems of correctly capturing all those possible interactions and permutations.

Quantum computing integrated with traditional high-performance systems, can enhance early drug discovery through precise physical simulations and by enabling the design of new compounds via data-driven machine learning and generative modelling.

Working together, traditional high-performance computing (HPC) and quantum computing can offer advances for biomolecular dynamics and chemical compound interactions. One example is captured by Density Functional Theory (DFT), a quantum mechanics based methodology that attempts to predict the varying electronic structures of molecules when put under different circumstances and conditions.

In early drug discovery, machine learning and DFT combined can test the effectiveness of potential drug molecules against specific diseases and identify new drug modalities, even considering their interactions with human biology. This is done by predicting the particle structure of molecules and their behaviour in drug-target interactions to treat a disease with its potential side effects.

Integrating quantum-enabled DFT with artificial intelligence and the pattern-recognition capabilities it offers in large data sets, can enhance the prediction of chemical compounds properties in early discover by removing the approximations associated with running DFT on classical computers.  Using machine-learning models alongside physical modelling methods, allows researchers to test therapeutic targets and drug modalities in larger quantities and at an accelerated pace versus traditional methods. It may be possible to advance hybrid quantum systems to provide greater precision in identifying chemical compound interactions through simulations, but the field is still in its early stages and has yet to reach practicality.

Enhancing Productivity in Drug Research by QC

That’s the potential of quantum computing, high-performance computing, and machine learning models in early drug discovery, to significantly improve the drug development life cycle. Biopharma companies now spend well over $2 billion on average to bring a new drug to market, and that’s over a time scale of over a decade to move from R&D through clinical trials to commercialisation. With the biomolecular modelling power potential of quantum computing and high-performance computing, researchers and private entities will be able to cast wider nets in the discovery of new compounds leading to an increase in new drug candidates, creating more productive and cost-efficient process.

Examples of QC Usage in Early Drug Discovery

With quantum computing enabled software, the early stages of drug discovery are likely to offer the most potential. Knowing where to deploy these novel capabilities will be key to adding value.

CADD with physical simulation and AI

Computer-Aided Drug Design (CADD) is an essential part of the modern drug development life cycle, from discovery through pre-clinical and clinical trials as well as in marketing. Specifically in the realm of physical simulation on high-performance computing (HPC) systems with machine- and deep-learning techniques, CADD in early drug discovery can predict compound properties and understand the impact of drugs within the body.

Disease understanding, target finding, hit generation/identification

At the early target-finding stage of CADD screening, where researchers work to identify a specific proteinand understand their associations and impact on a particular disease, vast amounts of biological data are analysed and complex biological systems need to be accurately modelled to generate a relevant target. 

Here, while CADD and high-performance computing excel in data crunching, quantum computing could make the process more precise. Quantum computing through physical simulation has the potential ability to extract novel insights and help better interpret biological data sets, which provides a method of validating promising targets beyond the reach of traditional CADD.

Once a disease-relevant target has been identified, the goal is to find molecules that generate a “hit” by interacting with the target protein. Typically, researchers use methods like High-Throughput Screening (HTS) which involves either the physical lab testing of compounds for reactions, or computational virtual screening, which churns through databases of chemical compounds to computationally screen for potential hits.

The unique processing of quantum computing could allow researchers to more fully ascribe characteristics to compounds and thereby more accurately depict their potential interactions, helping to break through the hit generation wall and optimise a new drug’s characteristics.

Generative chemistry and QC: Generation and Validation

An emerging component in CADD is generative chemistry, the process of using artificial intelligence and deep-/machine learning to generate chemical models. This technique in early drug discovery is nascent on supercomputing systems but is an emerging method of how researchers can search and devise new chemical structures in hit generation and identification. While prospective, when high-level screening in generative chemistry converges with the increased precision and accuracy of quantum computing, the potential for significantly improved drug discovery is unlocked. Where novel AI methods bring strong throughput, physical simulations and quantum computing can complement this with precision and validation.

Optimisation of drug modality and properties

After finding hits, researchers modify the chemical structure of drug candidates to improve suitability vis-a-vis the target and suitability for delivery to the target, and this optimisation process takes place on several levels, involving considerations of factors such as absorption, distribution, metabolism, excretion, and toxicity (ADMET) of the prospective drug, as well as optimisation for key properties like specificity and binding affinity.

Here, too, simulations of the quantum mechanical interactions between compounds and proteins, will prove more effective and efficient than conventional CADD in terms of improving specificity and binding affinity as well as ADMET characteristics and other properties, thereby increasing the quality and potential approval rate of new drugs. 

Quantum computing, or a hybrid approach incorporating high-powered computing, can potentially boost the low approval rate of new drugs known as the ‘Valley of Death,’ commonly attributed to high toxicity levels and the drug’s adverse interaction with human biology.

An interesting application in this area is the potential for quantum computing to improve the predictions of drug molecules' unbinding kinetics (Koff) from the target protein. Implementation of this capability in a drug discovery workflow has the potential to improve dosing regimens of drugs which can improve efficacy and safety of new drugs.

Optimal Quantum States: The Kvantify Solution

Companies today are looking to employ quantum computing and HPC to meet the challenges of their respective industries, most clearly in the need for expertise in the field and access to cutting-edge technology. 

At its current stage of development, noted as the Noisy Intermediate-Scale Quantum (NISQ) era, quantum computing systems are too faulty to reliably measure qubit states. Necessitating technical complications such as noise, topology, loss tangents, coherence times, and more, to be solved before quantum computing systems can be practically useful.

As it stands, quantum computers are too prone to error to achieve quantum value and potentially even quantum advantage. As a result, error correction is needed to make practical and reliable commercial use of quantum computing.

One promising approach to error-correction is limiting the quantum noise that prevents current quantum computing systems from practical usage. Our FAST-VQE (Fermionic Adaptive Sampling Theory for Variational Quantum Eigensolver) limits quantum noise in today’s noisy intermediate-scale quantum (NISQ) computers, realising its potential for practical applications. Currently, our FAST-VQE is in late-stage testing with the support of AWS Amazon Braket, and testing has shown that the FAST-VQE algorithm allows for the precise selection of optimal components of a quantum state. 

As a tool for quantum chemistry, FAST-VQE enables researchers to represent on current NISQ computers the electronic wave functions of qubits, essentially cutting through the quantum noise and error. Kvantify calls it a remarkable achievement and one with which companies in the life sciences and pharma industries will be able to peek into the potential of what quantum computers could provide and possibly outperform classical computers. 

For more information on our quantum-ready solutions, please do not hesitate to reach out at

Future Projections

The impact of quantum computing projects is potentially revolutionary for the pharmaceutical industry. A McKinsey report states the emerging quantum ecosystem could have an overall economic impact of nearly $700 billion by 2035, with the Pharmaceuticals and Medical Products industry taking on close to $200 billion of that total. 

And as early-stage drug discovery takes up a significant portion of pharma’s budget, even a 10-20% shift from lab-based R&D to CADD-based in silico target identification and drug design would lead to savings of $33-$66 billion.

Drug development is expected to become more precise through quantum computing, especially when integrated with machine learning algorithms for added pattern recognition in data analysis. This increased accuracy in CADD has the potential to deliver better and safer therapies to the clinic.

The reach of pharma will also become wider, as lowered expenses and timelines enable companies to more cheaply pursue treatment for neglected and rare diseases, opening new market opportunities while providing palliating effects for more patients.


Quantum computing is well-positioned and is in the early stages of revolutionising drug discovery, enhancing accuracy in modelling molecular interactions through physical simulation can be groundbreaking to developing new, better treatments.

When integrating machine learning with quantum computing, we can expect increased speeds and efficiency in searching large data sets through generative modelling and even the ability to recognise previously hidden patterns. This has the potential to drastically reduce costs and the time-to-market for new drugs.

In other words, by complimenting high-performance computers with quantum computers, ‘more and better shots on goals’ could be achieved.

Kvantify's Role in Quantum and AI Accelerated Drug Discovery

Our role in Quantum and AI Accelerated Drug Discovery came to life with Kvantify's FAST-VQE. We're proud to offer something that can really ramp up early drug discovery, making it more efficient and effective, especially in complex molecular modelling.

In early March 2024, we are launching our first product - a groundbreaking new tool to that will help accelerate drug discovery through classical means today, with quantum advantage tomorrow.

If this seems like the solution you've been searching for, contact us to experience the difference FAST-VQE and our tools can make in your drug discovery initiatives.

Nils Anton Berglund
Head of Strategic Alliances

Assistant Professor at Aarhus University and a biotech consultant. Nils has a PhD in Theoretical Biophysics from Southampton University, and experience working with biotech companies. His expertise lies in computational strategies in drug development.

Stig Elkjær Rasmussen
Quantum Engineer

Working among other things with optimization, Stig studies modelling of quantum systems and quantum machine learning algorithms. Stig holds a PhD in physics from Aarhus University where he is also an external lecturer.