Closing the loop in biology. Currently @FlagshipPioneer. Co-founder @LilaSciences. Manifesto: https://nitter.gq/t.co/nxm4usQGlQ
Joined July 2020
- Tweets377
- Following320
- Followers812
- Likes838
Pinned Tweet
9/Here's a manifesto exploring these ideas—feedback loops, Eroom's Law, AI-guided optimization, and what it means for the future of biotech.
If this resonates, I'd love to hear from you.
Can’t wait to read this
Tragedy formed our modern clinical trial system. Opportunity has the potential to form the future of our clinical trial system.
With this piece, we @CompoundVC walk you through the case of thalidomide that was the lightning rod for adverse event reporting, investigation new drug applications, and efficacy trials before drug approval.
We also discuss impactful regulatory updates coming to clinical trials in the US which could lead to faster feedback loops.
In our field's ambition to eventually cure all disease, we must learn from the past to make a better future.
Jake Feala retweeted
I've been hanging out with this rock star team at @generate_biomed for the past few weeks and while this result is worth a deep dive, the way they work is the real game changer. They've allowed AI to remove the friction in the organization. As AI-pilled as I am, it's taken some adapting to transform to the way the future will look, but there is no catering to the 'old' way.
And that's the only way the true impact of AI will be seen, not just in the frontier labs, but also in the AI native biotechs working to bring real cures to patients fater.
Things are going to change quickly and the hardest part is going to be how people and systems adapt to move at speed. And this team is doing it.
Congrats @jasonyi96, @liambai21, @wwang002, @KaggenA!
What will drug discovery look like when intelligence is abundant?
Our team at @generate_biomed began with this question, and we’re excited to share some results!
We took on in vivo CAR-T and designed a scaffold 70% smaller than ones in the clinic, with 3x the potency 🧵
This was a surprise and updated my thinking. Would have expected this level of performance improvement to have come from a fine-tuned protein model and lots of training data specific to the task, not off-the-shelf frontier LLM.
Great work @generate_biomed and @AnthropicAI !
Jake Feala retweeted
What will drug discovery look like when intelligence is abundant?
Our team at @generate_biomed began with this question, and we’re excited to share some results!
We took on in vivo CAR-T and designed a scaffold 70% smaller than ones in the clinic, with 3x the potency 🧵
Jake Feala retweeted
Making medicines is really hard. Excited to share some early work from the team @generate_biomed with our partners @AnthropicAI on how we can leverage agentic tools to rapidly advance science. While it is still early days, the signals are very promising and patients are waiting.
Over the past few months, Anthropic scientists have engaged in deep research partnerships with six organizations developing new medicines. We've also begun drug discovery efforts of our own.
In doing so, we’ve learned much more about how Claude can support the many stages of drug development.
In this blog post, we share some early results from these efforts, and describe how this work fits in with our wider ambitions in the life sciences.
anthropic.com/research/early…
Jake Feala retweeted
okay I’m just going to rapid fire this because I’ve gotta sleep and also this is just like talking as if you were a close friend:
-like half of bio is ~wrong, like the problem is us not AI- misleading in vitro models, misleading murine models and then all of our “rules” are highly context dependent and not applicable in at least 20% of in vivo contexts
-also everything is baked with people and politics which leads to certain narratives/models in the paradigm sense winning when they maybe shouldn’t- publishing is beyond corrupt, holding people’s livelihood ransom based on whether they generate positive results and hide the negative is just a bad system
-lit search synthesis by AI is therefore not at all close to god, just sort of a condensation of issues
-but it’s still good and bad- like astra max got thymic involution at peak human adolescence wrong on monday but I’ll have grad students use it for questions I don’t want to be bugged with
-protein folding tells you v little abt function- like timing, location, ptms, nearby molecular species, idrs, local environment etc
-no one is building a model that integrates different echelons of regulation for actual in vivo tissues
-everyone needs to undiscover the cell/discover tissues and organ systems and overall interacting physiology
-human data should be step 1 in everything
-idk as far as experimental stuff/basic research gaps- automation and scaling aren’t issues it’s just like if your goal is solving human health what are you starting at? I would love for RNA biology to be solved first- I’m still biased towards building an actual causative model of life
-I think this shit could all be way more variable based on the individual than anyone is ready to deal with but everyone violently disagrees with me on that- but like my baseline basophil count is in the CML range and I’m am biased as a weirdo
-I hate non-durable single target based therapies in oncology for solid tumors with a passion and in my schizophrenic head I’d rather have biological understandings that allowed us to develop actual curative treatments
-to be positive ML based DNA encoded libraries are absolutely sick and I have new molecular glues in like the picomolar range which means tox will probably be an issue but yolo
Jake Feala retweeted
People who cured / are curing their own cancer on X:
1/ @sytses - bone cancer, billionaire founder of Gitlab, developed 10+ personalized therapies for himself.
2/ @paul_conyngham - shrank his dog Rosie's mast cell cancer tumors after chemo and surgery weren't enough. Now backed by YC and Founders Fund
3/ @miriamgonp - stage 4 breast cancer, tumor sequenced, searching for vaccine manufacturer now
4/ @kevinace - stage 4 lung cancer, angel investor in 300+ companies, first drug stopped working after 18 months. Vaccine designed, now paying $300k out of pocket for the vaccine
5/ @masonbgrant: going founder mode on his his 32-year-old wife's rare liver cancer, sequencing her tumor with AI agents
6/ @ArunRLverma: former product leader at Atoms and Shopify, brain tumor, built a vaccine from his tumor. Now building an AI health agent startup based on what he went through
We're going to make it
Yup, “competence without comprehension” (Dan Dennett’s term) is the rule, not the exception, in almost every real-world endeavor.
So it’s best to get good at a thing by trial and error first, before trying to understand it.
Jake Feala retweeted
From clinical trials to personalized therapeutic engineering
The clinical trial apparatus is very obviously not working for patients:
Complicated, Kakfa-esque inclusion and exclusion criteria.
Minimal combination opportunities.
Randomization to arms that made sense when the trial started, but not anymore.
Years from trial start to readout.
Getting on a trial is like walking a tightrope where any one of ten things can immediately disqualify you.
Luckily, the successor (personalized therapeutics engineering) is rising. Find what's attackable and unique in a tumor, and engineer a therapy or set of therapies to attack it.
This paradigm is still very small. But it's growing, and at maturity, it will supplant trials.
The beauty is that where trials are "a million reasons to say no," therapeutics engineering is "one reason to say yes." Take whatever information you have on the tumor, and develop therapeutics that target that set of things.
Very early days, but I expect this to be mainstream a few years from now.
Still a lot to build and figure out, but this transition will be fantastic for patients with tough cases.
Jake Feala retweeted
Hugely frustrating fact about cancer:
The best centers (Sloan Kettering, Stanford, UCSF, etc) typically run patients through in-house genomic diagnostics that are incredibly limited (small gene panel; no RNAseq).
So many of our cases are folks who've had MSK-IMPACT or Stanford STAMP and UCSF 500 that missed obvious things. They're well behind Caris and Tempus, and miles behind Valius.
They're at the cutting edge of everything else; very sad that they're so behind here.
Great list! I’d expect #8 to be solved first - very amenable to high-throughput assays and closed-loop optimization with existing (sequence-based) protein design models
In light of the progress in mathematics, we at Edison Scientific and FutureHouse have assembled a set of Millennium Problems for Biology. They are chosen to be very hard to solve but very easy to validate in a simple laboratory environment. Any of these, if solved, would mark a major advance in biotechnology, and most of them would contribute materially towards curing disease. These are, in some sense, the “last reasonable eval” for AI in biology. This was work primarily by @MichaelaThinks and myself, with contributions from many others.
Short descriptions below. The full descriptions of the problems with acceptance criteria are at the Bio Millennium Problems website, linked in the next post. Share more if you have ideas. If they meet our criteria, we’ll add them to our list (with attribution and permission).
Jake Feala retweeted
In light of the progress in mathematics, we at Edison Scientific and FutureHouse have assembled a set of Millennium Problems for Biology. They are chosen to be very hard to solve but very easy to validate in a simple laboratory environment. Any of these, if solved, would mark a major advance in biotechnology, and most of them would contribute materially towards curing disease. These are, in some sense, the “last reasonable eval” for AI in biology. This was work primarily by @MichaelaThinks and myself, with contributions from many others.
Short descriptions below. The full descriptions of the problems with acceptance criteria are at the Bio Millennium Problems website, linked in the next post. Share more if you have ideas. If they meet our criteria, we’ll add them to our list (with attribution and permission).
I'm a cancer genomics expert and I want to analyze sequencing data from my friend's cancer, in my spare time and with my personal account.
Will I be granted access? Will report back with the answer.
Today we’re opening applications for the Life Sciences Verification Program.
Through the LSVP, life science professionals can use our models—including, for the first time, Mythos—with a new set of safeguards designed to enable the full range of biology-related work. We designed these new safeguards to provide a better experience for biologists and more protection from risk of misuse.
The program is launching in beta for teams of all kinds—from academic labs to startups, pharma companies, and more. We will continue to improve the program and expand access to individual Pro and Max plans over time.
Learn more about these access grants and apply: anthropic.com/news/life-scie…
Human cells badly want to form tissues, can't in the typical, open-loop lab approaches of fixed recipes and delayed readouts. Polyphron closes the loop to give cells the nudges they need, in real time, to reach their goal.
Huge congrats on the raise!
Today we’re announcing we’ve raised $20M in seed financing led by @QuietCapital , with participation from @GradientVC , @haystackvc and @CompoundVC and other amazing folks to build the verification layer for AI-driven biology.
We are also delighted to be welcoming @vqctran who is joining us from @GoogleDeepMind as a co-founder.
AI is not going to cure all diseases without a verification substrate - a scalable environment where models can test predictions and learn against living human biology.
AI is becoming increasingly capable of generating biological hypotheses, therapeutic candidates and experimental designs. But generating ideas is advancing much faster than our ability to test what those ideas will actually do in human biology. Existing systems force a tradeoff: the most biologically relevant approaches are slow and expensive, while faster in vitro and computational systems often lack the fidelity needed to capture human response.
Polyphron’s bet is that tissue is the fastest, cheapest, most parallelizable verification substrate that still contains the biology that matters and we believe the path to biological superintelligence runs through closed-loop interaction with these living and simulated human systems.
Thank you to @axios for covering the announcement.
Link to our new website in the thread.
Jake Feala retweeted
Actually, even stronger than this: while people are worrying about theoretical engineered pandemics, more than 1% of the population is already dying every year to disease and old age. Safety is important, but accelerating medicine is the best way to save human lives.
Jake Feala retweeted
I’ve expected this for a while but it still hurts. The “Life Sciences Verification Program.” It breaks my heart to know that biological knowledge will be locked down and gatekept like this. I want to take a moment to mourn, because I believe this marks an end for open science as I knew it.
I think back to my education. I owe so much to the open-access literature. I'm so deeply thankful to the culture of transparency among scientists. All the free knowledge I've been given. The free GenBank sequences. The free PDB structures. The education and know-show that my colleagues shared with me for the pure love of learning.
The Life Sciences Verification Program will kill that spirit. The most important biology tools of the next decade will be locked down and controlled. Some fragments of the old literature will remain free, but the real heart of where science happens will not be free. Frontier knowledge will belong only to people who the authorities decide deserve it.
I want to fight this, but I don’t think I can. I don’t believe bioterrorism is a serious threat, but convincing people of that is not realistic. The fear is embedded too deep in the psyche. If I tried, there are armies of internet trolls eager to accuse me of evil things. It would become my entire online life.
There is a growing and well funded fear-industrial complex dedicated to the idea that Bio-AI is a terrible risk and the people who practice it need to be monitored and policed. Hundreds of full-time professionals whose salary depends on their belief that biology presents a dire threat. What hope do I have of arguing with them?
On the other side, the traditional defenders of free information are historically weak. Biology professors are struggling just to keep the lights on. The last thing they want is to be called reckless enemies of national security. They aren’t going to fight this - I’m sure many of them are already signing up to ask for permission to be on the list.
So I’m writing this to bear witness. This is how it ended. Biology isn’t going to be free going forward. That was an idea of the pre-AI era and it is over now. The fight now isn’t over free access to information. It’s over who gets verified access to the closed information.
Big pharma companies get permission, of course. So do professors at elite universities. But we can petition Anthropic to expand that list. Maybe they might allow access for high school teachers, medical students and nurses? I hope they will approve farmers, iGEM teams, community biolabs, birdwatchers, park rangers and community gardens.
Will Anthropic allow those people to study biology at the frontier? As of today, the power officially belongs to them. We can only ask.
Today we’re opening applications for the Life Sciences Verification Program.
Through the LSVP, life science professionals can use our models—including, for the first time, Mythos—with a new set of safeguards designed to enable the full range of biology-related work. We designed these new safeguards to provide a better experience for biologists and more protection from risk of misuse.
The program is launching in beta for teams of all kinds—from academic labs to startups, pharma companies, and more. We will continue to improve the program and expand access to individual Pro and Max plans over time.
Learn more about these access grants and apply: anthropic.com/news/life-scie…
Jake Feala retweeted
In sum, when you actually know something about building a laboratory, lab automation, and what goes into synthesizing a virus and testing its properties, it becomes clear that AI does not impact this very much.
At best, it provides bad actors with a quicker way than the internet to learn about the stuff I described (which machines, which lab protocols, etc.). But it does nothing to impact procurement timelines, existing industry safeguards and regulations on procurement for lab facilities, the ~$100M cost (plus high operating expenses), the physical limits on experimentation velocity, or the fundamental knowledge gaps in virology that cannot be solved merely by 'smarter' AI without iteration in the physical world.
