The Translational Data Accelerator provides a suite of applications that support researchers with varying levels of technical expertise and through the different stages of the translational research lifecycle.

Whether you are evaluating study feasibility, exploring molecular datasets, building cohorts, conducting advanced analytics, or developing machine learning models, our platforms provide secure access to trusted research data and computational resources.

Atlas

Atlas is the primary self-service application for exploring de-identified clinical data represented in the OMOP Common Data Model. Atlas is built on the OHDSI open-source ecosystem and is widely used by academic medical centers and healthcare organizations around the world.

Atlas enables investigators to:

  • Evaluate study and clinical trial feasibility
  • Explore patient populations
  • Identify potential cohorts
  • Understand disease prevalence and characteristics
  • Define reproducible cohort definitions
  • Support protocol and grant development

The application provides an intuitive web-based interface that allows researchers to answer many common feasibility questions without requiring programming or database expertise. Cohorts developed in Atlas can be used to support requests for research datasets (or biospecimens via TakePART NW) through established institutional governance processes.

cBioPortal

cBioPortal is an interactive platform for exploring molecular and clinical cancer data. Originally developed at Memorial Sloan Kettering Cancer Center and now used by research institutions worldwide, cBioPortal provides powerful visualization and exploration tools for oncology research without requiring researchers to write code.

Researchers can use cBioPortal to:

  • Visualize mutations, gene expression, copy-number alterations, and other molecular features
  • Compare molecular data with clinical characteristics such as stage, age at diagnosis, treatment, and survival
  • Generate publication-ready visualizations, including OncoPrints and Kaplan-Meier survival curves

Explore examples of how to use cBioPortal here.

The Translational Data Accelerator supports a Fred Hutch-hosted cBioPortal environment, bringing cBioPortal’s exploration and visualization capabilities into a secure, institutionally supported research environment connected to Fred Hutch data resources and governance processes.

The Fred Hutch environment enables researchers to:

  • Explore investigator-generated datasets in a shared, supported environment without maintaining a separate cBioPortal installation
  • Controlled access to datasets so they are available only to appropriately authorized users
  • Share approved datasets with collaborators through established governance processes
  • Explore curated Fred Hutch molecular oncology data available through our curated molecular oncology dataset

Bringing your own study to cBioPortal

Researchers with molecular and clinical study data can request to make their dataset available through the Fred Hutch cBioPortal environment.

Study access and data use are managed through established governance and authorization processes. Once a study is approved and its data are prepared for cBioPortal, the dataset can be hosted in the Fred Hutch environment and made available to authorized members of the research team for interactive exploration. Learn how to bring your study to the Fred Hutch environment (access on Campus or via VPN).

You can find the latest release notes related to the Fred Hutch hosted cBioPortal on our Announcements page

Databricks

Databricks is the primary computational environment supporting advanced analytics, machine learning, and AI-enabled research within the Translational Data Accelerator ecosystem.

The platform provides scalable computing resources and integrated tools for working with structured, unstructured, and multimodal data. Databricks combines data engineering, analytics, machine learning, and generative AI capabilities within a unified environment.

Researchers use Databricks to:

  • Access approved research datasets
  • Perform statistical analyses
  • Develop machine learning models
  • Conduct large-scale computational studies
  • Build reproducible analytical workflows
  • Develop and evaluate AI and LLM-enabled research applications
  • Process complex multimodal datasets

Databricks supports a wide range of users, from biostatisticians and bioinformaticians to data scientists and software engineers. Researchers can work in SQL, Python, and R within a secure, governed environment.