a16z Podcast: What's in the Water at the George Church Lab?

Episode Summary

Executive Summary: George Church and Jorge Conde discuss the ethics, science, and future of genome editing and sequencing, using the Chinese CCR5 CRISPR babies as a springboard. Church argues the edit was medically plausible but ethically premature, and broadens the conversation to privacy, preventive medicine, whole-genome sequencing, and his lab’s strategy of turning “science fiction” into real technologies and companies.

Main Topics: CRISPR babies and germline editing ethics (Priority: 5/5): Church analyzes the Chinese CCR5 embryo edits, arguing the intervention was technically modest but ethically and regulatorily fraught. He says the target was medically defensible in concept, yet premature for implantation and public trust. CCR5 biology, mosaicism, and risk/benefit (Priority: 5/5): The discussion explains CCR5 knockout as a way to reduce HIV susceptibility, while noting mosaicism, possible incomplete protection, and tradeoffs such as possible increased vulnerability to other infections like West Nile. Regulation, ethics, and the pace of biomedical change (Priority: 4/5): Church emphasizes that regulatory oversight is essential and tends to shape and accelerate technological progress rather than merely slow it. He frames ethics as a mechanism for smoothing adoption, not stopping innovation. Whole-genome sequencing, privacy, and public-health utility (Priority: 5/5): The conversation returns to Church’s long-standing bet that whole-genome sequencing will become broadly useful once it is cheap, encrypted, and better understood. He argues current barriers are cost, privacy, and weak value perception. Personal Genome Project and informed consent (Priority: 4/5): Church describes the PGP as an early effort to make genomic participation transparent, consent-driven, and shareable like Wikipedia. He argues participants need 100% comprehension of risks and benefits, not legalistic forms. Church Lab culture and the science-fiction-to-reality pipeline (Priority: 4/5): Church explains how his lab recruits multidisciplinary, kind, risk-tolerant people and repeatedly pursues ideas that initially sound impossible, from sequencing to editing to future multi-omic and high-multiplex genome engineering. Future directions: diagnosis, prevention, and synthetic biology (Priority: 4/5): Church highlights underused opportunities in diagnosis and prevention, including sequencing for pathogens/allergens, IVF/PGD, matchmaking for genetic compatibility, mammoth hybrids, and large-scale editing for novel biology.

Key Arguments: The Chinese CCR5 edit was less a precision edit than a gene knockout; CRISPR is currently better at damaging genes than making exact changes. Germline editing in this case had a plausible medical rationale because CCR5 loss can reduce HIV susceptibility, and HIV remains a major global disease burden. Mosaicism exists in many gene-therapy contexts; some level of mosaicism does not automatically invalidate an intervention if enough corrected cells remain functional. Ethics debates often slow things temporarily but also attract attention, investment, and technical progress, so they can accelerate development over time. Regulation is necessary because it prevents harmful biomedical failures and helps normalize new categories like gene therapy, cell therapy, and digital therapeutics. Whole-genome sequencing will matter only when it is effectively free, private/encrypted, and clearly linked to actionable preventive care for the 1-4% who have serious findings. Most people do not need to know their genomes every day; the value is public-health scale—identifying the small subset for whom sequencing changes care dramatically. The Personal Genome Project was designed to fix weak consent by requiring participants to understand risks fully and to enable broad, open sharing of genomic data. The Church Lab succeeds by recruiting nice, multidisciplinary people and by pursuing radical, high-upside projects well before they are mainstream or company-ready. Future step changes in medicine will come from better diagnosis, prevention, and high-multiplex biology, not just incremental therapeutics.

Data Points: Annual HIV deaths: 900,000 - Used to justify the medical motivation for targeting CCR5 in germline editing People affected by HIV/AIDS: 37 million - Church cites the global burden to explain why prevention matters Fraction of certain European populations with CCR5 double null: Up to 10% - Referenced as natural precedent for CCR5 loss of function Properly edited T cells in adult trials: As little as 20% - Church notes this mosaicism can still be clinically sufficient in some adult HIV gene-therapy trials Population likely to get a blank genome report: 95-96% - Church argues most people will not have highly actionable Mendelian findings People with major impactful genome findings: 1-4% - Used to frame sequencing as a public-health tool for a minority with high stakes Mendelian-disease care cost: About $1 million per person - Cited to show why prevention and alternative strategies can save insurers and families money Genome sequencing cost: Below $1,000 / toward $0 - Church says utility will expand when sequencing becomes effectively free Genome price reduction over time: From $3 billion to about $7,000 - Used as evidence of the Church Lab’s sequencing impact Pig genome edits: 62 edits - Example of the lab’s progress in multiplex genome engineering Single-cell edits: 10,000 edits - Cited as a new milestone in large-scale editing

Pivotal Quotes: "I would say there's three reasons why it hasn't happened yet: one, cost. The cost should probably be zero dollars. Secondly, it's privacy. ... And the third is most people don't understand the value proposition." — George Church: Explaining why whole-genome sequencing is not yet universally adopted "We’re not looking for geniuses. We’re looking for people that are nice." — George Church: Describing the Church Lab’s hiring philosophy and culture "What CRISPR does well is often described as editing. It really is damaging, is what it is." — George Church: Clarifying the technical nature of current CRISPR interventions

Implications: The episode frames genome editing as powerful but still immature, and sequencing as a coming public-health infrastructure. For industry, privacy-preserving genomics, preventive medicine, and high-multiplex engineering are the biggest near-term opportunities.