Mendelspod Podcast

Join Gavin Knott, an associate professor at Monash University and an alum of Jennifer Doudna’s lab, as he dives into the fascinating world of RNA and AI in gene editing. Knott reveals RNA as a dynamic language and how AI tools like AlphaFold are revolutionizing protein design. He discusses the exciting potential of AI in reducing off-target effects in CRISPR technology and opens up about the challenges in the field. With his infectious curiosity, Knott envisions a future where proteins are designed with intent, pushing the boundaries of molecular biology.

This is a free preview of a paid episode. To hear more, visit www.mendelspod.comFor decades, drug discovery has followed two main roads: small molecules that can slip into cells but often lack precision, and biologics that offer specificity but struggle with delivery and manufacturing. In this episode, Christian Schafmeister introduces us to a third path: spiroligomers—synthetic, fused-ring molecules designed to combine the best of…

What if drug design didn’t depend on pre-existing biological data? In this episode of Mendelspod, Theral sits down with Aridni Shah, co-founder and CEO of Immunito AI, a Bengaluru-based biotech startup reinventing how we develop antibody therapies. Shah and her team are using artificial intelligence to design antibodies from scratch, bypassing traditional animal-based and data-heavy methods. “We no longer need to rely on pre-existing data,” she explains. “We’ve gone to the fundamental atomic level and learned basic physics, chemistry, and maths—why do two proteins interact?”* 0:00 Discovering new antibodies without relying on biological processes* 4:45 The platform - a structure first approach* 10:00 The business model* 15:05 Bringing mathematics and physics to biology At the heart of Immunito’s innovation is a proprietary mathematical transformation that translates raw 3D structural data into rich, generative models capable of designing antibodies for even “undruggable” targets. It’s an ambitious approach to rational drug design that prioritizes data efficiency, leveraging physics-based complementarity rather than brute-force sequence matching. Still early in development, Immunito is already seeing promising lab results and aims to partner with pharma while building its own pipeline. “We should have signed a few multimillion-dollar multi-target deals,” Shah says of their goals over the next few years, “and be looking to progress towards IND.” This conversation opens a compelling window into how AI, when grounded in mathematical and physical intuition, may radically transform biology. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.mendelspod.com/subscribe

This is a free preview of a paid episode. To hear more, visit www.mendelspod.comYes, the Trump economy has been tough on life science tools—but according to today’s guest, the business model was failing us anyway. It’s time to go back to rethinking the basics.Today we’re joined by Stephane Budel, founding partner at DeciBio, one of the leading market research and consulting firms focused on life science tools and precision medicine…

This is a free preview of a paid episode. To hear more, visit www.mendelspod.comOn today’s episode of Mendelspod, Theral sits down with Giovanna Prout, CEO of Scale Biosciences, to explore a how the company is achieving new orders of magnitude of scale in the single cell space."In the past, single cell was one cell per well—maybe 96 cells per experiment," Prout explains. "Now, with our Quantum Scale platform, we can scale from 85,…

What if you could step inside a molecule? On today’s episode of Mendelspod, we talk with Steve McCloskey, founder and CEO of Nanome, the company bringing virtual reality into the world of drug discovery and molecular modeling. A former nanoengineering student at UC San Diego and a self-described futurist, Steve founded Nanome to create what he calls the "ultimate scientific interface"—a way for scientists not just to visualize molecules, but to interact with them naturally in 3D space.* 0:00 The ultimate science app* 2:15 A use case* 7:15 How does VR interaction change things?* 12:47 Intuition at the nanoscale * 23:10 Can VR redefine biological models and become the norm?* 35:15 Hard to give career advice now with rapid rise of techNanome’s VR platform enables researchers to grab, rotate, and manipulate proteins, explore molecular dynamics, and even collaborate remotely inside a shared molecular environment. Built with both scientific rigor and the playful spirit of a 3D gamer, Nanome is already being used from classrooms to major pharmaceutical firms.One of the most compelling ideas discussed today is the possibility that VR can help biologists build intuition at the nanoscale, much like space exploration and science fiction helped earlier generations develop intuition about the cosmos. "Ninety-nine percent of people have no intuition for the nanoscale," McCloskey says. "You look at molecules on a 2D screen, and it's like a ball of spaghetti. But in VR, you pretty much instantaneously get it."​This new kind of "nano-intuition" could, McCloskey argues, open doors for faster discovery, better communication, and broader scientific literacy. As he puts it, "Space is fascinating because it's so vast and mysterious. But the nanoscale is even denser with phenomena—and we’re just beginning to explore it."​Looking ahead, Steve sees VR as becoming a normal part of the scientific workflow. Though today's headsets are still bulky, future versions could be as light as eyeglasses, embedding a whole new layer of spatial computing into everyday research.As Steve says: "If we can give scientists—and even the general public—intuition for the nanoscale, it might radically change how we approach biology, drug development, and science education itself."​ This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.mendelspod.com/subscribe

This is a free preview of a paid episode. To hear more, visit www.mendelspod.comWait! What? Sequencing as a service?That was the reaction when Mark Budde first set out to upend a basic ritual in molecular biology—confirmation Sanger sequencing—with his company Plasmidsaurus. You’ve heard it many times: Illumina machines sequence around 80% of the bases in the world. But as Mark points out in today’s interview, it’s not 80% of th…

Just weeks before the FDA’s sweeping rule to regulate laboratory-developed tests (LDTs) was set to take effect, a federal court struck it down. What does this mean for the field of precision medicine and for labs themselves? There are have been those on the side of regulation arguing its importance for reimbursement. Yet many, such as the plaintiffs in this case, the ACLA and AMP, have argued that FDA regulation could stifle innovation in a new field and even force labs to shut down. Today we talk with Sarah Overton, Senior Director of Revenue Cycle Management at Velsera, whose team helps clinical labs with everything from validation to reimbursement strategy. She walks us through the practical implications of the ruling. “It came just in time. Even meeting stage one was going to be incredibly burdensome,” she says in today’s interview. (Link to a webinar with Sarah on this topic here.)* 0:00 Impact of recent court decision on LDTs * 5:15 Supports an in-between regulation * 15:15 How are your clients choosing between LDT and IVD?* 20:00 Topic goes hand-in-hand with reimbursement * 25:20 Every test is differentInstead of scrapping oversight altogether, Overton argues for a middle ground — more robust than the current CLIA framework, but less rigid than the FDA’s approach. “We need something in between,” she says. “There are already mechanisms in place — like MolDX’s technical assessment — that address analytic validity, clinical validity, and clinical utility. That could be a model to build on.”We explore how Velsera’s clients are choosing between launching LDTs or IVDs, and what factors drive that decision. Overton emphasizes that reimbursement strategy must be integrated from the beginning — not bolted on at the end. “The test might be clinically excellent, but if it’s not reimbursable, it won’t sustain a lab,” she warns.“We need standards that ensure quality without crushing innovation,” she says. “There are ways to do that now. Let’s build on what’s working.” This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.mendelspod.com/subscribe

In today’s show, Theral is joined by Chris Petersen, Chief Technology Officer at Scientist.com, a company sometimes called the "Amazon for science"—though with a great deal more complexity. Chris pulls back the curtain on how AI is transforming the research services marketplace and offers a rare look into how AI is already reshaping the infrastructure of pharma and biotech.Calling this the "tinkering phase" of AI, Chris likens the current moment to the early days of the web—when best practices were still forming and every developer had to invent their own solutions. “It’s one of the most exciting times to be a software developer,” he says, describing how AI has enabled a leap forward in productivity across the board—from writing code to streamlining negotiations and customer service. One internal tool, Elisa, functions like a fine-tuned ChatGPT within Slack, answering employee and customer queries on the fly.AI, he says, is speeding up nearly every part of the business. "There are all of these problems that were impossible to solve a year and a half ago. And now you can solve them. One of the hardest things… is your old preconceptions of what you're capable of? You have to let some of that go because you're capable of so much more now."Petersen also talks risk: while AI promises a democratizing effect, making outsourcing more accessible to small players, he warns of the dangers of consolidation—where just a few massive models hold everyone’s data. To avoid that future, Scientist.com is building its own internal LLMs, training and fine-tuning models like Mistral and DeepSeek on proprietary data, all under an evolving platform they call Benchmate.Scientist is betting that AI will not only enhance their marketplace but change how science itself is organized and conducted. Stay tuned for more in this new series on AI. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.mendelspod.com/subscribe

In today’s program, we continue our series on disease subtyping with Dr. Jeffery Klco, a pediatric pathologist and researcher at St. Jude Children’s Research Hospital. Dr. Klco joins us to discuss groundbreaking work recently published in Nature Genetics, which redefines the genomic landscape of childhood acute myeloid leukemia (AML).Klco and his team have identified twelve novel molecular subtypes of pediatric AML—some driven by genetic alterations that had long gone undetected by standard computational tools. Chief among these is the UBTF tandem duplication, a complex mutation previously overlooked but now shown to be a distinct disease subtype that accounts for up to 10% of pediatric AML cases and is associated with relapse and poor outcomes.“We’ve been misclassifying childhood AML for years,” says Klco. “It’s the same disease as in adults—but it behaves very differently. And we’ve been using an adult framework to treat kids.”* 0:00 We’ve been misclassifying childhood AML* 7:25 Challenges with current treatment paradigm* 11:05 Doing clinical whole genome and RNA sequencing* 17:25 How might this new approach work for other pediatric cancers?The conversation delves into how St. Jude’s use of whole genome and RNA sequencing, paired with advanced analytics, has enabled more precise subtyping. In response, Klco’s team is already developing targeted therapies, including the use of menin inhibitors, which have shown early promise.But discovery is just one side of the coin. Implementation is another. Klco discusses the development of a new 357-gene panel—specifically designed for pediatric cancers and incorporating structural variants—that is now in clinical use at St. Jude. It fills key gaps in diagnosis, risk stratification, and minimal residual disease monitoring, especially in complex cases such as post-transplant patients or those whose tissue samples are incompatible with full genome sequencing.Asked about future potential, Klco notes that while most pediatric cancers may now be genomically defined, new methods such as long-read sequencing and methylation profiling still hold promise for sharpening diagnostic tools and stratifying risk.“Pediatric cancers are driven by different genomic forces than adult cancers,” he explains. “Even within the first 18 years of life, we see distinct subtypes emerge at different ages. If a child under three comes in with AML, I already have a good idea what subtype it might be.”It’s a compelling example of how detailed genomic subtyping is not only advancing our understanding of pediatric disease—but directly shaping the next generation of therapies. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.mendelspod.com/subscribe