IBM Watson fires its own cancer-fighting ‘moonshot’ | Hacker News Reader

IBM Watson fires its own cancer-fighting ‘moonshot’

venturebeat.com

54 pointsprateekj38 comments

Instead of Jeopardy, I always wanted Watson to go on "are you smarter than a fifth grader" and throw questions like "think of the second color of the rainbow. What large bird starts with the same first letter?"

Or: "If I call A 5, B 4, and C 3, then what do I call D?"

As a scientist (working on cancer, no less), more than just churning through terabytes of data, it strikes me that THESE sorts of questions are the ones you'll need to be able to answer. There's also a lot of "reading between the lines" you'll need to do, and a lot of "small pattern extrapolation". For example, the drug candidate I'm working on is one of a family of about let's say, 16 known compounds. All of them kill cancer cells, fourteen are cardiotoxic. The two that are not, lack a single oxygen atom. Maybe that oxygen atom lets the molecule block the mitochondrion (sort of important for heart cells)? Why? Because the molecule family looks sort of like Coenzyme Q10 - if you use your imagination and remember some of your chemical reactivity rules from Grad-level Ochem - and supplementation of Q10 alleviated the cardiotoxicity in one study on rats for one of the molecules.

Ultimately, my previous boss was bold enough to latch onto that and give it a shot - can we take this molecule that is the most potent of the group (but unfortunately cardiotoxic), delete the homologous oxygen atom and make something that still works and is less cardiotoxic?

>Ultimately, my previous boss was bold enough to latch onto that and give it a shot - can we take this molecule that is the most potent of the group (but unfortunately cardiotoxic), delete the homologous oxygen atom and make something that still works and is less cardiotoxic?

And? Did it work?

ELI5 - how do you delete the oxygen atom? How is the new compound replicable at a scale where you have lots of it?

THanks - I am fascinated by this - but am completely ignorant. Thanks

It worked well enough that I'm starting a nonprofit (http://indysci.org) in part to continue the work (by "previous boss", I mean, she quit her job as a professor - so the whole thing is the pharma equivalent of abandonware). It was 5 times less cardiotoxic by design, and not by design, it was 10 times more potent.

You delete the oxygen atom by doing something called mutasynthesis, if you want a better description email me, I'm literally going to the bank right now to open the account for the nonprofit:

http://www.ncbi.nlm.nih.gov/pubmed/?term=yonemoto+gerratana (click on "free PMC article")

It's scaleable since the hacked bacteria do all the heavy lifting.

I'm legitimately stumped. Which large bird starts with 'O'? Or is it second from the top.. so an Ibis?

Ostrich?

I'm going to go sit in the corner and cry now.

Thanks!

Only one I thought of relates to baseball. Didn't think it was large.

Edit: for some reason ramchip I keep missing that one when the subject of birds comes up

Large is relative. I thought Owl, Osprey. Ostrich (another comment) is of course larger. Guess your answer depends on your training set.

As far as I'm aware there have been zero peer reviewed publications showing Watson actually doing something in medicine. When a doctor thinks he or she has a better approach than what is being done, they use the scientific method to find out of their hypothesis holds up. So far I've seen lots of IBM conference presentations and press conferences. But where is the data? Where are the increased survival rates? Where are the reduced medical errors compared to controls? That is science. Everything else is snake oil unless you have that.

Indeed. It's a little annoying to keep seeing these IBM PR releases about fictional medical suppositions, when one is working on drug screening, diagnostics, and such from within a reality constrained lab!

The way i understand it, this system offers data to physicians ,it doesn't offer treatment, so it doesn't have to be clinically tested.

It is the the responsibility of the hospital and doctors to evaluate and see if it offer any benefit to them.That's the usual standard information systems are bound to.

But watson is quite new. I'm pretty sure testing will come with time.

This is truly incredible stuff. Hats off to IBM for pushing the field of machine learning forward.

This statement really struck me: "“In the Jeopardy days [in 2011], Watson was running off 90 servers and could store 15 terabytes of memory,” said Gold. That’s not your typical household integration. “Now it’s far more affordable and runs 240 times faster, so we can do far more with less.” Moreover, the size of IBM’s server is now smaller than a pizza box and can fit in any data center."

How is this possible?

"Smaller than a pizza box and can fit in any data center" sounds like marketing for "We put it in a standard rackmount form factor"

If we ignore cloud interfacing solutions[1], there is a simple answer. "Enterprise scale" solutions really mean "Enterprise SCALING".

Pre-sales for 90 servers in many cases would be impossible. Pre-sales of a single rack mountable item that can then be expanded from depending on the workload is really attractive, however. Write the software right, and it should be able to cope.

Pre-sales probably starts with a customer renting a single box to try out, possibly with some professional services time. The customer decides they like it, and commission a project to develop a Watson solution. IBM pre-sales then work out how many servers would actually be needed, once the customer is interested. As the solution can scale, when management goes "that cost looks little high" they can safely say "well, if we reduce it by sixteen servers, you'll be slightly slower but that might be ok. Then if the business decides to increase we can simply add more servers[2] at a later date". And, of course, adding more servers at a later date means more professional services time, meaning more money and more quality customer relationship building time.

[1] Which for all kinds of commercial reasons, we have to discount a cloud solution being the default here. Buzzword, yes, but enterprise is slow and many really exciting uses cannot be based in the cloud.

[2] Or, y'know, just turn this key....

I'm assuming they mean 1 server is a pizza box (ie. it's a 1U rack server), so you'd still need ~90U of space.

Either that or IBM has made some breakthroughs in the last 2 years.

You can almost hear the article author completely misunderstanding what he was told.

I'm betting it was something like, "For Jeopardy, Watson was using 15 Terabytes of data stored on 90 servers. And those servers were large back then, now they're your standard pizza box 1U's. And because of advances in storage technology we can store a lot more in the same space."

Isn't there some sort of basic tech literacy class for people who write these articles? It's like reading the sports section and seeing "The Seattle Seahawks are a group of big men and they ran at another group of big men, and they've been doing that a lot, and faster lately. Some of the men in the Seahawks group are smaller and faster, so they are harder to catch by the other group of big men. The Seahawks grouping of men is like a successful business, because they've been winning competitions against other groups."

There is some more prize gibberish in that article:

Saxena said IBM is ready to make a big a bet on Watson, as it did in the 1970s when it invested in the emergence of the mainframe.

Leaving aside the grammatical error (it's probably just an editing artefact), what bed did IBM make in the 1970s on the emergence of the mainframe? Is that a typo for 1950s? Or just unvarnished ignorance of computing history?

It's stuff like this that makes me mistrust a news source: if they get it blazingly wrong about something I know enough about to have a working bullshit detector for, what are the chances that they're not also getting it wrong about everything else?

To be fair, the System/360 (which was the first to run the classic OS/360 operating system) was released in '64[1], and was replaced in the '70s by the S/370.

Certainly mainframes were in use prior to the '70s, but the S/360 was the first "classic" mainframe in that it was designed for both business & scientific use[2], and it wasn't until the 70's that they began to become more popular in areas outside their traditional use.

[1] http://en.wikipedia.org/wiki/IBM_System/360

[2]

16 Terabytes of RAM on P750s.

From http://en.wikipedia.org/wiki/Watson_(computer)

"Watson is made up of a cluster of ninety IBM Power 750 servers (plus additional I/O, network and cluster controller nodes in 10 racks) with a total of 2880 POWER7 processor cores and 16 Terabytes of RAM. Each Power 750 server uses a 3.5 GHz POWER7 eight core processor, with four threads per core. "

The main change from the Jeopardy competitions is they don't need to find an answer in under a second. So you can probably reduce to computitional capability significantly.

I would bet on optimization of the software, lower costs of memory, and database sanitization.

I think prior to that it mentions its cloud based, so the head end is probably only 1U.

A stab in the dark would be chips made from Graphene or IBM's Silicene, if I am to understand what you quoted.

Even if all some AI system did was take a list of symptoms, anonymize them, suggest a diagnosis, and then take the MD's actual diagnosis, track outcomes, and compare results over time, it would be a huge advancement.

Right now, unless I misunderstand something structural about our medical system, most of the data gathered by MDs are wasted, because there is no digital record of much of it and patient privacy legalities prevent wholesale entering of symptoms and case outcomes in to any sort of freely available (inter)national database.

Why is it always cancer?

I'd be impressed if somebody was trying to cure ALS or Tourette's syndrome.

Maybe because cancer is ridiculously more common than both of those disorders combined?

Perhaps cancer in general, but not particularly types of cancer, when you recognize that there are multiple forms of lung cancer, breast cancer, etc. -- common forms of cancer are not a single disease.

The trouble with cancer drugs is that you take them for a short time and you either die or you survive.

The real money is in drugs that a large segment of the population can take for an indefinite time such as Lipitor, blood pressure meds, asthma inhalers, and the perpetually elusive obesity cure.

Sure, but obviously curing cancer is a waste of research effort... what they should be trying to cure is disease.

Why is it a waste of research effort? We've made dramatic non-palliative strides in the area of cancer treatment, and most importantly have been able to induce complete remission in a significant number of cases.

see my sibling comment.

Sorry, I don't understand. If you meant PaulHoule's comment I'm not sure how that answers my question, and I didn't find any other comments by you. Do you mean because cancer isn't a simple, isolated thing like polio or TB?

Yes, I meant PaulHoule's comment. If the reason for studying "cancer" is that "cancer" is more widespread than better-defined illnesses, then the reasoning is misplaced; it's surely better to focus on something more widespread, like "illness".

I have seen it expressed, though not in these terms, that asking for a "cure for cancer" is equivalent to asking for bug-free software. Cancer isn't something that afflicts you from without; it's just a failure mode for cells. They can fail for all kinds of reasons (indeed, note that one of our best methods of causing cancer, radiation, works on the principle of "make random changes to the cell's genetic code, and it will eventually become cancerous").

So I think the general claim is that focusing on cancer is a middle ground between seeking treatments for obscure well-defined illnesses and searching for a panacea for illness or disease in general. Cancer has a precise operational definition: unregulated cell growth. It isn't nearly as vague as illness or disease. We understand a lot about all the different pathways inside cells that can lead to it. A "cure for cancer" is an attempt to address the issue upstream, such that it will affect all different types of cancer in the body. Unlike seeking a panacea for all illness, this is not obviously a waste of research effort. There are highly similar mechanisms at work in every cancer. Even if a cure-all cannot be found (I would be surprised, personally), just trying to understand the mechanisms still has value.

Actually, a cyst meets your definition of unregulated cell growth. It's just slower / less greedy. The distinction between a benign tumor and a malignant one is generally ascertained by the process of "let's watch it and see what it does", so I'm not convinced that "cancer" is defined more usefully than "illness".

In my opinion, and my medically-educated mother's opinion (really, I get it from her), people don't look for "a cure for cancer" because they think they can make progress from that viewpoint. They do it because they're looking for funding.

edit:

You might compare "died of cancer" to "died of old age". Old age used to be an accepted cause of death. When autopsies started happening, it was quickly noticed that people who had died of "old age" always had some other, more immediate cause of death. But it turns out that if you take an old person and do your utmost to prevent / cure / treat all of those more proximate causes, eventually one will get past you and they will die.

Okay, but that's not actually how cysts work if you look it up. They are sacs that can have cancerous tissue inside them, but they can also have fluid or air. The operational definition of benign and malignant tumors that doctors use boils down to rudimentary observation, yes, but in the lab we've really identified many of the key mutations in the transition from benign to malignant. So I guess you could say we have a reasonably good idea of what cancer is and how it works, but it hasn't fully made its way onto the front lines yet, either in terms of assessment or treatment.

I understand your cynical point of view about funding, but I guess I would say two things. First, cynicism is a cancer unto itself (ha), and all academics are faced with the corrupting influence of money. Second, if you go and talk to the cancer researchers in the nearest university you'll probably find a great deal of them really do care about the work; they just might care about a very tiny corner of it rather than a cure-all solution.

In this case they choose cancer , because

1.It's a large field, plenty of money.

2.Genetic/molecular cancer medicine is advancing rapidly, creating tons of data, which means This field will most likely be highly computerized anyway.

3.Machine learning is the core of genetic cancer medicine , and a lot of data could be really helpful.

4.Assuming this is the best approach to to do cancer medicine ,watson will have strong political allies on his side , since cancer frightens everyone , including powerfull people.

> Why is it always cancer?

I guess your mother and your grandfather and your first cousin didn't die of cancer. Very difficult to find that situation even combining ALS, MLS, Parkinson's, and Tourette's syndrome.

Maybe some day Watson can cure the most deadly disease of all -- idiocy.

Where can one learn more about how Watson actually works?

...starting from the IBM website and Wikipedia one mostly gets to whitepapers and high level descriptions. I hope at least bits of it are open source or at least have the math and logic underlying them well documented.

Reply on news.ycombinator.com