December 6, 2023
Barry Bunin, PhD
Founder & CEO
Collaborative Drug Discovery
Super Supercomputers: Get Ready to Speak in Quintillions. Sometime next year a new breed of supercomputer is expected to come on line that will break the two quintillion mark, according to The Wall Street Journal. “That’s two billion billion operations a second, enough to explore the brain and discover drugs,” the article notes. To help visualize that, a quintillion is one followed by 18 zeros: 1,000,000,000,000,000,000. Imagine doing two of those per second. Housed at the Energy Department’s Argonne National Laboratory, the supercomputer, called Aurora, is among a new breed of machines known as “exascale” supercomputers. In a single second, an exascale computer can perform one quintillion operations, which Aurora is expected to double. This kind of computational horsepower should prove valuable to pharmaceutical research. The article says, “Researchers recently used Aurora to screen 22 billion drug molecules an hour, accelerating potential drug discovery.”
"Race for First Drug Discovered by AI Nears Key Milestone.” That’s the headline in a recent Bloomberg report about Insilico Medicine’s use of AI to develop an experimental drug for the incurable lung disease idiopathic pulmonary fibrosis. The treatment is in mid-stage trials in the US and China with some results expected early 2025. The article says, “The results of Insilico’s trials are being closely watched in the drug industry because the company used AI to identify a new approach to fight against the deadly disease and produce a novel molecule to treat it.” The article, however, also notes that while pharmaceutical companies worldwide are investing heavily in AI, there is still no clear evidence the technology can generate life-saving therapies. A string of other leading molecules that relied on AI have faced setbacks and Insilico’s could still fail in the process or take years to reach the market. When the super supercomputers, noted above, are able to screen 22 billion drug molecules an hour, and power AI to analyze the results, such milestones may disappear in the rearview mirror. As with any new technology, there is a combination of both hype and utility. The latest CDD Vault Deep Learning module uses our patented AI/ML algorithms to rapidly search ChEMBL via chemically rich vectors within the secure CDD Vault environment - see: https://www.collaborativedrug.com/deep-learning”
* * *
“Can Some CAR-T Therapies Cause Cancer?” That’s the question facing researchers after a recent FDA report that the innovative cancer treatment may sometimes cause cancer. The potentially lifesaving cancer treatment CAR-T, first approved in November 2017 for life- threatening blood cancers, is under scrutiny after the FDA said it had received 19 reports of new blood cancers in patients who received the treatment. The findings may not be that disruptive to the drug’s use. The New York Times quotes Dr. John DiPersio, Director of the Center for Genetic and Cellular Immunotherapy at Washington University School of Medicine in St. Louis, as saying his center had treated 500 to 700 patients. And, he said, “I haven’t seen a single one” develop a new T cell cancer.” The article notes that CAR-T therapy has been reserved for patients who would die without it. Dr. DiPersio says: “They are all going to die and they are all going to die quickly without this treatment. It saves their life. It works in a substantial portion of patients. The benefit is enormous.”
* * *
Will AI Become Hypothetically Speaking?. Can AI move beyond data analysis to form scientific hypothesis? That’s the question Derek Lowe explores this in his recent blog in Science. Lowe points to a report in Nature on the possibility of AI-driven hypothesis generation, and sees big promise—if it can be accomplished. He writes: “The idea is that you look over a lot of experimental results and come up with a "What If" idea that might explain them....-what if there were another regulatory pathway, hooked up to this thing over here, that also affects the pathway that I'm looking at? What if the reaction that I've found is actually being catalyzed by trace impurities in the stuff that I've been thinking is the real catalyst? What if there's something wrong with my cell cultures, some infection or contaminating organism, and that's why my results are much more inconsistent than they should be? What if this enzyme that I'm studying also works on a totally different substrate in a different tissue, because that might explain the odd in vivo effects of inhibiting it?
* * *
Blocking a “Come Here” Message to Cancer Cells Could Stop or Slow Metastasis. Discovery of a new type of stem cell—with a protein that acts as a “come here” signal to tumor cells—could lead the way to blocking or slowing metastasis, especially to the spine. The Washington Post, carries an article about research published in Nature, on a new type of stem cell that could help explain why when breast, lung and prostate cancers metastasize to multiple bones in the body, three to five times more cancer winds up in the spine than in the lower and upper limbs. C. Rory Goodwin, a neurosurgeon and spine surgeon at Duke Health, says “The skeletal stem cells in the vertebrae have a surprising capacity to attract cancer cells that you don’t see in other skeletal stem cells.” Matthew B. Greenblatt, one of the study’s authors and a pathologist at Weill Cornell Medicine, says, “We predict this discovery will lead to the targeting of these cells to disrupt the function and ultimately reduce the spread of cancer to the spine.”
* * *
Barry A. Bunin, PhD, is the Founder & CEO of Collaborative Drug Discovery, which provides a modern approach to drug discovery research informatics trusted globally by thousands of leading researchers. The CDD Vault is a hosted biological and chemical database that securely manages your private and external data.