Hacker News Reader

Comment by hprotagonist | Hacker News Reader

personalized medicine of this form is one case where at least for now the cost makes obvious sense. You’re paying like 8 FTE PhDs to hand-build you a therapy from your own body. It’s not even close to automated.

but: if it doesn’t work, sometimes you get your money back! (novartis has never had to pay out)

https://www.fastcompany.com/40461214/how-novartis-is-defendi...

View on HN

Maybe there's a possible job in helping them automate this?

safe bet people are working real hard on it; it’s just like, cell cultures are fussy and biology doesn’t care about you, so it’s very much an uphill fight.

one dream job of mine would possibly be working on the data side of this though

That won't lower costs, only increase profits

That's bullshit, because it can do both. Increasing the amount that can be made by 10x could both reduce unit costs by 8.5x and increase unit profit by 1.5x. Get outta here with your pessimistic external-locus-of-control communism

Automation has a tendency to do both increase profits and lower costs. This is because as the price decreases the market increases. There's 500,000 new cases of pancreatic cancer a year in the world. How many of them do you think will have access to a One Million Dollar treatment?

This is not to mention how much of this tech could be shared with other treatments for other diseases.

That might be a thousand times more expensive. It’s not software, it’s biological processes that are hard to automate.

There are lots of folks working on problems surrounding automation and reproducibility to enable science in this space. One such effort that I contribute to is [pVACtools](https://pvactools.readthedocs.io/en/latest/), (there are others as well). We do have a small team of developers who work in concert with scientists and clinicians to build these tools.

How would I apply for a job like that? I’ve always had a strong interest in bio and computers (and physics, and psychology, and…)

I don’t know if 8 PhDs is entirely accurate, biotech teams are a thing and usually there are a good mix of non-PhD technicians who support an individual project.

I image it like ML teams in tech - the PhDs are building the models and the non-PhD are exposing the model, jockeying the data, etc.

You're right that there are a variety of folks working on these things, both in academia and industry, but it's more complicated than "PhDs do this, others do that". It comes down to knowledge, ability, and experience, and the best research groups grab onto that wherever they can and slot people into the roles that are right for them.

Seems more like politics where I'm at. My company won't let you work on the models without a PhD. Literally all I would do as an "ML Developer" would be cleaning and moving data plus APIs to provide input/output.

Reply on news.ycombinator.com