Andrew Buchanan

speaker

155 appearances 1 recordings 1 series first heard Jun 2026 last heard 22 Jun

Andrew Buchanan’s voice in public audio — every appearance, attributed to the second.

Trend

recordings per month · last 12 months

Recordings per month over the last 12 months — 1 in all, peaking in Jun 2026 with 1.

Appearances

newest first · ▶ plays the moment

Talking Techniques · AI & Antibodies miniseries | Designing smart antibodies in the age of AI · 22 Jun 2026 podcast

So that's a single molecule pull down readout that doesn't require antibody purification.

And that enables them to measure

IgG affinity, aspects of developability, surprisingly robustly, and they can measure that for one to ten thousand antibodies per week.

And in the paper they provide

Some brilliant figures that help illustrate the fitness landscape of expression and affinity and show how they don't overlap terribly well.

But by using some structure informed AI so they combine Alpha Fold III and protein MPNN.

They are able to efficiently navigate this very rugged and bumpy landscape to predict, identify and confirm new antibody variants that are effectively better C D P matches than the already good starting antibodies.

So the novelty here is in this wet lab platform that can generate and visualize data that enhances the antibody optimization beyond what any old school team could do in that time frame.

And it really reminded me of an some an earlier paper by Ben Frobeski in twenty twenty three, which again illustrates the benefit of high quality, high throughput data.

And I think that's beginning to give it a glimpse of how the wet lab platforms will help define the next generation of molecule design with AI.

Mm.

I think today my today, my view is that the wet lab platforms and the expert insight that's better integrated and informed

with AI tech will define the next generation of antibody designs and the target selections, because choosing the right targets and translating that mechanism of action forward towards patients does remain the hardest of the challenges, as it's ultimately the human biology that decides on how good this approach is.

So as these smart molecules move closer to real clinical readouts, the AI label will matter less and the execution and quality will matter more.

So while there is a lot of attention rightly on AI and drug discovery platforms, that's not really still where the real value is being built.

And we see that with biofarmers commitment and links to AI Deepen.

So Novart is the CEO, joins the Anthropic Board.

We see Enterprise Solutions, Merck bring in Google's Gemini, Novo.

uh bringing open AI and that's to enable teams to make really high stakes development decisions faster.

Document them more clearly and defend them under regulatory and commercial scrutiny.

Showing 121–140 of 155 · page 7 of 8