Whether you are in the data science field, interested in developing computational solutions for clinical oncology, or a clinical researcher, we’ve curated a list of data sets, tools, and learning resources to showcase how these disciplines can and are working together to empower cancer research. Cancer data science can drive clinical oncology forward. For example, researchers develop computer models to serve as a cancer patient’s digital twin to capture real-time dynamics to create predictive models and, one day, guide treatment decisions. Also, national sequencing efforts could equip clinicians with the knowledge of how a patient’s genome impacts their response to medications.

Explore Clinical and Biological Data Online

With so much data available, we’ve pulled together a list of links to data sets for common cancer types across some of our NCI Cancer Research Data Commons (CRDC) resources. Collectively, these resources offer access to more than one million files of experimental and clinical data from landmark NCI studies and other grantee projects. Some of the clinical attributes include:

• demographics,
• diagnoses,
• treatments, and
• environmental exposure.

You can explore high-level trends in these data sets directly from the online portals or analyze the data with the library of robust computational tools provided through NCI’s Cloud Resources.

*Note: To find the links for each cancer site in the Genomic Data Commons, visit the exploration page to build a cohort.

Find Bioinformatics Tools for Predictive Oncology

In addition to what the CRDC offers, there are many analytical tools and pipelines to help you mine and extract meaningful insights from clinical data. We’ve added a few selections from our partners focused on supporting predictive and precision oncology analysis, but you can find more tools through the “Resources for Researchers” search engine.

• Accelerating Therapeutics for Opportunities in Medicine (ATOM) Consortium: This public-private consortium has developed the ATOM Modeling PipeLine (AMPL), an open source, free-to-use software for building and sharing models that advance in silico drug discovery.
• NCI-Department of Energy (DOE) Collaboration: NCI and DOE developed predictive artificial intelligence (AI) and machine learning (ML) models of drug responses in pre-clinical models of cancer to improve and expedite the selection and development of new targeted therapies. Its tools are available online through their capabilities catalog.
• Informatics Technology for Cancer Research (ITCR): This trans-NCI program supports investigator-initiated and research-driven informatics tool development. The ITCR program offers several resources for clinical research, including tools for integrating and analyzing electronic medical records, databases cataloging clinically actionable information for personalized cancer therapy, and training courses to advance your skills.
• Childhood Cancer Data Initiative (CCDI) Molecular Targets Platform (MTP): This knowledge base allows you to browse and identify associations among molecular targets, diseases, and drugs. Watch a recording on how to navigate, identify, and prioritize data in the CCDI MTP.

Learn About Cancer Data Science in Precision Oncology

If you’re new to the world of cancer data science and its application to clinical research, check out these blogs, news articles, and event recordings that cover some of the basics and examples of innovative work made possible through the intersection of these disciplines.

Training

What makes a good AI model?

Read five tips on how to use AI in your research or clinical practice.

Cancer Data Science Lifecycle

Apply our six-stage cancer data science process to your research. You’ll learn about everything from data cleaning (including our blogs that outline the importance of semantics and common data elements) to predictive modeling.

101 Course

Take our free, easy-to-follow courses on critical skills for each of the data science life cycle. You’ll get beginner-level introductions to topics like cloud computing and popular data science technologies.

Training Guide Library

Explore instructional guides and resources that elaborate on each of the cancer data science lifecycle stages.

Data Science in Clinical Care

How can you apply data science to your clinical profession? Here are some examples:

Cancer Detection

Automated AI Model Aids in Early Detection of Pancreatic Cancer

Use AI to find undetectable cancers on scans of a normal-looking pancreas long before clinical symptoms are visible.

Seeking a Better Biopsy? NCI-Funded Researchers Are Using Machine Learning to Identify Exosome Biomarkers

Read how NCI-funded researchers used ML to characterize a cancer biomarker based on exosomes. Their biomarker worked well using non-invasive sources, such as blood and urine, allowing the researchers to catch cancer early, even in tumors of undetermined origins.

Cancer Diagnosis

Biomedical Data Fusion Lab
Watch the recording and learn about leveraging data at different scales for personalized diagnosis, prognosis, and therapy in oncology and neuroscience.

NCI-Funded Researchers Develop a New Model for Interpreting Pathology
Classify cancer and predict its progress through a new NCI-funded approach. This approach helps annotate and analyze whole-slide images to contribute to your research. You can get the pre-trained model and see demonstrations, too.

Cancer Treatment

NCI Uses AI to Take the Guesswork Out of Assessing Prostate Cancer Images

See how our AI model offers another tool to help predict if cancer will spread or re-occur, giving us vital information for monitoring disease and tailoring treatment to each.

Affordable, Interpretable, and Equitable AI for Precision Oncology

Get advice by watching a video on utilizing AI to forecast patient response to treatment and track their progress in a clinical setting.

Machine Learning in Cancer Care Delivery: Moving from Model Validation to Clinical Workflow

Watch a video to learn how to transition from creating and validating ML tools and incorporate them into patient care.

Theranostics and AI—The Next Advance in Cancer Precision Medicine

Learn how AI and data are helping researchers “see” cancer in a new way, resulting in a more precise way of targeting cancer treatment. By using molecular imaging, we can identify areas of tumors most likely to respond to treatment and select the most effective therapies in those instances.