Cracking the Cancer Code with Watson Genomics

Share this post:

Since the time of Egyptians building pyramids, we’ve known about the impact cancer has on humankind. It is the second leading cause of death in the US1, based on data from 2010 through 2012, nearly 40 percent of Americans will develop it in their lifetime; one in four men and one in five women are at risk of dying from the disease2. US presidents have put tackling this disease on their political agendas; in 1971, President Richard Nixon declared a “war on cancer,” and President Barack Obama announced the Cancer Moonshot in January 2016, both programs designed to accelerate cancer research.

As we have learned more about the causes and progression of this disease, the needle on patient survival rates has slowly moved. One of the key milestones in understanding cancer happened in the last few decades, when it was firmly established as a disease of the genome (meaning it is caused by mutations in DNA). However, the rampant heterogeneity of the disease, (i.e. variation from patient to patient, tissue to tissue within the same patient, and even within the same tissue of a patient), complicates the understanding of the disease and its treatment, diagnosis and prevention. Since cancer incidence and progression is so ill-understood, it poses massive challenges to the interpretation of a patient’s molecular profile.

Although the complexities and differences within even individual patients makes understanding the disease more challenging, the genomic differences in individuals are also the very distinctions that can potentially be used for patient-specific treatment. These differences are what we hope to exploit and leverage with Watson for Genomics.

Our reasoning algorithms take into account the existing models (even those that are contradictory) of the ever-changing understanding of the biology of cancer. Working hand-in-hand with structured and unstructured data sources, these algorithms can provide information regarding cancer-driving genes and suggest molecularly targeted therapies for consideration by the treating physician on a patient by patient basis.

Watson for Genomics can help a physician facilitate personalized patient care. This concept is only gaining traction with doctors and researchers. In 2015 President Obama had announced another initiative called the Precision Medicine Initiative (PMI), fostered particularly by the genomic revolution. Recently, our research team published a paper about the intersection between PMI and Watson for Genomics, published on

Collaboration with Broad Institute on cancer drug resistance

While the survival rate of cancer patients has improved over the past few decades, many cancers eventually reoccur. One of the causes of this is that the cancer becomes drug resistant. The biology of cancer drug resistance is not fully understood in the scientific community. Yet the impact of these mutated cancer strains are staggering – every year in the US nearly 600,000 cancer deaths are attributed to drug resistance3 . Today, little is known about the cause of drug-resistant mutations. This is why, in partnership with the Broad Institute, we hope to apply Watson’s computations and machine learning approaches to study and understand thousands of drug resistant tumors. With over 10,000 patient samples as well as the results of laboratory studies, we plan to take a crack at this mystery through collection of molecular data of large cohorts of patients, carefully designed laboratory experiments on cell-lines, and the power of sophisticated algorithms.

Ultimately, our hope is that a stronger understanding of the molecular underpinnings of cancer treatment resistance will help provide researchers and clinicians with even more information to help confront this serious challenge. If Watson and the Broad Institute can together help uncover new knowledge about this important issue, we will be that much closer to this shared Moonshot.



1 CDC, “Statistics for Different Kinds of Cancer”, accessed at:

2 American Cancer Society, “Lifetime Risk of Developing or Dying From Cancer”, accessed at:

3 American Cancer Society, “Cancer Facts and Figures 2016”, accessed at

More Healthcare stories

Using SecDevOps to design and embed security and compliance into development workflows

IBM Research has initiated focused efforts called Code Risk Analyzer to bring security and compliance analytics to DevSecOps. Code Risk Analyzer is a new feature of IBM Cloud Continuous Delivery, a cloud service that helps provision toolchains, automate builds and tests, and control quality with analytics.

Continue reading

Impact of the SQL Relational Model 50 years later

Fifty years ago this month, IBM researcher and computing pioneer Edgar Frank Codd published the seminal paper “A Relational Model of Data for Large Shared Data Banks,” which became the foundation of Structured Query Language (SQL), a language originally built to manage structured data with relational properties. Today SQL is one of the world’s most […]

Continue reading

IBM Research AI at ICASSP 2020

The 45th International Conference on Acoustics, Speech, and Signal Processing is taking place virtually from May 4-8. IBM Research AI is pleased to support the conference as a bronze patron and to share our latest research results, described in nine papers that will be presented at the conference.

Continue reading