@Ronalfa profile picture
Ron Alfa @Ronalfa
Joined February 2011
  • Posts 3,813
  • Following 2,525
  • Followers 7,033
200 Photos and videos
Pinned Post
Ron Alfa
 @Ronalfa
Apr 18
1/ Spatial transcriptomics is among the richest view of human biology that we have: 18,963 genes mapped at subcellular resolution. It's also almost never collected outside of research settings. So we trained a foundation model to generate it from a clinical H&E image alone. Meet TARIO-2. 🧵 noetik.blog/p/tario-2-a-whol…

TARIO-2: A Whole-Transcriptome Foundation Model from H&E Alone

Spatial transcriptomics may be the richest measurement of a tumor we have. It's also almost never collected. Here's how we get around that.

noetik.blog
9
34
184
20,867
will take this w
Garry Tan
 @garrytan
17h
The founder in their 40s with taste and discernment is the new gentleman unicorn founder Because there can be 100x to 1000x of them working at their beck and call via agents and software factories all the time
15
2,799
at the sf party, beer pong is played for sport, very little actual beer is consumed
8
973
Ron Alfa reposted
Benjamin Perry @bots_and_bits
20h
OpenDDE - An Apache-2.0 licensed co-folding method that approaches IsoDDE in performance and beats Proteinix-v2 and ESMFold2🚀 Code - github.com/aurekaresearch/Op… Paper - github.com/aurekaresearch/Op…
3
20
98
6,026
👀
Diego del Alamo @DdelAlamo
20h
An open source reproduction of isomorphic’s IsoDDE w/ Apache license claims to beat Protenix-v2 on antibody antigen docking performance
1
22
4,354
Ron Alfa reposted
Alexis Ohanian 🗽
 @alexisohanian
Jul 4
Fuck Cancer
Ben Wilson
 @BenWilsonTweets
Jul 3
The Cancer Episode > My diagnosis and prognosis > What this means for How to Take Over the World > A few thoughts on death > See below for links to GoFundMe, etc.
16:16
13
11
260
31,201
Very practical about it: everything we do @NOETIK_ai is focused directly on using AI to tackle cancer.
MTS
 @MTSlive
Jul 2
SITUATION EXPLAINED: Why is the next wave of AI biology simulations not experimental perturbation data? We asked @Ronalfa, co-founder and CEO of @NOETIK_ai "Our view is the last wave of AI bio was, okay, we're gonna generate a whole bunch of data in the lab, experimental perturbation data, and then train models on that. This next wave is actually gonna be more simulations." "From the very beginning we've been focused on the concept of world models. Can we actually just simulate the perturbations? Because if you want to work with the data that's most relevant, which is the human data, you can't really perturb it." "Historically, people have tried to come up with virtual cell models as, okay, if we can just simulate everything, all the chemical reactions in the cell, think of the cell as just a bag of chemical reactions, and if we have a model that could simulate everything, you can fully understand the biology of the cell. My view is, that's gonna be really hard. I don't even think we could generate that level of data." "When we think of a virtual cell, we're literally just trying to simulate some biology of a cell in a patient tumor. Placing cells in that patient tumor in the simulation." "I don't really want to get to all the chemical reactions. I just want to know what genes are gonna be good drug targets for cancers, and can we understand patients that respond to certain drugs and patients that don't based on these simulations."
3:00
1
3
39
4,479
Interesting read
Carlos Outeiral
 @c_outeiral
Jul 2
Everyone's talking about AI-designed antibodies. Having sat through several such "revolutions," I was wondering, is this the real one, or just another flop? So I wrote a long-form essay on whether it holds up. Spoiler: the science checks out, not so sure about the business.
1
3
12
7,156
Ron Alfa reposted
MTS
 @MTSlive
Jul 2
SITUATION EXPLAINED: Why is the cancer drug bottleneck clinical success not drug discovery? We asked @Ronalfa, co-founder and CEO of @NOETIK_ai "Pharma companies have tons of programs that they're working on at any given time. They basically bring them into portfolio review and every year decide which targets they're gonna push forward. Generally there is an abundance of new targets and pharma companies are not moving forward all the targets that they could move forward into the clinic." "The bottleneck really is, one, clinical development is a bottleneck because it's very expensive. You can't afford to run every drug that you develop. But also it's a bottleneck because most things fail. Ninety percent of drugs fail in the clinic." "So if you increase the top of the funnel to push more things into the clinic and most things are still failing, it becomes really hard to impact probability of success by increasing the top of the funnel." "That's why our focus is, if we can make drugs more successful in the clinic, that's probably the best lever for getting new drugs approved." "If we can go from 90% of the drugs fail to 80% of the drugs fail, which is still insane that that many drugs fail, but that's where you're really gonna impact new medicines for patients."
2:54
4
5
57
35,846
Ron Alfa reposted
knower
 @knowerofmarkets
Jul 2
Awesome lineup on MTS today: @ahall_research just discussed AGI politics @Ronalfa on the AI x Bio explosion happening @evan_wineland & @kaandogrusoz discussing their home robot, Isaac 1 @jared_western telling us about Western Chemicals @kmad sharing his thoughts on FDEs versus consultants @rdn_nikita on Flexion's approach to robotics @jasonhausenloy to discuss the frontier lab brain drain @StevenGlinert about Nvidia & geopolitics timeline review and coverage of important stories like government equity stakes in OpenAI, the return of Fable, and other breaking stories at the frontier of the singularity
MTS
 @MTSlive
Jul 2
GOVT STAKE IN OAI | ANTHROPIC X SAMSUNG | NVDA REVENUE SHARING x.com/i/broadcasts/1dJrPPvMe…
11
6
46
8,586
Live in 10 on @MTSlive
2
318
🎯
11
2,578
Public data is the constraint: passing in virtual cell inference from hundreds of patients, and now we're cooking up some more interesting targets.
2
542
it wants to drug MDM2/TP53, what do you guys think?
4
21
5,034
Very good launch by the @AnthropicAI life sciences team today.
Claude
 @claudeai
Jun 30
Introducing Claude Science, a new app designed with every stage of research in mind. Artifacts traced to their code, environments managed on demand, and 60 optional scientific databases that you can connect. Available now in beta.
1:26
1
32
3,333
This is great. So many rare diseases that could use some impact. I think this can actually be scaled today.
Andrew Dunn @AndrewE_Dunn
Jun 30
NEW: Anthropic announces it is developing its own preclinical drug programs No specifics provided beyond a focus on neglected diseases "To build the right models, products and tools, we need to live it along with all of you," Eric Kauderer-Abrams, Anthropic life sci head, says
1
1
66
12,917
"It all comes down to the hard problem of is this the right target for this disease." @VasNarasimhan
1
11
1,875
Ron Alfa reposted
Matt Krisiloff
 @mattkrisiloff
Jun 30
I’m so excited to share this update on @Conception – We’ve generated the first early human eggs derived from stem cells. This is a big deal -- the potential to redefine fertility is real.
438
894
6,467
2,491,658
Bio x AI Cycling WhatsApp group: chat.whatsapp.com/LvyU8BVvOV… @AlfredoAndere @diegoarey @naywilliams @sahir_raoof @_odedf @A_Cembellin_Pri @antibody_42 @laurenap_ @brezina @cyrusnewday @ArdaGoreci @Sree_Surineni @joythw @g_palla1
2
11
457
Cool
Kenny Workman
 @kenbwork
Jun 29
Announcing the formation of Latch Biosecurity with our acquisition of TwentyTwo. Harmon, John and Evan are brilliant. Stoked to welcome them to the team. A few observations motivated this: - Most domains of basic and applied biology research have dual use potential. Not restricted to those obviously categorized in this way, e.g., surveillance. Each assay class and field of study requires focused thought and engineering to understand how harmful behaviors can emerge from otherwise productive research. - Miscalibrated refusals slow research progress - Practical AI x biology workflows require a new class of agent-native products for biosecurity Exciting roadmap ahead and lots of work to do. First product release this week.
11
3,495
Ron Alfa reposted
Sam Rodriques
 @SGRodriques
Jun 29
At Edison, we are on a mission to leverage AI to prevent or cure all diseases. Today, we are announcing a critical step: our first partnership using our AI scientist, Kosmos, to launch new biotech companies. Many of the most transformative new medicines begin life in biotech companies rather than big pharmas. We are partnering with Population Health Partners (PHP) to deploy Kosmos across their company creation pipeline. Kosmos will assist in the creation of PHP’s companies, allowing us to bring more medicines to patients. PHP has an outstanding track record of incubating successful biotech companies like @MetseraInc (acquired by @pfizer in a $10B deal), and The Medicines Company (acquired by @Novartis for $9.7B), to name just two. With PHP, we will focus on biotechs treating population-scale diseases like cardiovascular disease and obesity. If we can scale up the process of launching Metsera-style companies, we have an opportunity to transform medicine in the coming years.
1:57
6
35
168
30,404
Load more