Harnessing Big Data and ML to Drive Biotechnology’s Future
Big data in biotechnology: Even though Silicon Valley might be a fascinating place for big data, its most innovative applications frequently take place elsewhere.
In the world of venture capital, machine learning, and artificial intelligence are hot topics. The past several years have witnessed some spectacular exits, like Google’s $500 million acquisition of Deepmind in 2014, Twitter’s purchase of TellApart in 2015 for $533 million, and Intel’s $400 million acquisition of Nervana in 2016. But each of these was an IT play.
What happens when biology and machine learning collide?
Lygos, based in Berkeley, engineers and creates microorganisms that transform inexpensive sugar into valuable specialty compounds.
Fundamentally, they are creating and using a variety of hardware and software tools, then using them to study biology.
In the end, it is becoming faster and less expensive than ever before to program or design microorganisms. Modern data science and biotech advancements, as well as the rapidly declining cost of reading, writing, and modifying DNA (a trend that is even faster than Moore’s Law for computing), are driving this.
In other words, a new industrial revolution could very well be sparked by the most recent developments in software, big data, machine learning, biotech, and chemistry.
An innovative approach to machine learning:
Jeff Hammerbacher, a co-founder of Cloudera, famously remarked that “the brightest minds of my generation are thinking about how to make people click Ads. That is awful.
He was correct about the “suckiness,” but it’s possible that some of the brightest brains are working on something with a much greater overall impact.
Consider Eric Steen, co-founder, and CEO of Lygos, a fascinating firm that aspires to overtake DuPont but not in the conventional sense. As its main product, malonic acid (produced from petroleum) is used in a wide range of industries, including flavor and fragrance, electronic manufacturing, and coatings, Lygos creates microbes that transform sugar into high-value specialty chemicals.
However, what’s intriguing about Lygos is how he arrived at this position
Over millions of years, microbes have evolved into extremely productive factories. Because of how much information is encoded in the genetic code, microbes have extraordinary computational and machine-learning capabilities. Nature’s machine learning algorithm is evolution.
With the help of Lygos, it is now possible to direct and control the evolution of microbes so that they create certain products.
Every time a bacterium divides and grows, which happens every 20 minutes, it has the ability to perform a calculation. Millions of them are being processed by Lygos in a single vat at once.
They are creating and implementing a variety of these technologies to create microbial factories because they have access to a more potent machine-learning platform in nature than a computer could ever provide.
Data science outside of Silicon Valley: This is significant, awesome information that provides another justification for searching outside of Silicon Valley for the most intriguing ideas. Although Lygos is based in the Valley, its effects are not.
The use of big data by John Deere transformed farming, and the company still holds the view that the most significant data research does not take place within 45 miles of San Francisco.
Manufacturing and retail sectors, for example, have strong incentives to adopt data science and a wealth of data to work with, as stated by McKinsey & Co. The largest influence of big data won’t be on ad-clicking methods, even though they will use technology from Silicon Valley heavyweights (like TensorFlow from Google) and cloud providers like AWS will decrease the barrier for developers interested in dabbling in machine learning. Thankfully.
