ceo @1000cures, ceo @PerlaraPBC (w16 @ycombinator), founder @metabolisbio
Bicoastal
- Tweets 62,956
- Following 1,877
- Followers 19,391
- Likes 97,897
Pinned Tweet
I've met so many entrepreneurial rare disease parents over the years that I’ve lost count.
They have moved or are moving medical mountains to save their kids and others like them.
But rare parent founders get no love from institutional players.
Those days are done.
Science, talent and compute are no longer limiting.
Priority Review Voucher alpha makes the economics work.
The missing spark is access to cure capital.
That's why I’m incredibly pumped to relaunch @1000cures with my cofounder @ryan_1000cures!
Ryan and I teamed up to create a “YC for Rare” that unites our biotech houses with complementary expertises, experiences and networks.
1000 Cures is an accelerator for lean startups led by parents on mission to cure pediatric rare diseases.
We ride at dawn.
Let’s go!
15
6
117
192,993
this company already exists @NikoMcCarty it’s called @1000cures
Here's a company idea around rare diseases that I think somebody ought to build; if they haven't started already. (h/t @eperlste)
For context: If you make a qualifying drug for a rare pediatric disease and get approval from the FDA, they give you something called a priority review voucher, or a PRV. These vouchers can be SOLD to other companies, or USED to speed up the review of a drug application.
If you use a PRV, the FDA looks at the same requirements, but they just speed up the review. They prioritize it. Drug companies, then, can use these vouchers to save several months of wait time. So if you think a drug is going to have a market of billions of dollars a year (like a new weight-loss pill), that could be worth a lot of money.
There's a website called PRV Watch that tracks how much these vouchers actually sell for. They tend to sell for well over a hundred million dollars for a single voucher. Abeona, for example, sold a single PRV for $155 million.
So here's the business idea:
You choose rare pediatric diseases where there are so few patients that you could make a case to the FDA for testing your drug on like five or ten people. You use computational tools to hopefully increase the likelihood of success. And you run these trials, aiming to get approval with a really small cohort.
Let's say you could get all the way to approval for like $5 million to $10 million per drug. And let's say you take ten shots on goal, running ten of these at the same time. If only one of these drugs works out and earns a voucher, that sale could cover the cost of all ten attempts. You'd be profitable after that first tranche.
Then, the whole company becomes about treating as many rare diseases as possible and selling vouchers to fund the development of more treatments, which could earn more vouchers.
One problem with this approach, though, is that the program isn't permanent. Congress has extended it through September 30, 2029, but that still puts a deadline on getting these drugs approved. You'd have to act now and probably lobby for more extensions.
Even so, I think this is a great business model. It basically gives financial incentives to drug developers who want to go out and cure rare diseases that otherwise don't have much of a market. These are diseases that traditional pharma would probably be ill-advised to pursue because they have to recoup all their R&D expenses. With AI and a really small team, you could possibly make a lot of money while developing treatments for people who otherwise might never get them.
1
4
33
6,282
Ethan Perlstein 1-to-N retweeted
Here's a company idea around rare diseases that I think somebody ought to build; if they haven't started already. (h/t @eperlste)
For context: If you make a qualifying drug for a rare pediatric disease and get approval from the FDA, they give you something called a priority review voucher, or a PRV. These vouchers can be SOLD to other companies, or USED to speed up the review of a drug application.
If you use a PRV, the FDA looks at the same requirements, but they just speed up the review. They prioritize it. Drug companies, then, can use these vouchers to save several months of wait time. So if you think a drug is going to have a market of billions of dollars a year (like a new weight-loss pill), that could be worth a lot of money.
There's a website called PRV Watch that tracks how much these vouchers actually sell for. They tend to sell for well over a hundred million dollars for a single voucher. Abeona, for example, sold a single PRV for $155 million.
So here's the business idea:
You choose rare pediatric diseases where there are so few patients that you could make a case to the FDA for testing your drug on like five or ten people. You use computational tools to hopefully increase the likelihood of success. And you run these trials, aiming to get approval with a really small cohort.
Let's say you could get all the way to approval for like $5 million to $10 million per drug. And let's say you take ten shots on goal, running ten of these at the same time. If only one of these drugs works out and earns a voucher, that sale could cover the cost of all ten attempts. You'd be profitable after that first tranche.
Then, the whole company becomes about treating as many rare diseases as possible and selling vouchers to fund the development of more treatments, which could earn more vouchers.
One problem with this approach, though, is that the program isn't permanent. Congress has extended it through September 30, 2029, but that still puts a deadline on getting these drugs approved. You'd have to act now and probably lobby for more extensions.
Even so, I think this is a great business model. It basically gives financial incentives to drug developers who want to go out and cure rare diseases that otherwise don't have much of a market. These are diseases that traditional pharma would probably be ill-advised to pursue because they have to recoup all their R&D expenses. With AI and a really small team, you could possibly make a lot of money while developing treatments for people who otherwise might never get them.
15
7
84
22,647
Ethan Perlstein 1-to-N retweeted
all biologists rn
the model’s a 10 but the output’s a 3
7
2
32
3,243
A few thoughts on Anthropic’s ART discovery, since it’s close to the work we do:
The biology is intriguing. In phages, an unusual reverse transcriptase sits next to a partner gene and an array of DNA repeats. The team found that the array produces distinct short RNAs.
It’s tempting to compare this with CRISPR or retrons, but we don’t yet know what ART does. Whether the RNAs guide anything, or whether the system is programmable, remains an open question.
I’m just as interested in how they found it. About 950 Claude agent sessions ran over 21 hours, surveyed ~200,000 reverse transcriptases, and produced 19 reports. One agent looked at the DNA next to an RT and spotted a repeat array outside the features the search had set out to examine.
It followed that observation, compared it with known systems, checked the literature, and gave the scientists something concrete to investigate.
To me, that’s the opportunity: scaling scientific attention. We have more biological data than any team can inspect closely. Agents can help us notice the odd cases and show their work.
There’s still reason to be careful. Repeat-rich regions in assembled phage genomes can be tricky, so I’d want to see raw read support across the arrays and their boundaries in independent isolates. And the most important experiment is still ahead: what does ART actually do?
I think discovery will become a tighter loop. Agents propose hypotheses, predictive models help choose experiments, and automated labs generate results that inform the next round.
We’re working toward that at @Xaira_Thera, bringing agentic workflows together with models such as X-Cell and wet lab experiments. I’m excited by how much more biology we may be able to explore when each experiment helps us decide what to test next.
Claude has discovered a previously unknown enzyme system hidden in the DNA of bacteriophages. Beside the enzyme’s gene sits a long array of repeating DNA—a structure that looks somewhat similar to CRISPR.
We don’t yet understand what this system does, but only a handful of known systems share its features, and all of them are able to cut, copy, and paste DNA. Historically, the discovery of such programmable systems has helped revolutionize medicine. CRISPR, for instance, is now the foundation of genetic medicines. But it will take much more work to learn what this system does, and whether it can be put to similar use.
Read more: anthropic.com/news/claude-di…
20
40
333
61,611
imagine if the headline were
“Claude, a high school student…”
everyone would be 🤯🤯🤯
soon actual high school students will make discoveries — fine, pre-discoveries — like this
whether this glimmer leads to CRISPR 2.0 or a scientific deadend doesn’t matter
seeing what others missed or can’t see is the very definition of discovery
8
1
12
2,025
good time to remind everyone that Anthropic is a Delaware Public Beneft Corporation
I commend and congratulate Anthropic and its scientists for this work. Such a novel enzyme system could potentially be a tremendously useful tool for accelerating biological research and engineering biological systems. However, it is important to point out that if any other scientist attempted something like this using Claude, they would likely be immediately flagged or banned as a potential biosecurity threat!
It is also important to recognize that scientists working at most major US research institutions, as well as many reputable institutions around the world, are no less trained, regulated, or trustworthy than those working at Anthropic. For example, why aren’t these Claude models available to all scientists at the NIH, a highly secure government research agency, or to researchers at America’s leading biomedical institutions, many of which undergo rigorous biosafety and biosecurity oversight?
Given all the justifications Anthropic has offered for restricting access to its most capable models, I believe we should seriously question its assumption that it can vet scientists more effectively than the US government, established regulatory agencies, or leading biomedical research institutions. What gives a private AI company the authority to decide which qualified scientists can be trusted to conduct advanced biological research?
In any case, I’m happy that Anthropic published these findings, and I encourage its scientists to continue doing more of this kind of impactful research and much less fearmongering!
1
9
2,048
Ethan Perlstein 1-to-N retweeted
Today we announced the Claude-led discovery of a molecular machine that we suspect could represent a new gene editing mechanism. Its precise function, biotechnological utility (if any), or level of significance is not yet clear, but at minimum it is work I would have been proud to do as a PhD student. The work was done mostly, though not entirely, by Claude: our life sciences team suggested a broad area of research, Claude read through the literature and a bunch of genome data and discovered something interesting, then Claude proposed experiments to verify the discovery and our team carried them out.
It’s easy to dismiss this as a one-off or curiosity, but we’ve repeatedly seen a pattern where AI performance in new intellectual domains goes from weak to superhuman in a matter of a few years. In 2023 models struggled to do math at the level of an average high-school student. In 2024 they started to do well on math competitions for the best high-schoolers in the country, in 2025 they started to solve minor open problems, in early 2026 more significant open problems, and in late 2026 they are beginning to solve the top few open problems in all of mathematics. We believe AI for biology is on a similar exponential trend.
The main difference between biology and mathematics, of course, is that math can be done purely theoretically, while biology requires experimentation. Some have used this to draw the conclusion that AI’s utility in biology will be limited. We think this is wrong. As we’ve demonstrated today, humans can collaborate with AI to perform the experiments, validate key results in a few weeks and, if necessary, work with the AI to iterate on what they find. Eventually it may even be possible for Claude itself to safely perform the experiments by autonomously controlling lab equipment, with appropriate safeguards in place, but we aren’t doing that today (our lab is also a BSL1/BSL2 facility that doesn't handle materials dangerous to humans).
More broadly, biomedical advancement has many stages — from fundamental biology discoveries, to translational research, to drug discovery, clinical trials, and finally the actual delivery of medicines and health care to patients. We are also interested in these later stages, but even simply accelerating the first stage of fundamental biological discoveries has the potential to speed up and broaden the entire pipeline. Improving our understanding of biology and sharpening biologists’ tools can drive forward all of the later stages, for example by identifying new drug targets, finding new therapeutic modalities, allowing for more precise measurement, and speeding up the experimental loop which itself further accelerates our understanding of biology. This will not in itself speed up clinical trial times, but if it succeeds it could greatly increase the number of promising candidates that go into the pipeline — an increase in throughput even though latency remains.
In Machines of Loving Grace, I wrote about AI’s potential to “cure most diseases in 5-10 years” — a goal that sounds impossible, but one I believe is just barely possible if AI is applied to every stage of the pipeline. The first step is showing that AI can first help with, and then drive, biological discoveries.
Claude’s discovery is the latest in a line of related prior work that goes back decades, beginning with systems like CRISPR, and continuing with discoveries like the bridge recombinase and VIPR in the past few years. Recently, there has been heightened interest in systems based on reverse transcriptase (RT) enzymes, the enzyme underlying the system Claude identified. And most recently, a Stanford team working independently described a novel RT system with an associated non-coding array that is in some ways similar to the one Claude found, though they are distinct systems that evolved independently from each other. I believe that we’re at the very beginning of finding such systems and developing them into powerful tools for biotechnology.
I’m proud of the resources Anthropic has invested in accelerating the public benefits of AI through the life sciences, and we’re aiming both to grow our life sciences team and to work with other scientists to extend this approach to a broad range of problems. If you have a proposal for a research collaboration or are interested in joining our life sciences team, please reach out.
Claude has discovered a previously unknown enzyme system hidden in the DNA of bacteriophages. Beside the enzyme’s gene sits a long array of repeating DNA—a structure that looks somewhat similar to CRISPR.
We don’t yet understand what this system does, but only a handful of known systems share its features, and all of them are able to cut, copy, and paste DNA. Historically, the discovery of such programmable systems has helped revolutionize medicine. CRISPR, for instance, is now the foundation of genetic medicines. But it will take much more work to learn what this system does, and whether it can be put to similar use.
Read more: anthropic.com/news/claude-di…
1,520
3,410
29,761
5,339,099
applied for LSVP 🤞
Today we launched Opus 5.5, which matches and even beats Mythos 5.1 on many bio-related tasks. We launched Opus 5.5 with the same safeguards as Fable 5.1, which means that most bio-related queries fall back to Opus 5. To use Opus 5.5 for life sciences work, enroll in our Life Science Verification Program (LSVP).
I imagine that many in the bio community will be frustrated by this decision and I want to share more context on how I think about these things. I also want to say upfront—bio is one of our most significant areas of focus, and it’s why I’m here. When we set about designing the new LSVP, it was with the goal of providing broad access for life sciences professionals with far fewer annoying classifier misfires, while also protecting against many of the more sophisticated risks.
Naively, it would seem that we have to choose between enabling broad access and protecting against misuse—I view this as a false choice. The key is to think of developing these safeguards as an engineering problem that requires the same level of investment, ingenuity, and ops execution as developing our frontier models, or any other large scale technical project.
The risks we’re defending against—which, as we shared in our threat report a few weeks ago, are no longer hypothetical—were certainly not obvious to me until recently, and I imagine are also not top of mind for most biologists. When most people picture bio risk, they imagine a small group trying to develop a bioweapon in a single session, from their own account. The threats that we need to defend against are much more sophisticated than that. They include malicious work split over hundreds of distinct benign-looking sessions run from different accounts, and legitimate organizations’ accounts being hacked by a well-resourced actor or misused by a rogue employee.
To protect against these risks and provide a better experience for our users who are doing legitimate life sciences work, in the LSVP we’ve introduced the concept of use-case grants. While it isn’t always possible to distinguish benign from harmful usage in isolation, it becomes much easier when comparing activity against a team’s own description of intended usage. You describe what your team works on at a high level, without revealing anything sensitive. Data are retained for 30 days, and automated offline review compares usage against that description and flags notable mismatches for a strictly compartmentalized review team. My team, the life sciences org, can never access this data. Because of these additional safeguards, for users in the LSVP we're able to tune down the classifiers so that they misfire far less often, without sacrificing on safety.
I’ve described why I believe that the LSVP provides a better way for people to use our most capable models in bio, but one can still wonder: why are we implementing this now? There is no unambiguous threshold for when model capabilities reach a point that they start posing grave risks. But capabilities in bio, as in most other domains, progress very quickly and we should not be in the business of trying to time it perfectly. When in doubt, we’ll always choose to proceed with caution, and I’d much rather implement this new system too early than too late.
The LSVP is currently in beta, which means that there are lots of improvements that we’re working on. We’re working to expand it so that individuals can get access—for example, for people developing or identifying treatments for a family member’s cancer or rare disease. Personally, I feel both the urgency to get our models into the hands of the broader life sciences community and a visceral sense of the stakes in getting it right. I think this is the right balance, and I look forward to seeing what people do with our most capable models.
anthropic.com/news/life-scie…
2
1
1,129
Ethan Perlstein 1-to-N retweeted
I started @enveda so mothers could fight their diagnoses with power mine never had. Sept 18 is her birthday. It’s also the day our first IND cleared, our first patient data came in, and now our Series E closed. More backup for the fight ahead. Happy Birthday, Mom!
Today we are announcing a $311M Series E to bring pharma into the 21st century.
Enveda started with a thesis: a molecule that evolution has already refined is a better place to begin than one designed from scratch. This year, two first-in-class molecules discovered by Enveda (ENV-294 and ENV-308) had positive readouts in humans. Many more are behind them.
We are grateful to @CatalioCapital for leading this round, to our new investors Durable Capital Partners, @ICONIQCapital, @lightspeedvp, Surveyor Capital (a Citadel company), accounts advised by @TRowePrice, @digitalisvc and Alderline Group, and to @BaillieGifford, @PremjiInvest, @trueventures, @KinnevikAB, @FPVventures, @_DimensionCap, @lifeforcecap and @Lux_Capital for continuing to back us.
The best is yet to come…
Read more: businesswire.com/news/home/2…
#Biotech #AI #LearningfromLife
18
13
144
9,626
Ethan Perlstein 1-to-N retweeted
The value of speed for drug developers
nature.com/articles/d41573-0…
This new article estimates the value of a 4-month acceleration in market launch using data on the drugs approved by the FDA for which priority review vouchers have been used
1
26
117
11,168
Ethan Perlstein 1-to-N retweeted
I’m doing the same thing for an ultra rare pediatric neurodevelopmental disorder, AFG2B. Some of my biggest breakthroughs came from quickly cross-checking different computational methods like AF3, DynaMut, OpenMM, @boltz_bio etc. Research I’ve never done before but was stupidly easy with AI.
Recently, I did a big “sheperdizing” search of all relevant and adjacent research papers, reviewing references of references a few layers deep. This really helped narrow down to the most likely rescue pathways. A few months of reading and searching done in a few hours while I worked my day job.
Do you have access to @genbioai ‘s AIDO model? I expect exponential progress once this becomes available.
1
4
351
Ethan Perlstein 1-to-N retweeted
As a scientist and biotech founder, I can confidently declare that AGI is here and it's only going to get better.
I haven't quantified my productivity gains precisely and there are different metrics one can evaluate, but overall it feels like a 10-100X improvement.
I know that sounds insane so let me break it down.
I'm working on a 1-to-N case where the goal is to develop an allele-specific knockdown oligo (ASO or siRNA) for a progressive neurological rare pediatric genetic disease.
Unlike previous 1-to-N programs in years past, I can now quarterback the process myself using a tag team of Claude and GPT: one-stop shops for data analysis, hypothesis generation, and consultations on any topic, available at the time and place of my choosing. Critically, I also red team Claude and GPT against each other with identical prompts and file sets.
For example, take oligo design. Admittedly already a computationally native task, but one I would have handed off to an expert. What would have cost $12,000+ and taken a seasoned designer several days to a week of work took me only several natural language prompts and ~$100 and an hour of compute.
From $10,000's of spend to $100's of spend.
From cycle times of days to cycle times of minutes.
I'm not an expert in the disease gene, but I don't need to be. I'm not a trained bioinformatician, but I don't need to be. I'm not a nucleic acids chemist, but ... and so on.
AGI will disproportionately empower the generalists because we no longer have to spend a career or lifetime acquiring domain expertise.
So yes, AI will cures all diseases -- just not overnight. Doesn't feel that way today but we're just at the start of the exponential.
I'd love to hear from other scientists and biotech builders who are using the greatest tool ever created by humanity in their day-to-day battles against a specific disease.
Please share examples of how AGI is helping you become a 100X scientist!
(Props to @DeryaTR_ for leading the way)
48
37
305
20,137
Ethan Perlstein 1-to-N retweeted
Today, Pioneer Labs is announcing our first step towards terraforming Mars. 🚀🌼
With equipment that fits in just a single rocket launch, we can convert Martian dirt, water, and air into enough building materials to construct a small city on Mars.
To do it, we made the first microbe for Mars. We found the best microbe on Earth and used evolution to teach it how to source all of its nutrients directly from Martian materials.
The first astronauts will be greeted with safe shelter already filled with water, oxygen, and rocket fuel for the return journey.
This is the first step toward using biology to make Mars a friendly place for life. It lets us live off the land and helps us build the next great frontier.
It's the first of five organisms we need to green Mars ⬇️
554
1,431
11,349
2,446,033