So much information - so little time

It is estimated that the average researcher reads about 23 scientific papers per month, which adds up to around 300 per year. At the same time, it is claimed that a new scientific research paper is published nearly every 30 seconds, which equals more than a million annually. This means that the average researcher is missing out on 999,700 papers each year.

To be fair, researchers really only want to read papers specific to their particular field of interest, so the figure above is highly exaggerated. However, say you are interested in the protein p53, which is implicated in tumour suppression - so far about 70,000 scientific papers have been published about the protein. Even reading five of these papers each day, seven days a week, it will take you 38 years to catch up on all the published papers about p53.

The sciences are producing huge amounts of data and it is humanly impossible to keep up with this ever-growing body of scientific material. How can researchers deal with this volume of information? It seems the answer may be “Elementary, my dear Watson.”

Elementary, my dear Watson

In truth, Sherlock Holmes and his mate Watson have absolutely nothing to do with IBM’s Watson. However, researchers can uncover new relationships and recognise unexpected patterns among data that have the potential to significantly improve and accelerate the discovery process in research and science using Watson’s computing power.

Named after IBM founder Thomas J Watson, IBM Watson uses natural language processing and analytics to process information akin to how people think. This represents a major shift in an organisation’s ability to quickly analyse, understand and respond to big data. Watson’s ability to answer complex questions posed in natural language with speed, accuracy and confidence is transforming decision-making across a variety of industries. 

Available now as a cloud service, IBM’s Watson Discovery Advisor is designed to scale and accelerate discoveries by research teams. It reduces the time needed to test hypotheses and formulate conclusions that can advance their work - from months to days and days to just hours - bringing new levels of speed and precision to research and development.

Building on Watson’s ability to understand nuances in natural language, Watson Discovery Advisor can understand the language of science, such as how chemical compounds interact, making it a powerful tool for researchers in life sciences and other industries.

Researchers and scientists from leading academic, pharmaceutical and other commercial research centres have begun deploying IBM’s new Watson Discovery Advisor to rapidly analyse and test hypotheses using data in millions of scientific papers available in public databases.

In 2013, the top 1000 research and development companies spent more than $600 billion annually on research alone. Progress can be slow, taking an average of 10 to 15 years for a promising pharmaceutical treatment to progress from the initial research stage into practice. Using Watson Discovery Advisor, researchers can uncover new relationships and recognise unexpected patterns among data that have the potential to significantly improve and accelerate the discovery process in research and science.

Drug targets

In a retrospective study Baylor College of Medicine and IBM scientists demonstrated a possible new path for generating scientific questions that may be helpful in the long-term development of new, effective 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, which can eventually lead to better efficacy of drugs and other treatments - a feat that would have taken researchers years to accomplish without Watson’s cognitive capabilities. Watson analysed 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 per year. 

Drug re-purposing

Multinational pharmaceutical company Sanofi is exploring how working with Watson can speed up the discovery of alternate indications for existing drugs (drug re-purposing). Watson is able to understand and extract key information by reading millions of pages of scientific literature and then visualise relationships between drugs and other potential diseases they could target while providing supporting evidence each step of the way.

Drug safety and toxicity is a major driver of the high failure rate in clinical development and trials. Sanofi is exploring how Watson’s ability to understand, extract and organise toxicological information can enable researchers to make better informed decisions with respect to candidate progression

Genomic medicine

Despite tremendous discoveries into the genetic drivers of diseases like cancer over the past decade, big data makes it difficult to translate DNA data into life-saving treatments. Based on results from the clinical study, IBM Watson could soon help scale up the availability of personalised treatment options.

IBM Watson will be supporting the analysis in New York Genome Center’s clinical study to advance genomic medicine. The clinical study will initially focus on clinical application of genomics to help oncologists deliver DNA-based treatment for glioblastoma, an aggressive form of brain cancer.

Clinical trials

Johnson & Johnson is collaborating with the IBM Watson Discovery Advisor team to teach Watson to read and understand scientific papers that detail clinical trial outcomes used to develop and evaluate medications and other treatments. This collaboration hopes to accelerate comparative effectiveness studies of drugs, which help doctors match a drug with the right set of patients to maximise effectiveness and minimise side effects.

Typically, comparative effectiveness studies are done manually, requiring three people to spend an average of 10 months (2.5 man-years) just to collect the data and prepare them for use before they are able to start analysing, generating and validating a hypothesis.

In this research study, the team hopes to teach Watson to quickly synthesise the information directly from the medical literature, allowing 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. 

Mayo Clinic and IBM have announced plans to pilot Watson to match patients more quickly with appropriate clinical trials. A proof-of-concept phase is currently underway, with the intent to introduce it into clinical use in early 2015.

“In an area like cancer - where time is of the essence - the speed and accuracy that Watson offers will allow us to develop an individualised treatment plan more efficiently so we can deliver exactly the care that the patient needs,” says Steven Alberts, MD, the chair of medical oncology at Mayo Clinic. 

Researchers hope the increased speed also will speed new discoveries.

Clinical trials provide patients with access to new and emerging treatments, yet enrolling participants in trials is one of the more difficult parts of clinical research. Currently it is done manually, with clinical coordinators sorting through patient records and conditions, trying to match them with the requirements of a given study protocol. At any given time, Mayo Clinic is conducting over 8000 human studies in addition to the 170,000 that are ongoing worldwide. Watson’s cognitive computing ability will help sift through available Mayo clinical trials and ensure that more patients are accurately and consistently matched with promising clinical trial options.

“With shorter times from initiation to completion of trials, our research teams will have the capacity for deeper, more complete investigations,” says Nicholas LaRusso, MD, a Mayo Clinic gastroenterologist and the project lead for the Mayo-IBM Watson collaboration. “Coupled with increased accuracy, we will be able to develop, refine and improve new and better techniques in medicine at a higher level.”

A version of Watson will be specially designed for Mayo Clinic. As it progresses in its tasks and matures through this collaboration, it will learn more about the clinical trials matching process, becoming even more efficient and likely more generalisable. Watson also may help locate patients for hard-to-fill trials, such as those involving rare diseases.

Many clinical trials at Mayo Clinic and elsewhere are not completed due to lack of sufficient enrolment. Enrolment in general could be increased by the Watson project. In spite of well-organised efforts, even at Mayo Clinic, just 5% of patients take part in studies. Nationally, the rate is even lower, at 3%. Mayo hopes to raise clinical trial involvement to include up to 10% of its patients. Researchers say higher participation also should improve the quality of research outcomes.

To ensure Watson has the required expertise to assist with clinical trial matching, Mayo experts are working with IBM to expand Watson’s corpus of knowledge to include all clinical trials at Mayo Clinic and in public databases, such as ClinicalTrials.gov. The new Watson system is being trained to analyse patient records and clinical trial criteria in order to determine appropriate matches for patients.

In the cloud

Last January, the IBM Watson Group introduced three new cloud-delivered services: 

• IBM Watson Discovery Advisor aims to revolutionise how industries such as pharmaceutical and publishing conduct research.
• IBM Watson Analytics allows users to explore big data insights through visual representations, without the need for advanced analytics training.
• IBM Watson Explorer is designed to help users across the enterprise uncover and share data-driven insights more easily, while helping organisations launch big data initiatives more quickly.

All three of these new Watson services are fuelled by IBM Watson Foundations, a comprehensive, integrated set of big data and analytics capabilities that enable clients to find and capitalise on actionable insights.