Marc Tessier-Lavigne

speaker

407 appearances 1 recordings 1 series first heard Mar 2026 last heard 12 Mar

Marc Tessier-Lavigne’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 Mar 2026 with 1.

Appearances

newest first · ▶ plays the moment

Vanguards of Healthcare by Bloomberg Intelligence · Xaira’s War on Trial-and-Error Drug Development · 12 Mar 2026 podcast

So that is certainly one route.

There are various assays.

There's been a big effort by the industry as a whole over the past decade to develop assays that enable you to determine whether a protein is immunogenic.

Some of it involves in silico analysis.

Some of it involves, for example, looking at uptake of the protein that you've made, the synthetic protein, by the sentinel cells of the immune system, the dendritic cells.

There are other in vitro assays like that that you can use to determine whether your protein is immunogenic or not, and then you can modify it to fix it.

The beauty of AI, of course, is that you can not just modify to fix if the antibody that you've made in the first instance is immunogenic.

You can also generate massive amounts of data on synthetic antibodies to determine which ones are most likely to be immunogenic and which ones are less likely to be immunogenic, and then bake that into your model so that from the get-go, you create antibodies that are less likely to be immunogenic.

So there are a number of properties of antibodies that you want to make a good antibody in, one that might be

obvious is you don't want the antibody to aggregate because it might just fall out of solution.

And that's something that you can teach with appropriate data.

So some of the physical chemical properties of antibodies, we are generating data to teach our models how to create antibodies that don't have those liabilities.

I think the hardest one is the one that you identified, which is immunogenicity, because we don't have a perfect way of reading that out.

For physical chemical properties, we can read it out in the lab, so we can generate massive amounts of data and have a high degree of assurance that we'll be able to overcome that problem altogether using AI.

For immunogenicity, it's going to be semi-empirical initially.

but we think we can make a lot of progress.

And again, the combination of AI-driven modifications to the proteins and perhaps still some art of looking at the sequences and saying, well, maybe this is more likely to be immunogenic with the kind of implicit knowledge that people have developed over the past several decades is what will happen in the first instance.

Over time, I think we will overcome that too.

But you're right that that is probably one of the biggest areas that we'll have to attend to.

I think there will be lots of opportunities for both big pharmas and small companies.

Showing 261–280 of 407 · page 14 of 21