Quantum computing is opening up new possibilities to advance bioscience research by accelerating drug discovery and helping to develop treatments for cancer, writes Eugenia Bahit
Day by day, quantum computing is showing us its incredible potential for advancing biosciences and its ability to become integral to the future of several research areas. Many fields can benefit from quantum computing, and there are many compelling reasons for choosing to explore its capabilities. One of them is the need for more humane and effective science; the second one, is the promise of an emerging market.
The urgency for more constructive and efficient science
The need for more humane and effective science is clear. Animal Free Research UK reports over 3 million animal experiments are conducted in the UK, while Humane Society International claims that 50% are bred and killed without yielding any scientific benefit. According to their reports, in Europe, 32 beagles are used for each agrochemical and new drug tests, and 25,000 animals have died in cosmetic testing after the EU's ban. In the US, the situation seems to be more extreme. PeTA, the global organisation for the ethical treatment of animals, informs that more than 110 million animals are killed yearly for scientific experimentation. And although the evidence suggests that nearly 95% of drugs tested on animals fail in human trials, animal testing remains a regulatory requirement across almost all of the globe.
Dogs, cats, macaques, monkeys, pigs, guinea pigs, rabbits, fishes, birds, goats, llamas, hamsters, rats and mice are on the list of animal models for experimentation. "These animals are sentient beings", says Dr Luis Falcón[1], the President of GNU Solidario[2], a humanitarian organisation headquartered in Spain and involved in the GNU Project for advancing software for public health. "Science that harms does not deserve to be called science", he adds.
There are many scientists, like Dr Falcón, who think it's time to change the way to conduct science, proclaiming that animal experimentation is unethical and speciesist. In an interview with Nature Reviews Materials, Dr Donald Ingber, Director of the Wyss Institute at Harvard University[3], challenged the reasons for animal testing by saying that "the question is whether we are fooling ourselves because we convince ourselves that what we see is what we thought it should be".
Similar questions are the ones that guide several organisations to research and promote alternatives to animal testing. Some institutions, such as the Center for Contemporary Sciences[4] in the US, are promoting changes in legislation and education to reach more effective and human biology-based methods. Moreover, in partnership with an investment group, they are also offering opportunities for companies working on new technologies and products to replace animal experimentation. And it is here, among the technologies that can help to reduce animal experimentation, where quantum computing stands out as an emerging area in the realm of biosciences, with a market that could reach over 4,000 million dollars by 2028, according to market research firm MarketsandMarkets.
Drug discovery, for example, is where quantum computing has made the most significant advancements, followed by research areas such as cancer, genetics and genomics. All of them show promising results, whereas precision medicine and early diagnosis seem to be the next areas where quantum computing reveals its potential.
Accelerating the drug discovery process
Drug development is a complex, long process. Understanding disease to determine a biological target that can be modified by an external compound is an essential phase which can demand several years. Identifying a lead compound that can effectively modulate the molecule or protein involved in such disease, ensuring it can be safe, and understanding its metabolisation and interactions, is not as simple as just getting a name. It requires a long list of candidates and many tests to choose the most suitable one, and then several years to optimise it to reach its most effective and safe version.
Given this complexity, scientists have been researching ways to accelerate the drug discovery process. By leveraging quantum annealing mechanisms, they have managed to create algorithms capable of finding the best lead compound candidates in just a few minutes by binding prospects to biological targets among billions of molecules.
That is the case of the American POLARIS Quantum Biotech[5] (POLARISqb), who have developed QuADD, a software as a service which leverages quantum annealing and distributed cloud computing for molecular library generation. In dialogue with Dr Anna Petroff[6] and Dr SantiagoVilar[7], computational chemists at POLARISqb, Dr Petroff explains that finding a small molecule that matches a protein target is challenging because the number of them is enormous. She notes that with QuADD, they can build a custom library of billions and find a lead compound "in less than a minute". Meanwhile, Dr Vilar remarks on the potential of quantum computing in optimising results by improving data quality and speed.
Other companies, such as River Lane coupled with AstraZeneca, in the UK, are advancing in harnessing the potential of quantum computing to calculate the solubility of lead compounds to estimate their effectiveness in the human body.
It is clear that quantum computing has a big potential in helping to identify and optimise lead compounds and could also help in the pre-clinical phase, where the current methods notoriously fail. In this respect, Dr Aysha Akhtar[8], renowned neurologist and public health specialist, US veteran and Founder and CEO of the Center for Contemporary Sciences, emphasises that “Up to 95% of all drugs and vaccines that are tried end up failing at the human clinical trial phase because they don’t work or are too toxic and unsafe”. She adds, “This shows that animal testing is very bad at telling us which drugs and vaccines are actually going to work in humans and be safe for humans to use.”
Thankfully, scientists are making computational progress in predicting the toxicity of chemical compounds currently tested on animal models. An example is the work of Dr Teppei Suzuki[9] and Dr Michio Katouda, who developed a quantum machine learning model to predict the toxicity of over 200 phenols used in various drugs and antiseptics. Another example is the quantum machine learning model developed by Dr Saad Darwish, using genetic programming to predict the toxicity level of different chemical compounds.
Advancements like these are driving drug regulatory agencies—such as the FDA in the US, under the Modernization Act 2.0—to support more humane requirements for approving drug releases to the market. Even so, many pharmaceutical companies sustain that the efficacy and side effects of new drugs can only be tested on animal models. Fortunately, quantum computing is shedding some light here. Indeed, Dr Shahar Keinan[10], CEO and co-founder at POLARISqb, bets on the future of quantum computing for pharma. "Quantum computing has the potential to revolutionise drug design and development. [It] can aid in understanding how drugs interact with target proteins or enzymes, predicting their efficacy and potential side effects," she says.
Cancer detection and treatment: another potential area for quantum computing
According to official statistics, cancer is responsible for over a quarter of deaths in England, and the survival time after diagnosis is too low. WHO reports 10 million deaths worldwide from cancer just in 2020. Cancer kills, and scientists are not yet able to find a cure. Understanding the development and progression of cancer is essential to finding remission mechanisms and guiding the appropriate treatments.
Scientists are battling to beat these challenges from different perspectives. The most conservative ones keep trying to replicate cancer in animal models even though science itself has demonstrated that using animals does not work. However, other scientists are banking on developing organ-on-a-chip technology and exploring how to apply quantum computing mechanisms to foster better and more contemporary science. In this sense, Dr Akhtar thinks these technologies can help advance cancer research, proclaiming that "It is going to be one of the major technologies that are going to replace animal testing for drug development, but also for disease modelling."
And she is not wrong. Dr Dipesh Niraula[11], an applied research scientist at Florida's Moffitt Cancer Center[12], is working in the radiotherapy field and developed a quantum deep reinforcement learning (qDRL[13]) framework to support clinical decisions in radiotherapy treatments. He explains the importance of this framework by making a parallelism with the decision process of buying a shirt available in different colours: "Before a shirt is purchased, shirt options are like superimposed quantum states. We wouldn't know their decision until they pick a shirt." He moves this analogy to clinical decisions and notes, "When doctors don't have complete information on the patient's state, disease progression, treatment response, etc., the clinical decision will be based on the physician's professional experience leading to inter-physician variability. And quantum computing helps in modelling such intrinsic uncertainty in human decision-making."
But the potential of quantum computing for cancer does not end there. Scientists are exploring the potential of quantum machine learning from different angles. Recent findings have shown that quantum transfer learning could help in histopathological cancer detection, while quantum convolutional neural networks could assist in breast cancer detection and brain tumour screening.
Moreover, the utility of quantum computing extends beyond the applied field. Theoretical approaches are trying to reach a better understanding of cancer by leveraging the potential of quantum computing wave functions to model the quantum mechanisms of genetic mutations involved in cancer development.
Advancing in genetics and genomics
Many diseases have a genetic cause, but the genome size makes it hard to understand some mechanisms with current technology. DNA and RNA sequencing, analysing, and assembling are essential to understand such diseases and, at the same time, are potential candidates to leverage the advantages of quantum computing. This is the case with phylogenetic trees, fundamental tools for understanding the evolution of certain organisms. Here, some scientists are exploring reconstruction via graph cutting using quantum annealing. Another crucial tool where quantum annealing is playing a key role is the de novo assembly, where the overlapping of DNA sequences benefits from the annealing's optimisation process.
But the quantum computing potential in genetics and genomics is even more promising. Quantum gene regulatory networks, for example, can aid in diagnoses and developing targeted treatments, while quantum comparison algorithms could detect DNA and RNA mutations to advance disease diagnosis and understanding.
We can save more lives with an early diagnosis
The potential of quantum computing to predict and identify diseases under uncertainties is crucial. As Dr Niraula notes, "Human decision-making process in the face of uncertainty gets unpredictable", and as he previously said, such uncertainties are like quantum states, which can be modelled with the help of quantum computing.
Scientists are using quantum computing to solve such uncertainties in many ways. Quantum neural networks, for example, are being explored for applications as different as electroencephalographic signals classification and personalised treatment for osteoarthritis, while quantum machine learning and quantum deep learning are being studied for heart failure detection, dementia prediction and diabetic retinopathy classification.
Dr Joseph Davids[14], a clinical research fellow at Imperial College London, specialising in nanomedicine and neurosurgery, thinks that quantum computing could be the future tool to help in medical diagnosis. In this respect, remarks: "Quantum computing will be responsible for not just diagnostics but treatments too."
He also notes[15] the importance of initiatives such as the National Quantum Computing Centre in Oxfordshire, which is aimed at accelerating quantum computing development in the UK and fostering patient-tailored therapeutics and diagnostics.
Quantum computing: the future of biomedical sciences
Many are the advancements of this emerging technology in the biosciences field. In just a few years, scientists have succeeded in exploring the quantum computing potential not only in a theoretical stage but also developing practical applications. Quantum computing is already underway and carving out a promising path.
Dr Akhtar thinks switching to these emerging technologies, may be expensive initially but asserts that over time "you don't have to worry about feeding computer chips". Excited by the conception of combining quantum computing with biological sciences, she pictures a future where "biological models like body-on-a-chip technology will be connected with computing models, where quantum computing will probably increasingly play a role because it's going to take a combination of these different techniques to give the best-combined understanding of human biology and human diseases."
Emphasising the existence of "lots of data that computing technology can screen to combine the information with biological models like organ-on-a-chip technology", she concludes, "This is going to be the future of biomedical science."
References:
[1] https://en.wikipedia.org/wiki/Luis_Falc%C3%B3n
[2] https://www.gnusolidario.org/
[4] https://contemporarysciences.org/
[6] https://www.linkedin.com/in/anna-b-petroff/
[7] https://www.linkedin.com/in/santiago-vilar-9423b8a/
[8] https://www.linkedin.com/in/ayshaakhtar
[9] https://www.linkedin.com/in/teppei-suzuki-3b312634/
[10] https://www.linkedin.com/in/shahar-keinan-9b729b1/
[11] https://www.linkedin.com/in/dipeshniraula/
[13] https://doi.org/10.1038/s41598-021-02910-y
[14] https://www.linkedin.com/in/joseph-davids/
[15] https://doi.org/10.1007/978-3-030-64573-1_338
