Chai Discovery

“They reason quite literally by placing individual atoms in, in three D space. And often they're getting the resolution of these structures, the error down to less than the width of one atom. When we look at the error across the entire structure.”

“We ordered actually only 14 sequences to the lab. And I think four of those were histohumans. One of those was a hit to the Sino. One of them was actually overlapped and hit both. That one now allows us to move forward with that program and gives us a whole am…”

“So we actually wrote a scraper that would go and see what was in stock. We would go and pick out the protein. We would go look at what that protein sequence was. Now we need to make sure this is held out from training as well. Right? So we would take that, tha…”

“We were actually only targeting a success rate of one percent. That was the company-wide goal for the entire year.”

“I think if you see our mini protein results, we are, I think, close to 70% on those with picomolar affinities, like really, really tight binders for every single target that we tested. So all five targets we tested worked, and 70% of the designs that we ordere…”

“We have about a two week validation cycle in the lab, and two weeks later, we see that roughly close to 20% of these antibodies actually bind their targets in the intended way.”

“We actually even have a slide in our paper in the supplement where we actually look at an even harder subset. So not looking at things that are, you know, up to a sequence similarity with the model, but actually pushing all the way down to 25%. So really looki…”

“Almost every AI bio company before us has had some kind of very tight lab integration with what they are doing. And it almost too tight.”

“One piece of intuition around that is that you can almost think about structure prediction as The ImageNet moment for the field, where with structure prediction, we are asking a model to go from sequence to a predicted structure, and it's sort of like a classi…”

“The fact that we can get antibody hits in just 20 attempts, there's no reason that, that we couldn't generate intra drug candidates in that same number of attempts.”

“The model is not perfect. You know, it worked in 50% of the targets that we tried. Maybe it would have been more, right, for the caveats we talked about before, but, you know, it worked in 50% of cases.”

“I think you just learn that unless somebody is really taking care to keep the entire system in their head and is an effective technical steward of the architecture, that things just evolve and the sort of the entropy of the software takes over and slows down y…”

“We've literally had to do this in Chai's history, but we've had to go and bisect Git history. Run launch training runs, you know, with a sort of a binary search to identify a small enough range of pull requests to identify a bug, then go to that, that, that pu…”