Using quantum computing and supercomputer technologies to forward drug discovery
According to McKinsey & Company, introducing a new drug to the market takes up to 12 years—a long period influenced by several elements including technology constraints, legal requirements, patient recruitment and retention. More data than ever are produced by increasingly complicated scientific inventions and clinical research. Turning to artificial intelligence, machine learning, and advanced data architecture to increase data processing, user experiences, and outcomes, researchers are struggling to manage the volume and variety of data.
One less talked about issue in the drug research and development process, nevertheless, is the capacity of compute power to not only satisfy growing data needs of new and sophisticated clinical studies but also progress the necessary data infrastructure. Leveraging past decade’s tech innovations, technical hardware improvements are likely to drastically raise the computation, storage, and data transfer capability capacity. Future technologies not possible today across mainstream sectors, including drug discovery and clinical research, will be enabled by ongoing momentum.
Prospects in Clinical Trials for Supercomputers and Quantum Computing
Recent advances in compute capability enable scientists to tackle challenges too difficult for conventional computers. For the scientific and research community, the May 2022 debut of the Frontier, now regarded as the fastest supercomputer in the world, was revolutionary. Operating with a capability of 1.1 exaflops, this supercomputer technology has been in charge of processing data at astonishing rates—one quintillion calculations per second. For the life sciences sector, this is revolutionary since it allows scientific teams to test new discoveries faster and handle vast amounts of data. Still, a quantum computer may tackle a difficult mathematical problem—like Shor’s algorithm—a million times quicker than the fastest supercomputer already in use.
Many quantum computing innovations over recent years have opened the path for a hybrid compute method of hitherto unheard-of speed and complexity. This covers IBM’s breakthrough in quantum computing, which was reported in the scientific journal Nature concerning noise reduction and error minimisation in quantum qubits. Using the IBM Quantum “Eagle” quantum processor (with the power of 127 superconducting qubits on a chip), IBM finally solved a challenging problem that leading supercomputing approximation methods could not handle for years allowing the team to generate vast amounts of power that simulated the dynamics of spins to precisely predict properties such as its magnetisation. More recently even, we have seen technology behemoths like Microsoft and Quantinuum declare a quantum field breakthrough. By use of Microsoft’s error-correction algorithm applied to Quantinuum’s physical qubits, the two attained a record of logical circuit error rates 800 times less than their corresponding physical circuit error rates.
These successes hasten the path towards a future in which researchers could use highly performing computers to solve formerly intractable issues containing billions of data points, such molecular and atom simulations.
Ongoing Development of Storage and Data Transfer Capacity
Researchers using a machine learning algorithm to examine data from more than 16,000 clinical trials found in a study released by The National Library of Medicine (NIH) that the average complexity score across all trials jumped by over 10 percentage points in the past decade. Apart from the complexity of trials, it is clear that clinical trial teams of today deal with an unheard-of number of data. With an average of approximately 3.6 million data points, Tufts Centre for Study of Drug Development (CSDD) determined in 2021 that phase III clinical trials generated 300% more data points in the past ten years.
Life sciences businesses are looking for solutions to handle this data deluge by automating digital data
From intake to analytics, flows help to speed data cleansing and decision-making processes, hence encouraging quicker insights. Here is when using data architecture’s ability to be powerful comes forward. Though it establishes consistent methods for gathering, storing, manipulating, and distributing meaningful data for its consumers, data architecture is sometimes undervalued.
But the capabilities of data architecture must be evolved enough to handle, store, and centralise vast volumes of data—often originating from many sources—enabling clinical researchers to fully leverage it to derive relevant insights. Classical computing systems have limited processing capacity; but, in recent years we have witnessed significant advancements on the data storage and data transfer sides, therefore advancing the evolution of data architecture.
Under the direction of assistant professor Stephen M. Wu, a team of University of Rochester scientists created hybrid phase-change memristors offering super-fast, low-power, high-density computer memory in the field of data storage. These kinds of developments can create ultra-fast and efficient computer memory, therefore increasing the possibilities for the data volume that could be kept, accessed and used in clinical trials.
Regarding data transfers, Technical University of Denmark in Copenhagen created a single computer chip capable of 1.84 petabits of data per second—that is, downloading more than 200,000,000 images in one second. In the context of clinical research, a conventional trial nowadays produces under 1 terabytes of data from beginning to end—an amount that may be sent in a few hours over a standard connection. A medium-sized study can produce petabytes of data in precision medicine, where genomes data is gathered and kept in variants per patient. One petabyte needs 90 days for a standard connection data transfer. As these hardware developments remove the technological obstacles of data storage and processing it will release the possibilities of what can be done with the data of personalised medicine.
Opening Novel Drug Discovery Breakthroughs
Molecular dynamic simulations, quantum chemistry computations, genomic and bioinformatics, artificial intelligence and more will all be transformed by technological innovations, which will be especially important in the most intricate and data-driven spheres of drug development process. Recent developments are already changing drug discovery by providing before unheard-of capacity to find possible drug targets and generate new treatments. We will expand the access to answers in these volumes of data as we keep seeing the development of computing, data transfer, and storage capacity, enabling researchers to find novel therapies and accelerate timelines, thus obtaining remedies to patients sooner.
