The Genetics Podcast

This week on The Genetics Podcast, Patrick goes solo to look at what recent AI breakthroughs in mathematics could mean for biology. He discusses the OpenAI results on the Navier-Stokes equation and hundreds of open math problems, what makes math and code so easy for AI to improve on, how self-driving labs and surrogate endpoints could speed up drug development, and where the biggest gaps and opportunities remain.Show Notes0:00 Intro to The Genetics Podcast00:58 What the OpenAI math breakthroughs could mean for biology02:57 AI’s recent frontier math results and how mathematicians reacted09:12 Lean as a proof checker and why it matters for math10:44 Four conditions behind AI’s math progress and how biology compares16:22 Why verification speed decides where AI can accelerate biology19:35 How lab automation shortens feedback loops for variants and antibodies23:13 How validated biomarkers let early trials predict clinical outcomes26:11 Why validated surrogate endpoints could speed up drug development31:58 How startups and big pharma are building self-driving labs33:40 Virtual cell models built from perturbation data34:50 Genetics-driven drug discovery and the Phase II success rate question37:42 Verification speed and surrogate endpoints as biology’s biggest gaps40:38 Closing remarks

This week on The Genetics Podcast, Patrick is joined by Dr. Sek Kathiresan, CEO of Verve Therapeutics and SVP, Head of Obesity Research and Early Development at Eli Lilly. They discuss how protective genetic variants led to the development of a one-time gene editing therapy for heart disease in the face of early doubt, the phase one results for Verve-102, what recent trial failures mean for genetic target validation, and the unmet needs in obesity treatment.Show Notes0:00 Intro to The Genetics Podcast00:59 Welcome to Sek 01:40 Sek’s family history of heart disease and how it motivated him to pursue cardio therapeutics03:29 Sek’s background and research on the genetics of coronary artery disease and protective mutations08:03 Reactions to the Verve launch and the biological, technological, regulatory, and commercial risks11:00 Verve-102 phase one results and the case for lifelong LDL lowering16:14 Why cumulative exposure means lowering LDL earlier and for longer17:53 Recent trial failures in genetically validated targets and what the Lp(a) trial leaves unanswered20:33 Polygenic risk scores and who benefits most from LDL lowering23:36 Sek’s new role at Lilly and the remaining unmet needs in obesity treatment29:06 Whether one-time gene therapy could work for obesity and how GLP-1 drugs drive weight loss30:38 Food intake, energy expenditure, and the search for new obesity mechanisms32:15 Why Lilly is investing so heavily in genetic medicine34:35 Closing remarksFind out moreANGPTL3 study in NEJM (https://pmc.ncbi.nlm.nih.gov/articles/PMC3008575/)Verve-102 study (https://www.nejm.org/doi/10.1056/NEJMoa2601283)

This week on The Genetics Podcast, Patrick is joined by Dr. Luke Pilling, Senior Lecturer in Genetic Epidemiology at the University of Exeter. They discuss hemochromatosis and the case for earlier diagnosis, and the genetic drivers of multimorbidity, pharmacogenetics, and drug response.Show Notes0:00 Intro to The Genetics Podcast00:59 Welcome to Luke 01:37 An overview of HFE hemochromatosis and why diagnosis is delayed 04:24 Iron levels and overload risk in heterozygous and compound heterozygous carriers05:51 Aging-related effects seen in homozygotes 06:43 How hemochromatosis is currently diagnosed and detected in the UK09:26 Future screening and case finding, including the Our Future Health cohort11:28 Genetic and environmental factors that shape outcomes in C282Y homozygotes 13:28 The GEMINI project and the genetic drivers of multimorbidity15:56 Drug targets from multimorbidity genetics, including repurposing and opposite effects18:49 Using disease sequences in health records to look for earlier intervention points in multimorbidity21:21 Data gaps limiting multimorbidity research, from primary care records to proteomics23:17 Handling misdiagnosis and clinical code accuracy in health record research24:55 International collaborations and the need for more diverse biobank data26:08 Pharmacogenetics research and how genetic variants shape drug response29:04 Limited clinical examples of pharmacogenetics and lessons from non-responders33:08 Genetics of lifespan and healthy aging, and why geroscience pathways rarely appear in GWAS36:06 Luke’s future research focuses 40:39 Closing remarksFind out more:- Hereditary hemochromatosis study- Multimorbidity study- Pharmacogenetics framework study

This week on The Genetics Podcast, Patrick is joined by Dr. James Zou, Associate Professor of Biomedical Data Science at Stanford University. They discuss the concept of virtual biotechs and how AI agents link single cell data to clinical trial success, Paper to Agent and turning papers into living virtual authors, clinical AI from EchoNet to SleepFM, and the role of academia alongside big AI labs.Show Notes0:00 Intro to The Genetics Podcast00:59 Welcome to James 01:55 An overview of James’s virtual biotech and its AI scientist agents04:11 How single cell evidence behind drug targets predicts clinical trial success06:33 Behind the scenes of how autonomous the virtual biotech agents are08:06 Missing proprietary trial data as the next gap for the virtual biotech09:44 Scaling the number of collaborating AI agents as a new frontier for drug discovery12:29 How teamwork and structure evolved among the agents 15:53 Behind the idea of Paper to Agent and virtual authors19:01 What converting over 100 papers into agents revealed about reproducibility23:14 The story behind EchoNet, an FDA-cleared AI tool for cardiac ultrasound26:28 Predicting risk for over 100 future diseases from one night of sleep27:40 Transferring SleepFM from sleep labs to consumer wearables28:43 Combining multimodal health data to predict disease30:10 Longitudinal blood proteomics as the biggest data gap for health AI31:18 The role of academia as an open, neutral counterweight to big AI labs32:51 How James’s path from Cambridge math and Harvard machine learning led to biology34:41 Closing remarksFind out moreVirtual biotech paper (https://www.science.org/doi/10.1126/science.aeg6779)Paper2Agent paper (https://www.nature.com/articles/s41586-026-11044-y)

This week on The Genetics Podcast, Patrick is joined by Dr. Jeffrey Barrett, Group Leader at the Ellis Institute Finland and the University of Helsinki. They discuss using AI and federated biobank data to stratify common chronic diseases, how LLMs reading clinical notes in Finland's health registries can reveal treatment response, how AI is changing the way scientists research and train, and what the UK's COVID genome sequencing effort means for biosecurity.Show Notes0:00 Intro to The Genetics Podcast01:00 Welcome to Jeff 02:08 Jeff's research focus for his new lab, including molecular subtyping of common chronic diseases with AI06:18 Barriers to pooling biobank data across research environments for federated AI models11:00 Why Jeff expects larger datasets to make complex models better at predicting disease13:34 The three disease areas Jeff's lab will target first: cardiometabolic, inflammatory and neuropsychiatric19:41 Using LLMs to read clinician notes in Finland's health registries for treatment response24:50 How Jeff uses AI in his research and where humans still need to read and review31:08 Using Claude to understand papers and the gaps in its literature searches32:14 Jeff on AI solving open problems in mathematics and biology, and where researchers fit35:48 Simultaneous discoveries and why lab data generation still needs experts39:01 Jeff's COVID genome sequencing at Sanger and what it means for AI and biosecurity43:16  Closing remarks

SummaryThis week on The Genetics Podcast, Patrick is joined by Jenny Hsieh, Co-Founder and President of the Danon Foundation. They discuss how Danon disease affects males and females, why it is so often misdiagnosed, the current therapy landscape and research priorities, and how families can make informed decisions about gene therapy trials.Show Notes0:00 Intro to The Genetics Podcast00:59 Welcome to Jenny 01:28 An overview of Danon disease02:34 Sex differences in Danon disease and severe disease in females05:15 Why Danon disease is misdiagnosed and how it may be spotted earlier08:50 Therapy pipeline for Danon and the Foundation's global patient coordination13:46 Origin story of the Danon Foundation18:10 Research priorities and the burden of Danon disease beyond the heart22:39 Raising genetic testing awareness among ophthalmologists and cardiologists for Danon disease25:20 Recognizing broader Danon symptoms and launching the first Danon Awareness Day29:37 Informed consent and trial literacy for families weighing gene therapy33:14 Patient-friendly trial experiences36:19 Ways to support Danon research37:37 Closing remarksFind out moreDanon Foundation

This week on The Genetics Podcast, Patrick is joined by Dr. Inigo Martincorena, Group Leader at the Wellcome Sanger Institute. They discuss the discovery of widespread cancer-driver mutations in normal tissue, the role of somatic mutations in autoimmune disease, new sequencing technologies transforming the field, and the therapeutic potential of targeting these mutations.Show Notes0:00 Intro to The Genetics Podcast00:59 Welcome to Inigo 01:46 How Inigo's eyelid skin study revealed widespread cancer-driver mutations07:14 Findings from a follow-up study on esophageal tissue10:02 How NanoSeq technology scaled somatic mutation research across tissues12:34 The thyroid study linking somatic mutations to autoimmune disease17:07 How escaped B cell clones evolve into polyclonal autoimmune disease20:21 Immune gene mutations occurring in healthy aging lymphocytes21:21 Why driver mutation clones in normal tissue rarely become cancer24:16 Two therapeutic paradigms for targeting somatic mutations in disease28:13 Examples of somatic rescue mutations in the colon, liver, and blood29:20 Why clonal selection only occurs in dividing cell types31:20 The field's remaining blind spots in mobile immune cells and rare samples33:46 How new single-cell sequencing will link genotype to phenotype36:03 What Inigo has learned from collaborating across Sanger's expertise38:09 Closing remarksFind out moreEyelid studyEsophagus study

This week on The Genetics Podcast, Patrick is joined by Dr. Marleah Dean Kruzel, Professor of Communication at the University of South Florida. They discuss her personal path from watching breast cancer move through her family to testing positive for a hereditary mutation herself, the ART framework she developed for managing uncertainty, and the complexities of communicating genetic risk within families.Show Notes0:00 Intro to The Genetics Podcast00:59 Welcome to Marleah01:35 Marleah's family history of breast cancer across four generations02:48 Marleah's childhood living through her mother's five-year cancer battle04:26 Marleah's experience with genetic testing07:19 How a positive test result reshaped Marleah’s PhD research focus08:40 The central challenges of living with a positive test result 10:50 The ART framework for managing hereditary cancer uncertainty15:06 Applying the ART framework to Patrick’s experience with uncertainty and the importance of having a support system 19:53 Common misconceptions in how families communicate genetic risk23:54 The previvor identity and the chronic uncertainty it brings25:40 How Marleah has learned to accept and embrace uncertainty27:57 Balancing logic and emotion in hereditary cancer family planning31:28 How Marleah grounds her research tools in patients' lived experience33:21 How AI is eroding our capacity to sit with uncertainty35:57 Redesigning healthcare for a lifespan of genetic risk information38:10 Closing remarks

This week on The Genetics Podcast, Patrick is joined by Dr. David Hallett, Chief Scientific Officer at Recursion. They discuss the biggest shifts in drug discovery over Dave's three-decade career, where AI is genuinely transforming the field today versus where the hype outruns the reality, and how Recursion's perturbational maps have uncovered and validated a novel neurodegeneration target.Show Notes0:00 Intro to The Genetics Podcast01:00 Welcome to Dave01:57 The biggest shifts in drug discovery over three decades06:39 Where AI is delivering real wins across drug discovery today12:19 How AI-assisted trial simulation reveals which eligibility criteria to relax14:46 The three biggest reasons drug programs fail in the clinic 18:20 How Recursion's perturbational maps uncover new drug targets24:58 A four-step framework for validating a novel drug target28:44 How Recursion balances deep therapeutic focus with partnership breadth30:52 Why AI can't shortcut clinical trials, and what proof of real impact looks like34:57 The skills scientists need most in the AI era, and why trusting AI outputs starts with trusting the data39:52 Closing remarksFind out more:Recursion (https://www.recursion.com/)

This week on The Genetics Podcast, Patrick is joined by Dr. Danny Miller, Assistant Professor of Pediatrics and of Laboratory Medicine and Pathology at the University of Washington and Attending Physician at Seattle Children's Hospital. They discuss the case for making long-read sequencing the first genetic test every patient receives, the reference dataset his lab is building from the 1000 Genomes Project to resolve structural variants, publicly available methylation signatures as biomarkers for diagnosis and treatment response, and his vision for genome-informed care from newborn screening through the NICU.Show Notes0:00 Intro to The Genetics Podcast01:00 Welcome to Danny01:41 The case for long-read sequencing as first-line genetic testing02:52 Current barriers to wider use of long-read sequencing 04:14 Building a long-read reference dataset from 1000 Genomes for variant filtering06:47 How long-read sequencing can solve a missed diagnosis 08:12 The clinical case for complete telomere-to-telomere genomes10:39 What it will take to shift the clinical genetics status quo12:06 Making methylation signatures public to diagnose disease and track therapy15:58 Danny's path from programming and finance into clinical genetics18:19 Danny's lived experience with deafness and achondroplasia, and how it shapes his approach to genetic counseling21:46 Danny's optimism about AI in genomics and worry about AI in education27:15 The path to making genomes a routine part of the medical record29:47 The vision of same-day newborn genomic data guiding NICU treatment decisions34:31 Closing remarksFind out more:Miller Lab (https://millerlaboratory.com/)