When light is shone on a patient's body, an anticancer drug selectively destroys only cancer cells. It is like a bomber dropping bombs and designating the target with a laser. Wellcome Leap in the United States said on the 16th (local time) that a joint team from Algorithmog, a quantum software corporations in Finland, IBM in the United States, and Cleveland Clinic won the top prize at the Quantum4Bio (Q4Bio) competition for research on developing photosensitive anticancer drugs.
Wellcome Leap is a nonprofit research foundation established in San Diego by the United Kingdom's Wellcome Trust. Launched in 2023, Quantum4Bio is a research competition that uses quantum computers for life science and medical research, aiming to develop quantum algorithms for bio research that can run on commercial quantum computers expected in 3 to 5 years. An algorithm is the procedure or method by which computer hardware solves problems. Software is the program that implements this as code.
◇ Simulation of the process that kills cancer cells with light
The Algorithmog consortium team used IBM's quantum computer to simulate the key mechanism of photosensitive anticancer drugs. The principle is this: an anticancer drug that responds sensitively to light is injected into the body to seek out cancer cells. Then, when a laser is aimed at the lesion from outside, the drug generates reactive oxygen species and selectively kills cancer cells. The team simulated the process in which a single drug molecule interacts with photons, the particles of light.
Of course, not every step was done on a quantum computer. That is because no quantum computer yet has that level of capability. Some simulations were done on conventional computers. Sabrina Maniscalco, head of Algorithmog, said, "We demonstrated that running the same algorithm on more powerful quantum systems can extract molecular information that is impossible with classical simulation," and added, "This algorithm can also be applied to other molecular problems, such as designing new antimicrobials."
Quantum computers and artificial intelligence (AI) have emerged as technologies that can change the paradigm of drug development. Developing a single new drug takes more than 10 years and costs trillions of won, yet the success rate is under 10%. According to the latest report from the Biological Research Information Center (BRIC), AI has already cut the average drug development timeline by 40% to 60% and reduced expense by up to 70%. That is because every pathway by which a drug interacts with the human body can be simulated on a computer.
Quantum computers can currently accelerate drug development far more than AI. Han Namsik, a professor in the Department of Quantum Information Science at Yonsei University, explained, "Conventional AI explores only one pathway at a time, but quantum algorithms, thanks to quantum properties, can simulate multiple pathways simultaneously like SONOKONG's 'cloning technique.'"
◇ Solving a 10,000-year problem in 200 seconds
Quantum computers are often called "dream computers" capable of calculations 10 million times faster than supercomputers. Computers that use AI today represent the absence or presence of electrons as 0 and 1, that is, bits. By contrast, the unit of a quantum computer is the Qubit, in which the 0 and 1 states are superposed. If a classical computer has 2 bits, it can be one of four states—00, 01, 10, 11—while 2 Qubits can be all four at the same time. If there are 300 Qubits, 2 to the 300th power states—more than the number of atoms in the universe—are possible, leading to a dramatic increase in computational capability.
In October 2019, Google's Quantum AI team led by John Martinis, a professor at UC Santa Barbara and a Nobel laureate in physics last year, reported in the international journal Nature that a random-number verification problem that would take 10,000 years on the best existing supercomputer was solved in 200 seconds on a quantum computer. It was the first moment so-called "Quantum Supremacy," in which a quantum computer outperforms a supercomputer, was achieved. The team used the Sycamore quantum chip with 53 Qubits.
The Algorithmog team developed its algorithm with an eye toward quantum computers that could be commercialized in as little as three years. For the $2 million (2.9 billion won) prize, the organizers specified a circuit depth corresponding to 50 or more Qubits and 1,000 to 10,000 computational steps. Depending on the principle used to realize quantum states, there are slight differences, but calculations at this level can be completed in milliseconds (1/1,000 of a second) to a few seconds.
Quantum computers now under development operate on different principles. The most representative approaches are superconducting circuits and ion traps. Google and IBM implemented Qubits in a superconducting state in which electrical resistance disappears at ultralow temperatures. Among the six teams that reached the finals in this competition, five, including Algorithmog, used IBM's quantum computers.
An ion trap expresses a Qubit by confining ions—atoms carrying a positive or negative charge—in a state where two states are superposed. IonQ, founded by Kim Jeong-sang at Duke University in the United States, simulated in June last year how amino acid chains fold to form a protein's three-dimensional structure using an ion-trap quantum computer with 36 Qubits.
◇ Hopes for use in decoding genome genes
AI began to be used in earnest for drug development in 2017. According to U.S. AI consultancy Intuition Lab, there are more than 173 AI drug programs in clinical development, and 15 to 20 programs are expected to enter phase 3 trials this year alone. As quantum computers advance, AI-driven drug development could accelerate further. In particular, it could enable the development of personalized therapeutics based on each patient's unique genetic information.
The genome, which contains human genetic information, is composed of about 3.2 billion base pairs. Living organisms connect amino acids according to the order of bases to make proteins that govern all life processes. If human base information is converted into digital information, it becomes 6.4 billion bits or 750 megabytes. At that scale, it can all be stored with just 33 Qubits. Even storing all of humanity's genetic information would require fewer than 100 Qubits.
Researchers at the Wellcome Sanger Institute in the United Kingdom and the University of Oxford said on the 9th that they had achieved the world's first success in loading the full genome of the hepatitis D virus onto a quantum computer. They also reached the finals of the Quantum4Bio competition with this research. The team loaded information on roughly 1,700 bases that make up the hepatitis D virus genome onto IBM's quantum computer. They converted an entire specific organism's genome into a format that a quantum computer can process.
There are challenges before commercialization. The error problem in quantum computers must be solved to use them for genome analysis, which is directly tied to life. It also has to surpass the efficiency of highly optimized conventional computing algorithms. This Quantum4Bio competition demonstrated through simulations that quantum computers can be effective in tackling difficult problems in medicine and life sciences. The quantum era of bio is drawing near.
References
Q4Bio (2026), https://wellcomeleap.org/q4bio_prize_announcement/
Wellcome Sanger Institute (2026), https://www.sanger.ac.uk/news_item/genome-loaded-onto-a-quantum-computer-in-world-first/
arXiv (2025), DOI: https://doi.org/10.48550/arXiv.2506.07866
Nature (2019), DOI: https://doi.org/10.1038/s41586-019-1666-5