I have worked as an architect in HCLS technology for 10 years with Google and AWS.
Listening to you stitch all the technology and innovation together, moving from the information world to the physical world, as a patient, almost brought me to tears.
You said this will be the standard of care in 30-years, but the capability is here today. Listening to your recording is a profound moment in my life.
Really appreciate it. I agree the capability is here today. I'm trying to get it to more people so that the learning effects and economies of scale can bring the costs down.
I can attribute jumping several economic classes to the social skills I honed in high school and college. I have many friendships that are decades-plus and I had 150+ of my invited friends / family attend my wedding.
Emotionally, I do not long for new friends. It's a lot of work to maintain the relationships I have with my friends, family, wife and daughter.
I find aimless socialization these days to be laborious. I just do not give a shit.
I recently moved to NYC. I am at a point in my career where it's networking and politics that will get me ahead. I see a lot of my net-new socialization moving this direction.
I agree with you. This is brilliant. Tele-operation of the humanoid vs. waiting for AI is the key here. Then off-shore the tele-operation when you've smoothed out the edges.
I'll personally wait to own best hardware (Unitree) and purchase my own 3P tele-operation service contract.
But what I’ll say is, ideally they would demonstrate whether this model can perform any better than simple linear models for predicting gene expression interactions.
We’ve seen that some of the single cell “foundation” models aren’t actually the best at in silico perturbation modeling. Simple linear models can outperform them.
So this article makes me wonder: if we take this dataset they’ve acquired, and run very standard single cell RNA seq analyses (including pathway analyses), would this published association pop out?
My guess is that yes… it would. You’d just need the right scientist, right computational biologist, and right question.
However, I don’t say this to discredit the work in TFA. We are still in the early days of scSeq foundation models, and I am excited about their potential.
Cellular level computational simulation existed a very long time and it's more impressive by the day because of large collections of experimental datasets available.
However to infer or predict celular acitivities you need a ton of domain knowledge and experties about particular cell types, biological processes and specific environments. Typically the successful ones are human curated and validated (e.g large interaction networks based on literature).
In cancer it's even more unpredictable because of the lack of good (experimental) models, in-vivo or in-vitro, representing what actually happens the clinically and biologically underneath. Given the single cell resolution, its uncertainty will also amplify because of how heterogeneous inter- and intra- tumours are.
Having said that, a foundation model is definitely the future for futher development. But with all of these things, the bigger the model, the harder the validation process.
Thank you for spending the time to write this comment. As a solution architect, I don't write production-level code. I did not know of anything beyond unit / functional testing in verification automation.
reply