Biotech visionary. Looking for best drugs with goose data. Retweets mean honking endorsement.
Joined April 2022
- Tweets4.1K
- Following221
- Followers6.6K
- Likes869
Mr. @DaveARicks , @EliLillyandCo seems to have the momentum of a runaway freight train. Why are you so successful?
We've trained a Claude agent on a million @benchling ELN entries with the goal of making a model to optimize our experimental design process.
Our learning is that people haven't been doing their ELNing.
Goose Data (wa/wa) retweeted
people arguing developability is some magic sauce these companies have should look at the literature. you have a series of criteria which are hard coded and very trivial (no cysteines in CDRs, reduce possible glycan sites...etc) to internalize and you have a few softer metrics which are open source on the back end.
twist, genscript, thermo.....these guys don't have any functional readouts downstream so they have *zero* inference to the utility of the expressed molecules and very limited data on the runs that they do put in their workflows for things like aggregation or thermal stability
there is really genuinely nothing special here. i know many pharmas and CROs that have internalized high throughput antibody expression workflows.
TWST and companies like them are selling to a specific crowd (cash rich AI x Bio people) who have shown no ability to do this basic thing (make molecules and test them in vitro) and these companies will eventually run out of money propping up this revenue. likely without any drugs to show for it.
that said, this means the trade is short term bullish but long term bearish. there is no $100B+ or $1T TAM for this work. drug discovery is not being 'revolutionized'. it is being pushed forward incrementally by very smart people with S-tier judgement in translational science and a real ability to execute. not new people who can't even internalize in vitro work with hundreds of millions in cash
A remarkable overestimation of CRO revenue on the back of a slight improvement to in silico pipelines. Genuinely wreckless and very ignorant of market realities.
Yes the AI trade has tailwinds right now but a lot of this post is pure hyperbole and if you're actually putting money on these companies you're better served hedging than putting your foot on the gas of this trade
The bottleneck is and always has been the lack of core team meetings to discuss the pre-read slide deck.
Starting a working group to figure out a path forward.
The biggest effect of AI will be making drug discovery...
12%Cheaper
31%Faster
15%Better Drugs
42%nothing will change
599 votes • Final resultsReally is the bottle neck of the healthcare industry!
Imagine how much more successful Purdue would have been with such a thing!
An AI agent that can write actual prescriptions!
This team is building the future of healthcare and it's inspiring to watch. There is such a massive opportunity to improve healthcare with AI and it requires bold work like this. A huge congratulations to the Nolla team! Well deserved.
I'm biased, we got to work with Nolla at @TrainLoop_ai post training models for use cases like this. There might even be more models on the way, who knows.
One failure of AI is that the ads I see on this site are still trash.
No, I don't have chronic lymphocytic leukemia, I don't need a BTK inhibitor.
Thanks you very much 😤
Just because you hit statistical significance once, doesn't mean you'll hit it again. 😔
At least management is positive!!!
There's a lot of lingo here.
But when scientist and clinicians say something is "unimpressive", what they mean is that it's bad.
When they do say "bad", it probably means it's literally poison.
Peptide therapies are very exciting!
This is the team that runs our medicine factory in Research Triangle Park, Lilly’s first new U.S. site in 40 years when announced. It makes the leading medicine for obesity and diabetes for millions of US patients.
Building it was harder than we thought: 3.5 years from announcement to first pack, six years to peak capacity. Pharma patent clocks and build clocks don't line up unless you design a platform - one that runs follow-on products and that the third, fourth, and fifth plants can reuse. We did.
Thank you, North Carolina state and local leaders, and a workforce that delivered. You set the template.
We're now building/built in Concord, Kenosha, Huntsville, Houston, Lebanon ( a town in Indiana!), Richmond, and the Lehigh Valley.
What still slows delivery: permit timelines and associated NIMBYisms, grid connections that aren't ready when the steel is, and big labor gaps in the building trades. The fixes are straightforward. Moving colleges and state, local, and federal agencies all at once is not.
Building is hard. Doing it in America is worth it.
#WeAreLilly
Developability is the most important thing in antibody discovery.
Big pharma leads in this space which is why they lead in new antibody discoveries!
AI will allow startups to compete!
GPT-6 Astra is now the best-performing frontier model for antibody developability prediction in our benchmark. 🧬
Finding an antibody that binds is only the beginning. Will it aggregate? Will it stay stable? Can it actually behave like a drug?
GPT-6 Astra predicts these developability properties better than the other frontier models we tested. 🤯
And yes, the interactive visualization showing how an antibody actually works was also built with Astra in about 1 hour.
This model looks seriously powerful.
Congrats @gbt @sama
#insilicoSOTAFM
$30-35/hr
Hiring 1-2 experienced wet lab biologists in SF for a several-month contract (potentially longer). Experience with RL environments is preferred. In-person, five days a week, in Dogpatch area.
Interested —> DM with background and send a phone number. Optionally include some times of availability and ideally details of RL env work. I will give you a call if there’s potentially a good fit.
I always tell the team, you go into candidate selection with the drug you have not the drug you might want or wish to have at a later time.
