IBM Gives Cancer-Killing Drug AI Project To the Open Source Community 42
IBM has released three artificial intelligence (AI) projects tailored to take on the challenge of curing cancer to the open-source community. ZDNet reports: The first project, dubbed PaccMann -- not to be confused with the popular Pac-Man computer game -- is described as the "Prediction of anticancer compound sensitivity with Multi-modal attention-based neural networks." IBM is working on the PaccMann algorithm to automatically analyze chemical compounds and predict which are the most likely to fight cancer strains, which could potentially streamline this process. The ML algorithm exploits data on gene expression as well as the molecular structures of chemical compounds. IBM says that by identifying potential anti-cancer compounds earlier, this can cut the costs associated with drug development.
The second project is called "Interaction Network infErence from vectoR representATions of words," otherwise known as INtERAcT. This tool is a particularly interesting one given its automatic extraction of data from valuable scientific papers related to our understanding of cancer. INtERAcT aims to make the academic side of research less of a burden by automatically extracting information from these papers. At the moment, the tool is being tested on extracting data related to protein-protein interactions -- an area of study which has been marked as a potential cause of the disruption of biological processes in diseases including cancer.
The third and final project is "pathway-induced multiple kernel learning," or PIMKL. This algorithm utilizes datasets describing what we currently know when it comes to molecular interactions in order to predict the progression of cancer and potential relapses in patients. PIMKL uses what is known as multiple kernel learning to identify molecular pathways crucial for categorizing patients, giving healthcare professionals an opportunity to individualize and tailor treatment plans.
The second project is called "Interaction Network infErence from vectoR representATions of words," otherwise known as INtERAcT. This tool is a particularly interesting one given its automatic extraction of data from valuable scientific papers related to our understanding of cancer. INtERAcT aims to make the academic side of research less of a burden by automatically extracting information from these papers. At the moment, the tool is being tested on extracting data related to protein-protein interactions -- an area of study which has been marked as a potential cause of the disruption of biological processes in diseases including cancer.
The third and final project is "pathway-induced multiple kernel learning," or PIMKL. This algorithm utilizes datasets describing what we currently know when it comes to molecular interactions in order to predict the progression of cancer and potential relapses in patients. PIMKL uses what is known as multiple kernel learning to identify molecular pathways crucial for categorizing patients, giving healthcare professionals an opportunity to individualize and tailor treatment plans.
Thank you, IBM (Score:2, Insightful)
This isn't facegooggletwitterfruitbook claiming to be good because they contribute to open source projects of value to their own businesses. (Or pretend to like Go).
This is a company donating a bunch of real work that cost them real money to create to help cure cancer. Donations which have zero bottom line business value to IBM.
Maybe the big tech valley bastards can learn something about true philanthropy from IBM. I doubt it but the example is now there.
Re: (Score:1)
This is the shit IBM sold to governments, hospitals and universities for big money. But as the results were useless and health sector started to stop paying for it, the IBM is trying to get free PR by releasing code instead of moving it to /dev/null.
Re: (Score:2)
Re: (Score:1)
I wonder if the software can be extended to predict other useful compounds, like stuff that generates gravitons or gets you really high.
Wrong title. (Score:4, Informative)
IBM Discards Cancer-Killing Drug AI Project To the Open Source Community
I'm not stupid. IBM isn't giving this away out of the goodness of their heart, they simply no longer see how to use it to make a profit. They are "giving it away" because if it's good PR. If not for PR, IBM execs would prefer to simply archive it so that a competitor couldn't possibly profit from it.
Re:Wrong title. (Score:5, Interesting)
...IBM execs would prefer to simply archive it so that a competitor couldn't possibly profit from it.
Which probably means that IBM is convinced the whole approach is useless.
Re: (Score:2, Interesting)
Nah, it means they did enough to write a lot of vague patents in the 'AI prediction of cancer', so that when someone actually does the work and produces a working product, IBM is there to take the profit from them in the form of patent licenses.
Patents don't need to work these days, they only need to have legalese claims to working.
Re:Wrong title. (Score:4, Insightful)
IBM Discards Cancer-Killing Drug AI Project To the Open Source Community
I'm not stupid. IBM isn't giving this away out of the goodness of their heart, they simply no longer see how to use it to make a profit. They are "giving it away" because if it's good PR. If not for PR, IBM execs would prefer to simply archive it so that a competitor couldn't possibly profit from it.
Which probably means that IBM is convinced the whole approach is useless.
IBM began as the Computing-Tabulating-Recording Company (CTR) in 1911, and changed its name to IBM in 1924. To be sure, they acquired an unsavory reputation for hardball business tactics. But they have also had a reputation as a good employer, and more recently have shown a more enlightened attitude to open technology, with its early embrace of open-source. They have been leaders in the development of new computing technologies.
And they have entered and exited various businesses during their existence. Cancer AI now is one of them. I would hardly think that what they are releasing is "useless." More like they cannot continue to support it anymore, and like many other business who have faced this situation, they have elected to release it to the OS community so that it can continue to grow. If that gives them good PR, then more power to them.
Re: Wrong title. (Score:1)
And maybe cure some cancer. Execs get cancer too.
Re: (Score:2)
Redhat is an IBM band ... in case anybody has been asleep for a while.
Re: (Score:1)
Mozilla (Score:2)
Didn't Mozilla start the same way? Netscape gave away their source code because IE was killing them.
Re: (Score:2)
I'm ok with IBM doing the right thing for PR reasons. More than ok that that, actually. More companies should do the right thing for PR reasons. Looking at you Google. Looking at you Apple. Looking you Microsoft.
Not even going to mention Facebook, that cesspool is beyond hope. Rotten from the top.
It only runs on Watson (Score:2)
lol (Score:2)
Re: (Score:2)
Re: (Score:3)
I'm not a dumbfuck. I'm a Holistic Doctor (HD)
Sorry, those are contradictory statements.
that has understood the cause of diseases that no Western doctor ever has: Down's Syndrome, epilepsy, schizophrenia, cancer, etc.
Oh god.
Yes, I'm sure cancer is caused by our Qi being out of alignment or some such.
You are the cancer of our species, dude.
Re: (Score:2)
Sorry, those are contradictory statements.
Not at all. These days it's not uncommon to treat patents for both the primary condition and the resulting symptoms, such as depression or limited mobility.
This guy sounds like a quack but the basic idea of holistic treatment is well established and widely practised in modern medicine.
Re: (Score:2)
He doesn't mean that. He's a "Holistic Doctor", which means he's a naturopath. Holistic just sells better.
Re: (Score:2)
cancer is your path to enlightenment
You may be mistaking it for corprus.
John Kanzius (Score:1)
"Not to be confused with Pac-Man"??? (Score:2)
Uh, right.
I read that mouthful. There's no way that acronym was not deliberate.