Hacker News Reader

Comment by throwawaylolx | Hacker News Reader

I don't know why the author is so critical of his peers. DeepMind didn't come up with a novel biological insight; they simply pointed their unparalleled AI resources towards a deep learning problem. Is it really surprising that a team of world class deep learning scientists with virtually infinite resources managed to outperform pharmaceutical companies at a deep learning problem? I don't think so.

View on HN

> DeepMind didn't come up with a novel biological insight; they simply pointed their unparalleled AI resources towards a deep learning problem.

What this essentially comes down to is a bunch of teams with high domain expertise and high technical capability was beaten by a team with higher comparative technical capability and lower comparative domain expertise. One has to wonder if one of the teams with higher domain expertise (ie work in the field every day) could achieve better results by improving their technical capabilities, or more aggressively applying their domain knowledge.

The question really comes down to whether DeepMind can beat any hard "game" (chess, go, protein folding etc) better than humans with deep domain expertise.

> What this essentially comes down to is a bunch of teams with high domain expertise and high technical capability was beaten by a team with higher comparative technical capability and lower comparative domain expertise.

The fact that the experts are entrenched in departmental trench wars and crippled by bureaucratic crap on every step is probably a contributing factor. If they were free to advance actual research instead of contributing to some middle manager's manager's agenda, thinks might look different.

This doesn't affect chess and go however ...

Regarding domain expertise, I would guess that this is something you can catch up with within a year.

That's what people outside drug development think. Meanwhile, 70 % of all drugs fail at Stage 2 testing. That's right, experienced pharmacologists at established companies go forward with compounds they believe will succeed, and still in 2 out of 3 cases they are not better than placebo.

I'd hasten to add it's the same in software. We think the experts are right on the edge of performance, accuracy, and success while using all the right tools - meanwhile from the inside, it's a shambles in all but the highest stakes environments with the most qualified teams.

I think the parent is talking about expertise for protein structure prediction, not drug development

It's not like pharmaceutical companies are strapped for cash.

If they don't invest in R+D it's a choice.

Why should they invest in r&d when someone like alphabet will do it for them?

I believe they do spend a lot of money on R&D. Someone came along with a new idea and applied it to their field.

Now that Alphabet has sown them a better way, money should pour into the new technique.

While big pharma does do internal r&d, de facto, new drugs and drug strategies largely come by letting the government fund researchers (typically professors at large research institutes), encouraging these professors to do a spin-off (I deliberately don't want to use "startup") to absorb the scientific risk, followed by an acquisition for IP.

Sure, but I'd be surprised to see a pharmaceutical company profitably competing for talent with FAANG in the race of hiring top ML scientists.

While they might virtually have infinite resources, I would be quite interested in how much this actually cost. Can anyone share estimates?

I think the key to the success was translating folding into deep learning problem and solving it (maybe there is still something to win by optimizing the models). The academia and industry were not treating folding as deep learning problem and they were using 'old' methods with slight improvements from year to year. The usage of deep learning was a breakthrough and will speed up research in this field.

That's most definitely not the case: https://scholar.google.com/scholar?q=deep+learning+protein+f...

The Go community was rather forthcoming to what AlphaGo brought, because it provided new insights to a field that some of them study quasi-religiously.

Let's see how pharmaceutical companies are dealing with that, or the next fields DeepMind will enter and shake up like that.

Eh... no? If you look at abstracts, Zhang group (the second place) also used "deep-learning based contact predictor". Really, the usage of deep learning was kind of standard long before CASP13.

If you read OP, AlphaFold's main innovation is predicting distance instead of contact (regression instead of binary classification), which was also independently developed by Xu, and further, using probability distribution over distance instead of simply choosing the best distance.

Yes, you wonder how Alphafold compares with the combined wisdom of Foldit. You start out with a secondary structure prediction from Zhang lab and ask a panel of experienced humans to fold up the protein into a plausible tertiary structure. As a human you have a good idea what larger pieces fit where, but AI still has problems with the large picture, cf. the leopard sofa problem.

Reply on news.ycombinator.com