"You always have to start by asking yourself, did evolution spend a lot of time optimizing this? If yes, my job is going to be insanely hard. If no, potentially there are some lowhanging fruit. And so ..."

">> Right? By the way, just on that, often people who are trying to forecast AI will discuss basically how hard did evolution try to optimize for intelligence and what were the the things which optimiz..."

">> So in one way this is actually a very interesting RL problem, right? It's a long horizon RL problem 20-year horizon length and then there's a scalar value of how many kids you have. I guess that's ..."

"longevity is selected for and does arise in modern man and in which we are optimal. And so I think that puts human aging and longevity and health really in this category of problem in which evolution ..."

"antibiotics. Yeah, it's actually an excellent question that I haven't heard posed before. Um so we think about where do antibiotics come from? To your point, we could synthesize them. They're just met..."

"terms of number of genomes per unit resource they're consuming. There are trillions of bacteria in a drop of water that you might pick up. So there's trillions of copies of the genome, massive analog ..."

"evolved around it. Do we see evidence of these like historical antibiotics that some fungi came up with and the bacteria evolved around and all there's evidence for remnant in their DNA? >> I'm I'm go..."

"binds a indogenous retrovirus that is no longer present but was at one point actually resurrected by a bunch of researchers and it was demonstrated that it is the case. We have this indogenous gene wh..."

"a situation where you can go in and take human cells and make just a couple edits in that trim 5 alpha gene and it's currently protecting against a virus which no longer exists and you can edit it bac..."

">> Isn't the mutation rate per base pair per generation like one in a billion or something? >> It's quite low. So you're saying that in our genomes we can find some extended sequence which encodes how..."

"really happens because I've got my backup copy, my original. And so you can end up with drift. So you're saying that even though the per base pair mutation rate might be one in a billion, if you've go..."

"about for memory, but it's in like the tens. It's not it's not that you need massive kilobase scale rearrangements. It's actually a fairly small number of edits. And basically you can just align the s..."

"Okay. Um, back to aging. You'll cancel your evening plans. I've got so many questions for you and I keep going. Um, so the second reason you gave which was that there's selective pressure against peop..."

"are still going to experience for the first medicine some amount of decline over time. And this gives you an example of if you think about evolution as a medicine maker in this sort of anthropomorphic..."

"All right, back to Jacob. All right, so um uh evolution didn't select for aging. What are you doing? What's your approach at New Limit that you think is um is likely to find the true cause of aging? >..."

"going after this by trying to find combinations of these transcription factors that are able to actually remodel the epiggenome so that they can bind to just the right places in the DNA and then shift..."

"also measure any potential detrimental effects that that emerge. >> So there are canonical examples where you can seemingly reverse the age of a cell for instance at the level of a transcriptto but si..."

"question, but it will help me understand why an AI model is necessary to do any of this work. So, you mentioned the Yamanaka factors. From my understanding, the way he identified these four transcript..."

">> So we can think of these transcription factors as these basis directions and you get like a little bit of this thing, a little bit of that thing, and some combination. And evolution has designed th..."

"would be uh maybe a little bit overstating to say evolution is like using the gradient, but there is a system kind of like uh if you've heard of evolution strategies where uh basically the way you opt..."

"engender upon biology. >> Yeah, you're you're sort of hinting that uh if we analogize it to some code base, we're going to find a couple lines that are like commented out that's like daging, you know,..."

"what's occurring. And so that ultimately leads to transcription factors being some of the final aectors in these signaling cascades. So a lot of the drugs we have that for instance inhibit a particula..."

"They're just too small. And then the other classic modalities we have are recombinant proteins. We make a protein like a hormone in a big vat. We grow it in some Chinese hamster ovary cells. We extrac..."

"we need to start addressing transcription factors as first class targets rather than treating them as like uh maybe some ancillary third order thing that's going to happen. >> Interesting. So the drug..."

"order primitive. Ultimately, the genome's nucleic acids. The RNAs that come out of it are nucleic acids. So, if you can get nucleic acid into a cell, you can drug pretty much anything in the genome ef..."

"history and very large population sizes. They've evolved to get into our cells. Maybe we can learn something from them. Even better Trojan horses. So, one type of virus people use a lot, it's called a..."

"we're probably going to have to solve delivery the way that our own genome solved delivery. So we have this same problem that arose during evolution, which is how do I patrol the body, find arbitrary ..."

"will get solved. We've got many many stepping stones along the way. But if I could like clone myself and work on an even riskier endeavor, that's probably what I would do. >> This is actually in in I ..."

"delivery folks credit, they're currently ahead. There are currently no reprogramming medicines for aging and there are medicines that deliver nucleic acids. So like they're still winning the race agai..."

"same time hs are another example where there are many circumstances where one of the and I'll summarize this is uh mostly examples taken from a wonderful book by Frederick Applebomb who uh trained wit..."

"exist. >> Is this related to why has so many downstream positive effects that seem even not totally related to its um effects just on making you leaner? Yeah, I think it's one one example because it's..."

">> Interesting. How big will the payload have to be? >> Uh, how many transcription factors? >> Yeah, >> I think just a countable number. I think some of those that we found today that have efficacy ar..."

"either. >> And is it would it have to be a chronic treatment or could it just be a onetime dose? >> In principle, it could be one time. I think that would be an overstatement for today, but I can sort..."

"data that these positive effects can last several weeks after a dose. And so you could imagine even without many leaps of faith up toward this upper bound limit of what's possible just from the data w..."

"okay so I don't know your skin starts to sag because of the effects of gravity over the course of decades. Is that a cellular process? How would how would some cellular therapy deal with that? >> The ..."

">> Interesting. Okay. What is Arum's law? Iram's law is a funny portman who created by a friend of mine Jack Scanell where he inverted the notion of Moor's law which is the doubling of compute density..."

">> So, you need to then be able to find medicines that treat most people. All of us will one day get sick and die. So, arguably the TAM for any really successful medicine could be everybody on planet ..."

"Yeah. In this particular case, if we we bound ourselves to we must use small molecules as our modality, then there are lots of targets which are very difficult to drug. There are many other modalities..."

">> every marginal discovery increases the odds you make the next discovery or something like that. >> So there are multiple ways one might approach this problem. the most common today. Um, this is oft..."

">> Right. And you basically described um one of the models you guys are working on at New Limit. You you're training this model based on this data where you're taking the entire transcriptto and just ..."

">> This is so similar to in LLMs. You have first imitation learning with pre-training that builds a general purpose uh representation of the world and then you do RL about a particular objective in ma..."

"Potentially. It's more like developmental biologists locked in a room as my friend Cole Trapnel would say. >> Um, it seems like what you're describing seems quite similar to perturb. And we've had per..."

"ask questions like, well, I've turned on genes A and C. What did it do to the rest of the cell? So that's the general premise of the technology. And so it's useful to just set that up because it expla..."

"They've all improved to the degree where now you can actually operate at scale. And then groups like ours have had to do a bunch of work in order to actually enable combinatorial perturbations, turnin..."

"right now is at least at New Limit you were like we know the end ca end use case we're going after it would be like if um cursor or whatever is like we're gonna in like 2018 is like we're going to bui..."

"world has ever seen. Um and over time we can expand to this more general corpus of predicting every possible gene perturbation in every possible cell type. And so I think that's maybe the way the anal..."

">> Interesting. And then um this is more more a question about the broader pharma industry rather than just new limit which is that in the future how are people going to make money if you have you kno..."

">> The the reason I'm interested in this is that healthcare is already 20% of GDP. I think it's grown like notable percentages in the last few years. It's like this is a fraction that is quickly growi..."

"these are things that biotech probably isn't going to be able to solve as an industry alone. That's probably a larger a larger economic problem. But when you think about how will this affect sort of t..."

">> Okay, final question. Um, so pharma is spending billions of dollars per new drug it comes up with and surely they have noticed that the lack of some general platform or some general model has made ..."

"something like 70% of molecules approved in a given year come from originally small biotechs rather than large pharmas even though you look at the actual like dollars of R&D spend on the balance sheet..."