The newly launched Watson for Drug Discovery is a cloud-based offering that aims to help life-sciences researchers discover new drug targets and alternative drug indications. IBM’s natural language processing enables the Watson Cognitive Platform to read and understand the relevant literature from the corpus of nearly 31 million sources and detect connections and patterns that humans may not see. The average researcher reads between 200 and 300 articles in a given year, while Watson for Drug Discovery has already ingested 25 million Medline abstracts, more than one million full-text medical journal articles, four million patents and is regularly updated.

 

Watson for Drug Discovery will be augmented with Pfizer’s private data and help researchers look across disparate data sets to surface relationships and reveal hidden patterns through dynamic visualizations. Predictive models will then be applied to evaluate the available content to rank research targets and help generate and prioritize hypotheses. Finally, data visualization will help turn ideas into action by mapping these detected connections between entities, which can then be filtered for easier learning and application.

 

“We believe that the next great medical innovations will emerge as researchers and scientists find new patterns in existing bodies of knowledge. In order to do this, they need access to R&D tools that can help them efficiently navigate the opportunities and challenges presented by the explosion of data globally,” said Lauren O’Donnell, vice president of life sciences for IBM Watson Health. “IBM is honored to collaborate with Pfizer, and put Watson for Drug Discovery to work to support efforts in bringing life-saving immunotherapies to doctors and patients worldwide.”

 

Cancer is one of the leading causes of death worldwide, and presents an almost maddening complexity to researchers seeking more effective treatments and cures. One of today’s most promising treatment areas is immunotherapy, which works with a patient’s immune system to recognize and target cancer cells using a combination of vaccines, immunomodulators and small/large molecules.

 

Oncology researchers at Pfizer will use Watson for Drug Discovery to analyze massive volumes of disparate data sources, including licensed and publicly available data as well as Pfizer’s proprietary data. With this new tool, Pfizer researchers will analyze and test hypotheses to generate evidence-based insights for real-time interaction. The customized technology can also support efficient safety assessments.

 

“Pfizer remains committed to staying at the forefront of immuno-oncology research,” said Mikael Dolsten, president of Pfizer Worldwide Research & Development. “With the incredible volume of data and literature available in this complex field, we believe that tapping into advanced technologies can help our scientific experts more rapidly identify novel combinations of immune-modulating agents. We are hopeful that by leveraging Watson’s cognitive capabilities in our drug discovery efforts, we will be able to bring promising new immuno-oncology therapeutics to patients more quickly.”

 

Watson is a question-answering computer system named after IBM’s first CEO, industrialist Thomas J. Watson. It was originally developed to answer questions on the ABC show Jeopardy!, competing in 2011 against former champions Brad Rutter and Ken Jennings. Watson easily won the first-place prize of $1 million.

 

As far back as 2012, IBM teamed up with Memorial Sloan-Kettering Cancer Center to develop a Watson-based system that could help doctors create individualized cancer treatment recommendations for their patients. In the following years, IBM joined forces with WellPoint, the Cleveland Clinic, the New York Genome Center and a dozen other leading cancer institutes to analyze DNA and create products and tools to improve cancer treatment options. IBM is now also teaming up with the American Heart Association, the American Cancer Society and the American Diabetes Association.