In the pharmaceutical industry, where innovations like Artificial Intelligence Market and machine learning accelerate and enhance the accuracy of drug discovery and development, advanced technology is now commonplace. Add the powerful class of computers known as supercomputers, which are frequently utilized for data-intensive scientific endeavors and are far superior to general-purpose computers in terms of speed and performance.

But what implications does this extremely potent technology have for medicine?

Cambridge-1, the UK's most powerful supercomputer, was launched this year by US tech company NVIDIA to help UK healthcare researchers use powerful AI and simulation to solve pressing, large-scale medical challenges.

Cambridge-1, the result of a $100 million investment made by NVIDIA, will enable researchers to gain a deeper comprehension of brain disorders like dementia, make use of artificial intelligence (AI) to design novel medications, and more precisely identify variants in human genomes that contribute to disease. The projects will be carried out with the assistance of the computer's initial users: King's College London, Oxford Nanopore Technologies, Guy's and St. Thomas' NHS Foundation Trust, and AstraZeneca

NVIDIA's EMEA industry lead for Artificial Intelligence Market and life sciences, Craig Rhodes, says that medical research was the obvious target for Cambridge-1's processing capabilities.

Rhodes explains, "We've got enough data, sometimes too much data, but computation is the bottleneck right now." Therefore, we wanted to eliminate that computational bottleneck as part of our investment in a cure for Covid and are currently researching cancer cures. With all of the data we have, we can do things much better; When it comes to dealing with clinical change, drug discovery change, and various clinical practices, we can produce results that are significantly more efficient and effective.

According to him, "Us deploying Cambridge-1 is showing clinical and drug discovery organizations and organizations doing large, national population sequencing that supercomputing will give them an answer much faster and much more accurately" by removing the computational bottleneck.

"Taking all of those dimensions into account: NVIDIA has a one-of-a-kind opportunity to go and do something really, really good and worthwhile given the UK's status as an AI hotbed, its investment in key UK organizations, and its desire to accelerate the search for cancer and COVID cures.

For more insights on the Artificial Intelligence Market , Download a Free Sample

Data supercomputing can help researchers learn more about cells and potential compounds in the early stages of drug development, where it can be of great benefit, according to Rhodes.

He explains, "It's really important to look at microscopy data, look at the structure of compounds, and figure out how that compound might change subtly when it locks into a target." However, in order to accomplish that, a lot of computational power is required.

The cutting-edge technology comes into play here. In a recent study, researchers at the Institute for Bioengineering of Catalonia in Barcelona found that instead of taking months to process microscope data, using machine learning instead took just seconds.

Additionally, Cambridge-1 has the potential to significantly accelerate and optimize the later stages of drug research. This will make it possible to process large volumes of valuable pathology and genomics data, which will improve researchers' capacity to generate genetically validated drug candidate targets.

Rhodes asserts, "If we just had two whole genomes and we were looking at a drug that we are trying to use for a specific condition, then it is not a very good representation of the whole world." It won't give us a very clear picture of how this person might react to this drug if they take it with another person who has the same genetic information as us.

"This has been seen a lot; We saw it with the Covid-19 vaccine, where some vaccines worked very well on some demographics and others didn't,” he says. We can understand some of the subtleties in those differences by understanding the genetic makeup, but doing so computationally is extremely difficult.

According to Rhodes, one human genome is about one gigabyte, or about the size of a short film. He says that processing a lot of genomic data is like having a computer look over and analyze "every single scene, voice, motion, interaction" in thousands of Netflix movies at once. Supercomputers like Cambridge-1 make tasks like these, which are extremely challenging for general-purpose computing systems, a breeze.

Scientists will be able to process millions more samples than a conventional computing system could handle thanks to the processing capabilities of supercomputers. This will allow for a better representation of diverse populations and a more precise determination of which drugs are most likely to be successful. Drug companies save time and money by doing this; Researchers will be able to create clinical trial cohorts that are more diverse, precise, and, as a result, more likely to respond favorably to the drug candidate thanks to the computer's recommendations based on data.

Using Cambridge-1 to "teach AI models to generate synthetic brain images by learning from tens of thousands of MRI brain scans, from various ages and diseases," two of NVIDIA's partners, King's College London and Guy's and St. Thomas' NHS Foundation Trust, are aiming to improve diagnosis and treatment. It is hoped that the AI-based data model, which can generate an infinite number of never-before-seen brain images with selected characteristics, will facilitate an earlier diagnosis and treatment of brain diseases.

Numerous therapeutic areas benefit from enhanced imaging processing. According to Rhodes, NVIDIA's technology has the potential to greatly benefit oncology.

Let's say we are attempting to treat a specific kind of lung cancer. Lung cancer comes in many different varieties; We are aware of some types of lung cancer, but not many others,” he explains. Because we do not know what that cancer was, people continue to die.

"We want to look at hundreds of thousands of lung cancer pathology images, so the system has to go through hundreds of thousands of images to figure out what's different and what's the same." What sets this apart? What seems reasonable? What seems strange?”

Although pathology image processing is time-consuming and laborious for conventional computers, Rhodes claims that these are straightforward activities.

He adds, "We need to look at hundreds of thousands of specimens if we then say we want to look at the genetic data for this drug, for lung cancer." so that we can determine from the DNA signature that this group of patients with this particular type of lung cancer share this particular characteristic.

“Does this imply that this medication will be effective in conjunction with something very specific if they have it? Or does this imply that it will not function properly or at all?

"And now, what is getting better is to actually look at the image and then predict where to look in the DNA data based on the image."

Although a pathologist can quickly identify cancerous cells on an image, the type of cancer being displayed or the disease's current stage are not always obvious.

“However, if we are able to go into the DNA sequence data of that patient and identify where the particular markers we’re expecting for these types of cancer are,” Rhodes explains, "then we can then go, "actually, we can see there’s a genetic issue here, or this genetic makeup would determine that it’s this particular type of cancer."

He asserts that even if DNA analysis of a patient yields no relevant or useful information, this result may indicate that researchers are dealing with a new type of cancer. Additionally, this finding would be useful in determining which treatment strategies might or might not work for a particular patient.

According to Rhodes, "the activities that we are doing at this level, they appear to be very, very simple – it is watching, it is looking – but they are hugely important to the outcome, whether it is a patient or a drug going through a particular process."

Is the future of research supercomputing?
There are other supercomputers being used for medical research in addition to Cambridge-1. Marconi100, a supercomputer built in Italy by IBM and NVIDIA, was made available to the School of Chemistry at the University of Nottingham this year. “Harness the power of both machine learning and of physics-based molecular simulation to accelerate the discovery of compounds with predicted therapeutic values as leads in drug discovery efforts,” according to Jonathan Hirst, the professor in charge of the project.

In another location, Japanese researchers are utilizing Fugaku, the world's most powerful supercomputer, to accelerate the discovery and creation of individualized therapies. Fugaku has already contributed to efforts to identify existing medicines that can be used to fight Covid-19, reducing more than 2000 potential drug candidates to just a few dozen, despite only officially launching in March of this year.

In 2019, eight European locations were chosen to house supercomputing centers with the intention of assisting in the development of personalized medicine, drug design, and material design, among other applications. The European High-Performance Computing Joint Undertaking, which was established to provide the EU with "a world-class supercomputing infrastructure by the end of 2020," allocated the centers.

The high-performance computing required to truly optimize drug discovery and development must be widely available if supercomputers are to truly transform medical research. High-performance computers like Cambridge-1, designed specifically to support researchers in their endeavors, are a significant step in the right direction in democratizing advanced technology across the industry.

Rhodes states, "We want people to understand that computational blockages or challenges with computation shouldn't affect scientific research." We've done this to show the UK government what could be done and the extremely valuable assets that UK businesses are currently producing in the UK.

What fundamental changes will supercomputing bring to clinical research in the future? He continues, "I don't know the answer to that, but I do know that we're giving them the assets they need to really make a big change." When I think back on my life, I will truly believe that NVIDIA was the driving force behind our impact on clinical change, the lives of patients, and the NHS in this country.