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

There are probably a lot of data that are not as useful as it might appear at face value.

And there's going to be a need for additional data that's generated, again, in very tightly controlled clinical settings to address specific issues that can then be fed to the models to get the answers that we need, who's most likely to respond.

So I think there's going to be a combination of all of that.

In terms of, you know, the drug making per se, there's clearly some pharmas have data where they've looked at the structure of small molecule drugs together with proteins.

Those are going to be very valuable for further training data sets.

So there is data within the industry that can be accessed, but a lot of data will have to be generated online.

A lot of it will be bringing together certain technologies and certain AI models that are being developed in startups to bear on those data sets in collaborative actions.

I think you're going to see a whole range of situations, Sam.

You'll see some big pharmas that might be able to go it alone.

There are many that will look to partner with smaller companies that have advanced the technology to a point where they want to take advantage of that and

And the small company looks forward to being able to work with the data sets that the partner has.

I think it's a real issue.

And I think that's why, just to tie back to your previous question, why some of the data that pharma companies or biotech companies or others have,

in their vaults may not be as useful as data that are obtained in prospective trials.

It's absolutely the case that the AI will learn best if it's given both the positive results and the negative results.

And that way you can reinforce the model towards the positive results away from the negative results.

That is why many of us are generating data, even for, for example, antibody protein interactions.

There are, you know, a number of antibody structures that are present in the databases, a very small fraction, actually, of the 200,000 highly curated protein structures.

But those are obviously the ones that were successful, and we haven't seen all the ones that were not successful.

And so we are, so that has some value, absolutely.

Showing 301–320 of 407 · page 16 of 21