The desire to improve people’s health and lives inspires us as inventors. Drug discovery is a time-consuming and laborious process. By conservative estimates, it now takes at least 10 to 15 years and $500 million to $2 billion to bring a single drug to market. Furthermore, there is a widening productivity gap: research and development spending continues to increase, yet the number of new therapeutic chemical and biological entities approved by the US FDA has been declining since the late 1990s.
Lack of efficacy and adverse side effects are two of the primary reasons a drug fails clinical trials, each accounting for around 30 percent of failures. Computational models and machine learning methods that can derive useful insights from large amounts of data on drugs and diseases from various sources hold great promise to provide information to help reduce these attrition rates and potentially improve the drug discovery process.
It is for these reasons that we began developing cognitive tools and platforms for drug discovery and development, which we envision could eventually benefit patients by accelerating the discovery of innovative treatments. In one of our projects, we invented an approach for predicting the associations between therapeutic indications (a valid reason to use a certain medication) and side effects (a secondary, typically undesirable, effect of a drug or medical treatment), and a visual analytics tool to help explore these associations. In other words, our invention could provide information to help drug discovery researchers identify which drug indications are typically linked to which side effects. This invention was recently granted a patent, US 9536194 entitled “Method and system for exploring the associations between drug side-effects and therapeutic indications,” and our work continues to gain positive momentum. The invention is among the capabilities applied to joint research with Teva Pharmaceuticals on drug repurposing.
One of the research areas of the Center for Computation Health at IBM T.J. Watson Research Center is translational informatics, which focuses on the development of novel techniques to extract insights and knowledge from biological and clinical data to support biological scientists, clinicians, and their patients. We have been working in this area for several years, and have developed a suite of advanced machine learning tools and computational models and platforms that can be used to derive insights from a wide variety of data sources ranging from pharmacological knowledge bases (e.g., PubChem) to real world data (e.g., electronic medical records) and are designed to help researchers improve the efficiency and effectiveness of drug discovery and development. Our methodologies have been applied to many specific use cases including drug repurposing (helping to find new uses for existing drugs in the market), indication expansion (helping to identify potential new indications for drugs still in various stages of the development pipeline), drug safety (helping to detect and predict a drug’s safety profiles) and personalized medicine (personalized efficacy/safety profile predictions).
IBM Research staff member Ping Zhang (left) and Program Director, Center for Computational Health Jianying Hu (right) with their newly patented invention to help drug discovery researchers identify which drug indications are typically linked to which side effects.
For the invention that was just granted a patent, we developed machine learning models to help predict therapeutic indications and side effects from various drug information sources. We also built an association engine to identify significant linkages between predicted therapeutic indications and side effects, and a visual analytics system to support the interactive exploration of these associations.
We believe this approach could help researchers in pharmaceutical companies to generate hypotheses for drug discovery. For instance, strongly correlated disease-side-effect pairs identified by our invention have the potential to benefit drug discovery in many ways. One can use the side-effect information to repurpose existing treatments (e.g., drugs causing postural hypotension could be potential candidates for treating hypertension). For example, If a new drug is being designed for a disease that is strongly correlated with severe side effects, then special attention should be paid to controlling the formulation and dosing of the drug in the clinical trials to help address the issue.
As an example, our invention identified that a side effect of weight loss is closely associated with the indication of mood disorders (e.g., bipolar disorder, depressive disorder, panic disorder). The identification of this association could lead to the hypothesis that it may be beneficial to investigate certain drugs for mood-disorder for potential repurposing towards weight control.
Being an inventor at IBM is inspiring, because we have the opportunity to help solve real-world problems. Our team is dedicated to this research and we continue to search for new ways to improve people’s health around the world through innovation and invention.
At the 18th European Conference on Computational Biology and the 27th Conference on Intelligent Systems for Molecular Biology, IBM will present significant, novel research that led to the implementation of three machine learning solutions aimed at accelerating and guiding cancer research.
Our team of IBM researchers published research in Radiology around a new AI model that can predict the development of malignant breast cancer in patients within the year, at rates comparable to human radiologists.