Show Notes
“The future of biology is programmable.” To give you a sense of how important this topic is, Patrick Hsu is the first guest we’ve had back to Ground Truths since the podcast began a few years ago. Nearly 2 years ago we spoke about digital biology but so much has happened since to help unravel the complexities in the language of life. Patrick is a co-founder and core investigator at the Arc Institute. His work is at the intersection of biology, engineering, and AI. At the ripe old age of 33, he’s a pioneer of genome editing technologies and my go-to for tracking progress in digital biology. Besides that he’s a great explainer with frequent use of metaphors and simplification of very complicated stuff.
Here’s what we discussed with the relevant links: —Our recent review in Nature Biotechnology on the large language of life models (LLLMs). GBAI-generalist biological artificial intelligence
—The 2 multi-agent AI Nature papers in May for accelerating drug discovery: Full House (Robin) and Google Deep Mind (Co-Scientist) Multiple drug candidates were described as an outgrowth of multi-agent AI , with a 200-fold decrease in workflow time. Below Figure from synthesis of the Robin and Co-Scientist papers: “AI devises hypotheses and ways to test them,” Nature News and Views, by Oliver Elemento
—Biomni published in Science 9 July An infographic by NotebookLM below helps give you a sense of what—a diversity of important tasks— accomplished by Biomni
—New workbench offerings by the AI companies including Anthropic’s Claude Science, Open AI, Google DeepMind and the competition with academic spinoffs —Spud Cells —Virtual Cell project, which it’s such a big deal, and the importance of perturbations —Alzheimer’s disease with OpenAI for finding mechanistic new targets for drug discovery —The Nature Editorial —Mid-Journey’s Ultrasound Imaging of the “Whole-Body” (and we’re not imaginative enough, the need to think big) —Mitigating the worries about humans as lab assistants and nefarious use to produce bioweapons ******************************************************************************* A big thanks to Ground Truths subscribers from every US state and 212 countries. Your subscription to these free essays and podcasts makes my work in putting them together worthwhile. If you’re not a subscriber, please join! If you found this interesting PLEASE share it! Paid subscriptions are voluntary and all proceeds from them go to support Scripps Research. They do allow for posting comments and questions, which I do my best to respond to. Please don’t hesitate to post comments and give me feedback. Let me know topics that you would like to see covered. Many thanks to those who have contributed—they have greatly helped fund our summer internship programs for the past two years. It enabled us to accept and support a record number of 61 summer interns here with us now! These are high school, college and medical students selected from thousands of applicants. We couldn’t do this expanded program without the funds coming in through Ground Truths. Finally, thank you Alan Heldman MD, Bob Fleischman, Jeanie, Maureen Susannah, Mary Powers, and more than 400 others for tuning into my live video with Patrick Hsu! Join me for my next live video in the app.
Get full access to Ground Truths at erictopol.substack.com/subscribe
More from Ground Truths
Michael Lewis is one of the best storytellers of our era. You know and likely have read his books Moneyball, The Big Short, The Blind Side, Flash Boys, Liar’s Poker, and The Undoing Project, to name a few. He has long been one of my favorite authors. This was the first interview that Michael did on his new book, BLOCKERS, published on October 6th, streamed live on Substack Sept 29th. It’s his 21st book and 4 of the last 5 have been about the US government: The Fifth Risk (2019), Premonition (2021), and Who Is Government? (2025). His ability to talk with the silent civil servant heroes of multiple government agencies (the Blockers), members of the DOGE destruction team, and much more from the current US government administration.
In recent weeks the results (announced by press release) of 2 major placebo-controlled randomized trials—ZEUS, testing an interleukin-6 (IL-6) blocker vs inflammation, and HORIZON, testing a drug to lower Lp(a), indicated failure to achieve their primary endpoint. Are we maxxed out for new treatments vs heart disease, as this article in the WSJ (←gift link) today asserts? “Cardiovascular drug development is starting to look a bit like the modern oil industry. The easy wells have been tapped. What’s left is in deep water: expensive, technically perilous and less likely to strike oil.”
This podcast is meant to complement my recent Ground Truths post Why Our Memory Slips After Age 50 which has all the relevant links to the 2 papers, key graphics, and other related publications. If you haven’t had a chance to read that, it would be helpful to spend a few minutes to do so before listening to this podcast. Links to the 2 papers for those who don’t want to look back Zemke et al in Science https://www.science.org/doi/10.1126/science.adt8307 Belk et al in Nature https://www.nature.com/articles/s41586-026-10939-0
Deb Cohen is an award winning investigative journalist and correspondent who trained as a physician and has worked with the BBC and BMJ. She just published BAD INFLUENCE, a book which takes us deep into the wellness-industrial complex—the confluence of the supplement, testing, tech, and pharma industries, and how they sell health anxiety to healthy people.
Sarah Urbut is a physician-scientist at Harvard Medical School and the Massachusetts General Research Institute. She and her colleagues published a landmark paper in Nature on patient health journeys for 3 different cohorts—the Mass General Brigham patients, the UK Biobank, and the All of Us research program—with a cumulative 683,000 individuals, 348 diseases, and up to 52 years of follow-up. Integrating electronic medical records and polygenic risk scores (N=36). An AI model was developed and validated called ALADYNOULLI, with each individual represented latent disease signatures (N=21), mathematically and biologically driven. An example of a patient’s dynamic life journey (health arc) is provided in the Figure below (adapted from the paper, made with NotebookLM). The longitudinal story for each individual is key here. Using a Gaussian process, the risk continually shifts like a GPS navigation when you make a wrong turn and you are re-routed. Considerable effort was dedicated to addressing selection bias with inverse probability weighting.