Ep162: Sean McClain on AI for Biologic Drug Discovery

Episode Summary

Executive Summary: Sean McClain describes how he founded AbSci as a scrappy, bootstrap-style biotech to engineer E. coli into a cheaper, faster antibody-production platform, then expanded into AI-guided biologics discovery. The conversation traces his entrepreneurial upbringing, the company’s pivot from services to platform-enabled drug discovery, and AbSci’s current push to launch a TL1A antibody candidate into the clinic amid both AI hype and biological uncertainty.

Main Topics: Entrepreneurial upbringing and family influence (Priority: 5/5): McClain credits his father’s leadership and risk-taking mindset, and his mother’s creativity, for shaping his entrepreneurial drive and belief that science is a creative act. Early business lessons and selling skills (Priority: 4/5): As a child, he started a lawn-mowing business, learning courage, customer outreach, and the importance of selling—lessons that later translated to biotech fundraising and BD. University of Arizona and shift toward biotech (Priority: 5/5): He entered college aiming for engineering but became fascinated by biochemistry and synthetic biology, realizing biology could be engineered like a system. Building AbSci’s E. coli antibody platform (Priority: 5/5): McClain explains how AbSci engineered E. coli to produce complex biologics by tuning redox state, transcription, translation, and folding conditions to mimic mammalian-like output. From platform services to drug discovery (Priority: 5/5): A protein-affinity finding from a client project revealed that the ACE assay could screen not just yield but affinity at massive scale, pushing AbSci into internal drug discovery. AI’s role and limits in biologics discovery (Priority: 5/5): McClain argues AI can design biologics with desired attributes today, especially when paired with high-quality wet-lab data, but cannot yet predict biology itself or fully choose targets. Company growth and clinical ambitions (Priority: 4/5): AbSci has become a public company with 200 employees and more than $530 million raised, and it is preparing its first TL1A antibody candidate for clinical testing in 2025.

Key Arguments: Entrepreneurship is learned through action, risk, and repeated selling, not just credentials; McClain’s lawn-mowing business and cold outreach foreshadowed his biotech path. Science is creative and aspirational: building a therapy starts with believing a solution is possible before the data exists. E. coli can be engineered to produce complex biologics at lower cost and faster speed than mammalian systems for certain modalities. AbSci’s platform created value because it solved real manufacturability problems and could reduce cost of goods and development timelines. The discovery of an affinity signal in the ACE assay showed the platform could be used for biologics design, not just expression optimization. AI becomes powerful in drug discovery only when paired with high-quality wet-lab data; the company’s advantage is generating that data at scale. Current AI can help design biologics against hard targets, but it cannot yet fully predict biology or determine the right target for disease. AbSci’s strategy is to use AI and wet-lab iteration to open up previously inaccessible biology and move toward clinical assets.

Data Points: Founding age: 21 - McClain started AbSci shortly after graduating from the University of Arizona. Graduation year: 2011 - He graduated from the University of Arizona in 2011. Initial lab space: 200 square feet - AbSci began in a small basement lab in Portland, Oregon. Startup equipment budget: $40,000 - He used roughly this amount to buy surplus lab equipment and launch the company. Early revenue timing: 18 to 24 months - This was the period spent engineering E. coli before securing early customer engagement. Number of pharma companies that responded to cold outreach: 3 - McClain cold emailed top pharma BD contacts and received interest from three large pharma companies. Proof-of-concept proteins: 5 - A pharma partner asked AbSci to test five protein modalities in its proof-of-concept study. Affinity correlation: 0.9 - ACE assay affinity readouts correlated strongly with gold-standard SPR/BLI data. Throughput comparison: millions vs. ~1,000 antibodies per week - McClain contrasted AbSci’s high-throughput screening with conventional affinity testing. Potential COGS reduction: 25% to 50% - He cited modeled reductions in drug-substance cost of goods for some proteins. Potential manufacturing time reduction: 50% - AbSci claimed it could cut manufacturing timelines roughly in half. Company headcount: 200 employees - AbSci is described as a public company with a substantial employee base. Capital raised: >$530 million - McClain has raised more than $530 million as founder and CEO. Series E financing: $65 million - The company raised a $65 million Series E led by Casting Capital and Red Mile.

Pivotal Quotes: "In order to create a therapy for patients, you actually have to believe that it's possible." — Sean McClain: On the creative mindset required to pursue drug discovery before the data exists. "Every protein has its own personality." — Sean McClain: On why different biologics require tailored folding and expression solutions. "AI is, we're not at the point where AI can predict biology, but where we're at right now... is the ability to actually design biologics with the attributes you want." — Sean McClain: On the current practical power and limits of AI in drug discovery.

Implications: The episode shows how platform biotech can emerge from scrappy beginnings when paired with experimentation and persistence. For listeners, it highlights a pragmatic AI future: useful for design, not magic, and still dependent on strong wet-lab data.