(lively synth music) >> Announcer: Live, from Santa Clara, California, it's The Cube, covering Open Networking Summit 2017. Brought to you by The Linux Foundation. >> Hey welcome back everybody, Jeff Frick here with The Cube. We are at Open Networking Summit, joined here in this segment by Scott Raynovich, my guest host for the next couple days, great to see you again Scott. >> Good to see you. >> And real excited to have a long-time Cube alumni, a many-time Cube alumni always up to some interesting and innovative thing. (Scott laughs) Sam Greenblat, he's now amongst other things the CTO of Nano Global, nano like very very small. Sam, great to see ya. >> Great to see you too Jim. >> So you said before we went offline, you thought you would retire, but there's just too many exciting things going on, and it drug you back into this crazy tech world. >> Just when you think you're out, they pull you back in. (all laugh) >> All right, so what is Nano Global, for people that aren't familiar with the company? >> Nano Global is a Amosil-Q, which is the compound, which is a nano compound that basically kills viruses, pathogens, funguses, and it does it by attaching itself at the nano level to these microbiol, microlife, and it implodes it, and technically that term is called lysis. >> (Jeff) That sounds very scary. >> It's very scary, because we try to sell it as a hand processing. >> You just told me it kills everything, I don't know if I want to put that on my hands, Sam. (all laugh) >> No it's good, that it kills some of the good bacteria, but it basically protects you for 24 hours. You don't have to reapply it, you can wash your hands. >> (Scott) It's like you become Superman or something. >> Absolutely, I literally use it to wash off the trays on the planes, and the armrests, while the guy next to me is sneezing like crazy, to try to kill any airborne pathogens. >> So what about the nanotechnology's got you traveling up to Santa Clara today for? >> Well, what I'm doing is, one of the things we're working on, besides that, is we're working on genomics, and I worked with some other companies on genomics besides Nano, and genomics has me totally fascinated. When I was at Dell, I went to ASU, and for the first time, I saw, pediatric genomics being processed quickly, and that was in a day. Today, a day is unheard of, it's terrible, you want to do it in less than an hour, and I was fascinated by how many people can be affected by the use of genomic medicine, and genomic pharmacology. And you see some of the ads on TV like Teva, that's genomic medicine, added tax, a genomic irregularity in your DNA, so it's amazing. And the other thing I'm very interested in is eradicating in my lifetime, which I don't know if it's going to happen, cancer, and how you do that is very simple. They found that chemotherapy is interesting, but not fascinating, it doesn't always work, but what they're finding is if they can find enough biometric information from genomes, from your proteomics, from your RNA, they can literally customize, it's called precision medicine, a specific medicine track for you, to actually fight the cancer successfully. >> I can't wait for the day, and hopefully it will be in your lifetime, when they look back at today's cancer treatments, and said "now what did you do again? (Sam laughs) You gave them as much poison as they could take, right up to the time they almost die, and hopefully the cancer dies first?" >> I'll take the - >> It's like bloodletting, it will not be that long from now that we look back at this time and say that was just archaic, which is good. >> It's called reactive medicine. It's funny, there's a story, that the guy who actually did the sequencing of the DNA, the original DNA strand tells, that when he was younger, he basically were able to see his chromosomes, and then he was able to get down to the DNA and to the proteins, and he could see that he had an irregularity that was known for basically cancer. And he went to the doctor, and he said "I think I have cancer of the pancreas." And the guy said "your blood tests don't show it." and by the way you don't get that blood test until you're over 40 years old, PS-1, the PS scan. And what happened was they actually found out that he had cancer of the pancreas, so... >> Yeah, it's predictive isn't it? So basically what you're doing is you're data mining the human and the human genome, and trying to do some sort of - >> We're not doing the 23andme, which tells you you have a propensity to be fat. >> Right, right, but walk us through what you're doing. You're obviously, you're here at an IT cloud conference so you're obviously using cloud technology to help accelerate the discovery of medicine, so walk us through how you're doing that. >> What happens is, when you get the swab, or the blood, and your DNA is then processed, it comes in and it gets cut to how many literal samples that they need. 23andme uses the 30x, that's 30 pieces. That's 80, by the way, gigabytes of data. If you were to take a 50x, is what you need for cancer, which is probably low, but it's, that takes you up to 150 gigabytes per person. Now think about the fact, you got to capture that, then you got to capture the RNA of the person, you got to capture his biometrics, and you got to capture his electronic medical record, and all the radiology that's done. And you got to bring it together, look at it, and determine what they should do. And the problem is the oncologic doctors today are scared to death of this, because they know how, if you have this, I'm going to take you in and basically do some radiation. I'm going to do chemotherapy on you and run the course. What's happening is, when you do all of this, you got to correlate all this data, it's probably the world's largest big data outside of Youtube. It's number two in number of bytes, and we haven't sequenced everybody on the planet. Everybody should get sequenced, it should be stored, and then when you get, that's called a germline you're healthy, then you take the cancer and you look at the germline and compare it, and then you're able to see what the difference is. Now open source has great technology to deal with this flood of data. LinkedIn, as you know open source, cacafa and one of the things that's great about that is it's a pull model, it's a producer, broker, subscriber model, and you can open up multiple channels, and by opening up multiple channels, since the subscribers are doing the pull instead of trying to send it all and overflow it, and we all know what it's like to overflow a pipe. It goes everywhere. But doing it through a cacafa model or a NiFi model, which is, by the way, donated by the NSA. We're not going to unmask who donated it but, (laughs) no, I'm only kidding, but the NSA donated it, and data flows now become absolutely critical, because as you get these segments of DNA, you got to send it all down, then what you got to do is do, and you're going to love this, a hidden Markovian chain, and put it all back together, so you can match the pattern, and then once you match the pattern, then you got to do quality control to see whether or not you screwed it up. And then, beyond that, you then have to do something called Smith-Waterman, which is a QC time, and then you can give it to somebody to figure out where the variant is. The whole key is all three of us share 99.9% of the same DNA. That one percent, tenth of a percent, is what is a variant. The variant is what causes all the diseases. We're all born with cancer. You have cancer in you, I have it, Jeff has it, and the only difference between a healthy person and a sick person is your killer cell went to sleep and doesn't attack the cancer. The only way to attack cancer is not chemotherapy, and I know every oncologic person who sees this is going to have a heart attack, it's basically let your immune system fight it. So what this tech does is it moves all that massive data into the variant. Once you get the variant, then you got to look at the RNA and see if there's variance there. Then you got to look at the radiology, the germline, and the biometric data, and once you get that, you can make a decision. I'll give you the guy who's my hero in this is the guy named Dr. Soon. He's the guy who came up with Abroxane. Abroxane is for pancreatic -- >> Jeff: Who is he with now? >> NantHealth. (both laugh) And why I, he discovered, he knew all about medicine, but he didn't know anything about technology. So then this becomes probably the best machine learning issue that you can have, because you have all this data, you're going to learn what it works on patients. And you're going to get all the records back, so what I'm going to talk about, because they wanted to talk about using SDM, using NFA, opening up hundreds of channels from source to, from provider to the subscriber, or consumer, as they call it, with the broker in the middle. And moving that data, then getting it over there, and doing the processing fast enough that it can be done while the patient still hasn't had any other problems. So I have great charts of what the genome looks like. I sent it to you. >> So it's clear these two fields are going to continue to merge, and the bioinformatics, and IT cloud. >> Sam: They're merging, as fast as possible. >> And we just plug our brain and our bodies into the health cloud, and it tells us what's up. >> Exactly, if Ginni was here, Ginni Rometty from IBM, she would tell you that quantum, she'd just announce it first commercially, an available quantum computer. Her first use for it is genomics, because genomics is a very repetitive process that is done in parallel. Remember you just cut this thing into 50 pieces, you put it back together, and now you're looking to see what's hidden, and it doesn't look like it's normal. If you looked at my genetics, one of the things you'll notice, that I will not consume a lot of caffeine. And how they know that is because there's a set of chromosomes, and my 23 chromosomes, that basically says I won't consume it. Turns out to be totally wrong, because of my behavior over the day. (all laugh) But what the Linux Foundation was interesting is everybody here wants to talk about, are we going to use this technology or that technology. What they want is an application, using the technology, and NantHealth that I talked about, can transport a terabyte of data virtually. In other words, it's not really doing it, but it's doing it through multiple sources and multiple consumers, and that's what people are fascinated by. >> All right, well like I said, Sammy gets into the wild and wooly ways and exciting new things. (Sam laughs) So sounds great, and a very bright future on the health care side. Thanks for stopping by. >> Thank you very much. I hope I didn't bore you with... (Jeff and Sam laugh) >> No, no, no, we don't want more chemotherapy, so that's definitely better to have less chemotherapy and more genetic fixing of sickness. So Sam, nice to see you again, thanks for stopping by. >> Thank you very much. >> Scott Raynovich, Jeff Frick, you're watching The Cube, from Open Networking Summit in Santa Clara, we'll be back after this short break. Thanks for watching. (synth music) >> Announcer: Robert Hershevech.