Demonstrating end-to-end scientific discovery with Robin: a multi-agent system

futurehouse.org

18 points·eamag··4 comments

>Discovery of ripasudil as a novel dAMD treatment: Using the data from the first round of drug candidate testing, Robin proposed a second set of drug candidates. We tested them in the same experiment to find there was a new top hit: ripasudil, a drug already used in the eye.

https://bmcophthalmol.biomedcentral.com/articles/10.1186/s12...

(2020)

>Ripasudil alleviated the inflammation of RPE cells

>The inflammation of the RPE cells could also lead to the occurrence and development of age-related macular degeneration (AMD)

I assume you posted this as a counter, but it really isn't. It doesn't distinguish between wet and dry amd (the claim is very specifically for dAMD) but more importantly something that can help prevent a problem is very different from something that can treat a problem.

A seatbelt helps prevent you flying through a windscreen, but EMTs don't rush to put a seatbelt on your mangled body.

This happened on twitter I feel as well, where someone did a cursory search and posted that this was all solved and everyone patted each other on the back but the result wasn't a counter.

From my last comment on this, this is in the paper.

> Notably, while ROCK inhibitors have been previously suggested for treatment of wet AMD and other retinal diseases of neovascularization, Robin is the first to propose their application in dry AMD for their effect on phagocytosis

Is this the first entirely automated scientific research/ discovery?

"no" but there is clearly an arms race going on

Reply on news.ycombinator.com