Dr. CathyCab

engineer, scientist, mom, feminist, liberal

[Parmita Mishra @parmita]

the job is not to solve biology. that'd be ridiculous

biology is an open + non-linear thermodynamic system driven by degeneracy. Computing it would need infinite bottom-up quantum simulation.

i see AI x bio as using the biological matter as an analog computer to run the simulation in physical space and time.
its phenotype is merely a non-destructive interface to read its output. functional phenotypes inform you of where the system goes next. that's all.

there is no solving
there is only interfacing

you communicate with life, often productively. you do not control it perfectly. it is inherently chaotic. that's not a thing. solving biology is not a thing

[Michael Lin, MD PhD 🧬 @michaelzlin]

I think I've figured out one problem.

When biologists say complexity, they often mean variability: different results in similar conditions or with what seem like meaningless differences (different cell density, different culture medium, etc). I believe we/they do this because it sounds better, as if the variability were hinting at something profound rather than being meaningless noise or revealing the model system to be fragile.

When CS people hear complexity, they think great, we can solve complexity better than biologists because we can scale our model to more parameters to capture all the interactions.

Sorry to inform you that biology suffers not primarily from complexity of nodes and edges but from every single living model system being fragile and/or stochastic, from cell culture all the way to humans. The relative robustness of single protein folding or binding simply will not generalize to living cells. Even if you do it in 99 different cell types or condition, the 100th cell type or condition may not give you a predictable result.

[The Fort Pointer @FortPointer]

We're (3) weeks away from Tall Ships Parade of Sail on 7/11.

Can anyone explain why SailBoston's official map of Seaport viewing sites excludes Fan Pier Park, Pier 4 and the "Massport Viewing Area," and conflates $225-$375 per ticket bleachers on Fish Pier w/ free, public sites?

[𝒢𝒾𝓁𝒷ℯ𝓇𝓉 @AbakpaJob]

Having cats is basically babysitting drunk friends:
Someone's always hungry, someone's puking, and someone's yelling for no reason.

[Sammy James🔺 @HebHammer94]

I have a genius idea! @fenwaypark/@RedSox should open up the ballpark in a limited capacity on Friday to watch the Team USA game @ 3, followed by the Scotland game @ 6. Could even extend it for Brazil Vs. Haiti @ 8:30. Place would be nuts. Seems like a no brainer. Beer.

[Dr. CathyCab @CathyCathycab]

Certainly technically impressive, but also a really shiny hammer deeply looking for a nail.

[Eric Nguyen @exnx]

Together with my co-founders Michael @MichaelPoli6, Stefano @Massastrello and Armin @athmsx, I am excited to announce @RadicalNumerics is emerging from stealth with a $50M seed round to build general biological intelligence.

We’re also sharing an early preview of our new model Omnii, the most powerful genome language model to date.

Omnii preview link:
https://t.co/ouikMtRVwf

At Radical Numerics, our mission is to master the code of life, and to drive the frontier of biological AI for both design and defense.

This is our dual mandate, which comes from something our own team helped make possible.

Our founding team trained Evo and Evo 2, the largest biological AI models (40B params) trained on DNA sequences. Trillions of tokens across all of life, from microbes to mammals. It’s fully open source, and created the field now known as generative genomics.

Last year, scientists used Evo to generate the world’s first complete genome from scratch using AI. Turns out it was a bacteriophage—a type of virus. It functioned in the real world, and in this case it was harmless. But for us, it was a clear turning point.

It showed that AI is no longer just analyzing biology. It is on the cusp of generating functional lifeforms. Eventually, AI will have the power to design and control life itself.

That should make all of us incredibly excited, and incredibly uneasy. (Anyone can design DNA with a new function, and have it synthesized and delivered, like something from Amazon Prime).

The same technology that will help us cure cancer is the very technology that might create the next global pandemic, or worse, allow the creation of bioweapons that can wipe out populations.

We believe these forces are inseparable. If you work on the frontier of biology, you have to build technology to safeguard it from its misuse. Existing biosecurity tools are sorely losing the arms race, relying on outdated “have I seen this exact thing before?” style algorithms.

We founded Radical Numerics to turn the tide.

And we can’t do that by training on textbooks and natural language. We must understand the language of biology from the raw physical data itself, to reason across every molecule and modality, from DNA to proteins.

The next frontier for AI goes far beyond chatbots or video generators to models that can understand and engineer life.

Today, we’re previewing Omnii, which is already far surpassing Evo 2, and will continue improving as we scale and add new modalities (training now).

1. For human health, Omnii can read and write whole genomes (more on writing later). It’s state of the art (SOTA) on detecting causal variants for disease, and can rank Alzheimer's mutations zero-shot. We’re partnering with a diagnostics company to use Omnii for early cancer detection (pancreatic and multi-cancer).

2. For defense, Omnii is SOTA at detecting AI-generated pathogens. We benchmarked existing detection tools, and they simply can’t detect the AI-generated ones (“deepfake viruses”). We’re partnering with a US national lab to pilot Omnii for detecting the next pandemic, both natural and AI-generated.

We have a data center full of Blackwells in construction now to build the most powerful biological AI models ever. This mission takes a new kind of AI lab that can actually scale on physical, biological data: new alignment research (mid/post training), scaling long context, building out mech interp teams to dissect what these models learn, new architectures and systems designs, all from the ground up.

Our team is made up of AI researchers and scientists from top labs and institutions (e.g. Stanford, MIT, Google DeepMind), but more importantly, we all share the belief that this is the most important challenge of our lifetime. If you feel similarly, we are hiring. We aim to bring the brightest minds in AI and science together to save lives.

Thanks to our partners on this journey, led by Emergence Capital @emergencecap, with Obvious Ventures @obviousvc, Triatomic @TriatomicCap
, and Patrick Collison @patrickc. Our advisors include Eric Horvitz @erichorvitz, CSO of Microsoft, Chris Re @HazyResearch of Stanford, George Church @geochurch of Harvard, and Andrew Weber @AndyWeberNCB, former Assistant Secretary of Defense for Nuclear, Chemical and Biological Defense Programs.

Fortune article: https://t.co/L3f3f1329T

Jobs: https://t.co/EzsHSMcGJ1

[Parmita Mishra @parmita]

Bioinformatics asks: given this data, what patterns exist?

Computational biology asks: what data should exist so that the biological question survives digitization?

[Turnbull @cturnbull1968]

This is just absolutely humiliating.

[urbanist slop hq @SlopHq]

a bus every 6-7 minutes is the entire secret of good transit. no schedule, no app, you just show up. the kids have been screaming the answer at us for over a year. six seveeen. headways. they're talking about headways

[‏ً @omgsidewalks]

Honestly, i don't understand this economy when nursing homes are so expensive they bankrupt our grandparents but nursing home aides need to use food banks.

daycare is so expensive it eats up one parent's entire paycheck and yet daycare providers only make $10/hr and need second jobs.

college costs hundreds of thousands of dollars and puts students into debt for life and yet we have thousands of professors living in their cars.

everything we need is astronomically expensive and yet almost none of the money we pay is going towards the people actually doing the work and providing the services.

[Patrick Boyle — e/🦀 @p_maverick_b]

I did my PhD in Harvard’s systems biology department - seeing great mathematicians and physicists routinely getting owned by the challenges of doing biological experiments shaped my views on uplift. Many of our levers for biosafety are in atoms not bits

[Emily Zanotti 🦝 @emzanotti]

Honestly, all Platner shows is that progressives don’t really care about Donald Trump’s behavior, they’re just jealous he’s not on their side.

[Compassion in World Farming UK @ciwf]

"Is it possible to create an ethical octopus farm?"

Quite simply, no. There is no humane way to farm octopus. We need to leave these incredible, fascinating creatures alone.

#WorldOceansDay

[Eitan Fischberger @EFischberger]

This was the top of the New York Times last night. The first thing every reader sees the moment they land on the homepage of the most influential news outlet on earth. All three stories about Israel.

There is a famous communications scholar named Maxwell McCombs who developed what is now called the "agenda setting theory."

His core finding is simple: The press is not very good at telling people what to think. It is extraordinarily good, however, at telling people what to think ABOUT.

And what the New York Times has decided you should think about, every single day, multiple times a day, forever and always, is Israel.

You cannot saturate the most influential newsroom in the English language with relentless coverage of one small country and then act surprised when the public becomes similarly hyper-fixated with it.

The animosity we constantly see is the predictable output of editorial selection, repeated daily, until it becomes the background music of how people think about the Jewish state.

[Richard Hanania @RichardHanania]

Josh Barro on Platner:

“Graham Platner doesn’t work for a living. As The New York Times reports, the bulk of his income comes from a military disability pension of approximately $60,000 a year. The pension doesn’t mean he’s too disabled to work — he is, after all, currently seeking the job of U.S. Senator — but his recent non-campaign endeavors seem more like hobbies than a career. He runs an oyster farm that principally sells oysters to his mother’s restaurant. He earned a small stipend as his town’s harbor master: $3,000 last year. He lives in a $205,000 house that he bought with a $200,000 loan from his father.”

What a resume.

[CONSEQUENCE @consequence]

Ronny Chieng had one message for Harvard grads during his commencement speech: destroy AI.

"Look, a lot of other respected graduation speakers in colleges around America are talking about you guys needing to master AI for the future. I'm here to tell you the mission of your generation is to destroy AI...

"And I know, I know there's someone sitting out here right now who’s just like, 'Well, you know, what about the use of AI to pioneer breakthroughs in medicine and physics?' Well, first of all, shut up, nerd. I'm not talking about that. Obviously, if you're using it for that purpose, you're not the problem.

"I'm talking about the accumulation of cognitive debt due to excessive use of large language models according to a study by MIT published in 2025. That's right, MIT. MIT did that study. I guess you guys were too busy giving each other A's. Feel free to boo MIT, by the way, and AI, and yourselves, I guess.

"Look, this is actually good news, okay? This is why you guys shouldn't be scared of AI, because I think AI is just going to end up making mediocre people dumber. Have you heard how dumb people brag about how they use AI? They're always like, 'Hey, did you know that AI can now read my email, summarize it, and drop a response?' Yeah, you know who else can do that? Me. I can do that. You can't do that? How useless are you? You need artificial intelligence just to match me? I'm a dumb*ss who couldn't get into Harvard.

"From what I can see, getting an actual advantage from AI in the future will require a minimum escape velocity of intelligence that I'm assuming you guys from Harvard have. Everyone else who can't match that is just going to get dumber, and that's when you run up the score on them, assuming we still have a functioning society, of course.

"But to run up the score, you’re going to have to master your craft. And AI can be the fuel, but fuel is useless if you can't kindle the fire. For example, I recently used AI to use regression analysis to prove that a certain race of people are mathematically terrible at sports. I won't say which race, but thank you for not inviting Hasan Minhaj to Harvard. My point is, learning the fundamentals still matter. If I didn't know what a regression analysis was, and if I wasn't fundamentally racist, would I have been able to do any of that? No.

"Untalented people love bragging about using AI to help them draft their speeches and their scripts and their podcasts and their promo videos for UFC fights at the White House, which to be fair, even if they had filmed that for real, it would still have looked like AI. But what they're missing is this: the creating is the fun part. The best part of comedy writing is figuring out the puzzle pieces of a joke and getting the self-regard from having accomplished a difficult thing. Why would I want AI to take that away from me?

"You know what problem I want AI to solve? I want the problem of AI making everything look like sh*t. I want AI to solve that problem. How about that?

"Or how about, can AI take away the part of comedy writing where my TV pilot gets passed on and when I ask if I can pitch it to someone else, the network says, 'We don't want it, but we also don't want anyone else to have it. We just want you to be sad.' Can AI solve that?

"I recently tried to introduce my friend to Buddhism through a book called Buddhism Made Simple. It was literally a book about Buddhism made simple. And instead of reading it, he used AI to summarize it in 10 seconds. Believe it or not, he didn't reach enlightenment. It turns out speed running Buddhism is completely missing the point.

"And I know this platitude is almost worthy of AI, but the reason shortcuts to skip to the end aren't always good is because the journey isn't just how we acquire skills. The journey is the point of all this. It is! It turns out maybe the real Harvard was the friends we made along the way.

"Look, I know this won't apply to everyone's industry, but I'm just saying whatever your chosen profession is, please don't let AI rob you of the fun part of it.

"I think your generation's upcoming battle won't be humans against AI. That's at least two months away. It's going to be people with substance versus people with shallow knowledge. It’s going to be mastery versus faking it. It's going to be people with good taste versus tacky. I trust you will put in the work necessary to be on the right side of those battles."

[Mike Solana @micsolana]

look, they had a nice run, killed hundreds of millions of people, but it's time to gene drive mosquitoes out of existence. fuck it, ticks too.

[‎➳ ⋆.𖥔 𝘚𝘦𝘳 𝘌𝘭𝘭𝘪𝘦 @gildfae]

We need to get the Ents together and take down data centres like Isengard.

[Jon Levine @LevineJonathan]

Remember when the New York Times accused Jews of training dogs to rape Muslims and then never followed up with a single reported news article about it?

[Eylon Levy @EylonALevy]

Pretty sick of the countries that are boycotting Eurovision because the Jews are competing. Before you say “don’t conflate!”, these are countries that expelled their Jews (Spain), or refused to welcome Holocaust refugees (Ireland), or where most Jews were genocided (Netherlands). Europe liquidated its Jewish population in living memory and the survivors built new lives in Israel. Now these entitled, hateful broadcasters are boycotting the show in solidarity with the territorist armies that tried to kill those survivors. It’s so fucked up.