Suggestions
Dylan Reid
Mentor at Techstars
Just a human in the loop @Zetta
Highlights
Google DeepMind just dropped #AlphaFold3 [1] and it looks like it's going to be a BIG deal for AI-driven drug design...
BUT as Demis Hassabis said in his interview with Bloomberg [2] - the real impact of these technologies will be measured in the clinic and we're at least "a couple of years away."
It's a good reminder that the true test of AI for drug development will not be how well models generalize or perform on benchmarks, but how interdisciplinary teams are able leverage them to develop drugs that are safer, more effective and more accessible to patients. The DeepMind team knows this which is why they founded Isomorphic Labs to develop a therapeutic pipeline with Eli Lilly and Company and Novartis.
But while we wait to see how those programs progress, we're going to have plenty of exciting results to look out for... 👀
Just last week - the results of Recursion's successful Phase I trial for REC-994 were published in the Journal of the British Pharmacological Society [3], making it one of only a small number of AI-designed drugs to be tested in humans and advance to PhII [4]. While it may not be as exciting as a major new model drop, these are the kinds of milestones we should be celebrating if we want to measure AI progress in patient lives (instead of model parameters :)
Big congrats to Chris Gibson, Ron Alfa, Shafique Virani, Tim Considine JD, Tony Donato, Joel Ellis and everyone involved in this work and of course major props to the Isomorphic Labs and Google DeepMind teams who are investing in both technology and clinical translation. Can't wait to see how this changes what we can do for patients.
[1] https://lnkd.in/gW4UCddC
[2] https://lnkd.in/gfWsgu4n
[3] https://lnkd.in/gqKrMC_C
[4] https://lnkd.in/gDXkwDyA
cc Andrew Dunn, Andrii Buvailo, Patrick Malone, Elliot Hershberg, Leonard Wossnig, Jacob Rinaldi
2023 was WHIRLWIND year for AI x Bio and feels like 2024 is going to be even WILDER. Some thoughts (+ a few predictions) on what we'll be talking about this year...
-
”Bio Bucks" - We'll see an uptick in pharma AI partnerships but more scrutiny over deal details. Investors will discount commercial milestones more heavily and value quality over quantity in partnered programs.
-
"Public-Private Chasm" - The pace of AI breakthroughs in the lab compared to the glacial speed of trials will further the chasm between early and late stage companies and how public and private investors perceive (and value) AI-native biotechs.
-
"Insilico Assays" - As AI achieves experimental accuracy on a more measurements, diagnostic and tool companies will take AI more seriously with a string of acquisitions + investments.
-
"Multi-Modal" - Large, multi-modal models (LMMs) will steal the spotlight from language + diffusion models - opening opportunities in complex biology and non-generative applications like target discovery.
-
"AI-Native SaaS" - Open source models will continue to dominate bio and chemistry, but scientists will struggle to integrate them into their work. Software companies that build delightful UX on open source AI will own the workflows and capture much of the value.
-
"Interpretability" - As foundation models get bigger and more powerful, the ability to learn from their internal representations will become increasingly important for drug discovery. Expect interpretability to become a hot area of AI research, though likely under a different name.
-
"RLXF" - Reinforcement Learning through Experimental Feedback (RLXF) and other forms of supervision will prove powerful in aligning biological foundation models and will drive the biggest gains in utility and adoption this year.
-
"High-Context Data" - The move to self-supervised learning will change how companies run experiments and upend what training data is valuable. The specifics will vary by application, but expect an emphasis on data with higher dimensionality and more relevant “biological context”.
-
"Service Enabled Tech" - More AI companies will go to market as full-stack services, adopting service-based business models and competing against existing CROs. AI services will be able to charge more than software and fit more easily into existing budgets - but scope and operational efficiency will be important.
-
"Platform-Pipeline Fit" - AI biotechs will be forced to think about drug development sooner and face more scrutiny over the programs they pursue. The best companies will be incorporate clinical development into their platforms from the beginning and prioritize programs that are scientifically and commercially attractive while being maximally aligned with their platforms.
What's on your bingo card for 2024?...Ron, Elliot, Patrick, Leonard, Brandon, Amee Kapadia, Dr. Shelby, Pablo, Bogdan, Luis, Drausin, Andrew, Brian, Chris, Rohan, Rob, Joshua, Dave?
