Aubin Ramon

speaker

346 appearances 1 recordings 1 series first heard Jul 2026 last heard 6 Jul

Aubin Ramon’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 Jul 2026 with 1.

Appearances

newest first · ▶ plays the moment

Talking Techniques · AI & Antibodies miniseries | Nanobody thermostability prediction and the great data challenge · 6 Jul 2026 podcast

So what we did is that we have a library, like in the case of I don't know, like an in vitro library.

Um and you uh up screen your

library against your target and then you have a hundred, one thousand good candidates you want to take into the lab to do more characterization.

And so we uh measure the predicted emerging temperature.

with nanomer.

Then we selected three um highly stable variant around seventy degrees.

And free pool is stable variant around sixty degrees.

So there's like a ten degrees different here.

Um 10, 15 degrees different.

And then we were able to show that when we try to do further characterization, so to purify, express and purify the protein, we struggle.

We didn't were able to express the free-pool variant.

Whereas the three highly predicted variant were able to express, and the measured magic temperature was the one we predicted, at like one degree difference.

It can it can be useful um as a first triage.

Um I would say.

So you have a list of um variant and such a a gross a predictor can tell you okay which one to keep going with, which one is disparate.

And then if you have heat that you really, really like but the melt the predicted temperature is is bad, you can then flag it for further engineering um stability optimization.

Yeah, so that was a very interesting um observation and we were a bit disappointed by that.

Because we're thinking like, okay, we have uh language model uh fine-tuned train on antibodies, they're gonna perform better for an antibody property, for a property uh on of on nanobodies, antibodies.

Um and actually no, so our first intuition was is due to the fact that thermostability is uh

shared property among every food, every protein um

Showing 121–140 of 346 · page 7 of 18