IBM announced advancements in its Watson cognitive computing capability, moving beyond analyzing data for known answers in record speeds to providing supporting evidence to back up those answers and prompt new questions.
The new capability helps researchers accelerate scientific breakthroughs by simplifying the complexities and revealing the connections in massive amounts of data, said Rob High, vice president and CTO of the IBM Watson Group, in an interview with eWEEK.
“We’re entering an extraordinary age of data-driven discovery,” said Mike Rhodin, senior vice president of IBM Watson Group, in a statement. “The proliferation of data is exasperating an organization’s ability to discover connections between disparate data. This announcement is a natural extension of Watson’s cognitive computing intelligence, empowering research, developers and industry experts, with powerful insights and connections in data, giving scientists the ability to make connections with data that others don’t see, which can lead to significant breakthrough discoveries.”
High said IBM Watson Discovery Advisor is a system that can visually reveal patterns and pinpoint connections in data to accelerate the discovery process. The cloud-based Watson Discovery Advisor is designed to scale and accelerate discovery by research teams, reducing the time needed to test hypotheses and formulate conclusions that can advance their work, from months to days and days to just hours, by bringing new levels of speed and precision to research and development, IBM said.
The upgraded Watson Discovery Advisor not only maps previously unknown correlations between key data points and understands nuances in natural language, but also understands the language of chemical compounds and how they interact, addressing critical dimensions in discovery in life sciences and other industries, IBM said. The system can be taught to learn and understand biology, intellectual property (IP), and law, among other industries, having profound implications for their R&D efforts.
Researchers and scientists from leading life sciences organizations, including academic, pharmaceutical and research centers, have begun deploying IBM’s new Watson Discovery Advisor to overcome the complex challenges of absorbing, analyzing and creating hypotheses from the millions of scientific papers available in public databases. According to the National Institutes for Health, a typical researcher reads about 23 scientific papers per month, which translates to nearly 300 per year. A new scientific research paper is published on average every 30 seconds, making it humanly impossible to keep up with the ever-growing body of scientific material available, IBM said.
The implications are astounding to advancing R&D across a variety of industries. For example, in 2013, the top 1,000 research and development companies spent $638 billion annually on research alone, according to Booz & Company. Meanwhile, it takes 10 to 15 years on average for a pharmaceutical treatment to go from initial research stage into practice, according to the Pharmaceutical Research and Manufacturers of America.
Leading life sciences organizations are deploying Watson Discovery Advisor to advance discoveries in on-going research projects, including Baylor College of Medicine, Johnson & Johnson and The New York Genome Center.
High said Baylor and IBM scientists demonstrated a possible new path for generating scientific questions that may be helpful in the longterm in developing new treatments for disease. In a matter of weeks, biologists and data scientists using the Baylor Knowledge Integration Toolkit (KnIT), based on Watson technology, accurately identified proteins that modify p53, an important protein related to many cancers, which can eventually lead to better efficacy of drugs and other treatments.
This is a feat that would have taken researchers years to accomplish without Watson’s cognitive capabilities, High said. Watson analyzed 70,000 scientific articles on p53 to predict proteins that turn on or off p53’s activity. This automated analysis led the Baylor cancer researchers to identify six potential proteins to target for new research. These results are notable, considering that over the last 30 years, scientists averaged one similar target protein discovery a year, IBM said.
IBM Watson Delves Even Deeper Into Data
“On average, a scientist might read between one and five research papers on a good day,” said Dr. Olivier Lichtarge, the principal investigator and professor of molecular and human genetics, biochemistry and molecular biology at Baylor College of Medicine, in a statement. “To put this in perspective with p53, there are over 70,000 papers published on this protein. Even if I’m reading five papers a day, it could take me nearly 38 years to completely understand all of the research already available today on this protein. Watson has demonstrated the potential to accelerate the rate and the quality of breakthrough discoveries.”
Meanwhile, Johnson & Johnson is collaborating with the IBM Watson Discovery Advisor team to teach the computer to read and understand scientific papers that detail clinical trial outcomes used to develop and evaluate medications and other treatments. The collaboration hopes to accelerate the conduct of comparative effectiveness studies of drugs. These studies help doctors match a drug with the right set of patients to maximize effectiveness and minimize side effects.
Typically comparative effectiveness studies are done manually, in which three people spend an average of 10 months–or 2.5 man-years–just to collect the data and prepare it for use before they are able to start analyzing, generating and validating hypothesis, IBM said. In this research study, the team hopes to teach Watson to quickly synthesize the information directly from the medical literature and allow researchers to start asking questions about the data immediately to determine the effectiveness of a treatment compared to other medications, as well as its side effects.
In addition, IBM and the New York Genome Center are working together on a clinical study to advance genomic medicine using Watson. The collaboration will initially focus on clinical applications of the cognitive computing system to help oncologists deliver DNA-based treatment for glioblastoma, an aggressive form of brain cancer that kills more than 13,000 Americans each year, IBM said. The clinical study is designed to evaluate Watson’s ability to help doctors cut through the data to identify personalized treatment options for gliobalstoma patients, based on their specific genetic mutations.
“We’ve been concentrating on the problem of how we enable people to answer the questions they’re asking. Now we’ve extended that capability into the discovery phase–to help people find out about the things they’re not asking but should be,” High told eWEEK.
Watson is now able to provide supporting evidence for the answers it provides to queries. “The accuracy of that supporting evidence is as important, if not more important, as the accuracy of the answer,” High said.
“How this applies to the cognitive computing world is we see this as not just about doing the work of humans, but enabling humans to work better,” High added. “It introduces what I think of as ‘inspiration’ into the equation. We can inspire people with new ideas that they by themselves might not have come to. We are helping people ask the questions they might not have thought of otherwise.”
Discovering something new is applicable to many domains such as medicine, law or finance that require deep insight into a large body of information and protocols. Cognitive computing will allow human experts to interact with large bodies of data and research and the knowledge and insight of many other experts in their field.
IBM Watson Discovery Advisor for life sciences is an integrated package of technologies delivered as a cloud service. The technology makes a map of information by reasoning over patterns it sees in available data, transforming raw data into new knowledge.
IBM Power Systems will support Watson’s data learning by providing faster access to big data. Watson Discovery Advisor has the potential to transform industries and professions that rely heavily on data, including law, pharmaceuticals, biotech, education, chemicals, metals, scientific research, engineering, and criminal investigations, IBM said.
Three years after its triumph on the television quiz show Jeopardy!, IBM has advanced Watson from a game-playing innovation into a commercial technology. Now delivered from the cloud, powering new consumer and enterprise apps, Watson is smarter, faster and smaller, with a 240 percent improvement in performance and a 75 percent reduction in physical size.