The challenge of genomic data

In the era of precision medicine, the bottleneck is no longer generating data, but making sense of it. Large-scale projects like The Cancer Genome Atlas (TCGA) have produced petabytes of data, but for a clinical researcher, querying this information without advanced coding skills used to be nearly impossible.

What is cBioPortal?

The cBioPortal for Cancer Genomics is an open-source platform designed to lower the barriers between complex genomic data and biomedical researchers. It provides an intuitive web interface to visualize and analyze multidimensional cancer genomics data—such as mutations, copy number alterations, and mRNA expression—alongside clinical patient outcomes.

Investigating the stack

To understand how a platform of this scale operates, I used the BuiltWith technology profiling tool to audit the site’s infrastructure. The audit confirms that the platform relies on a cloud-native architecture primarily powered by Amazon Web Services (AWS).

You can view the full technical breakdown here: cbioportal.org Technology Profile (builtwith.com).

Cloud infrastructure: powered by AWS

To handle thousands of concurrent users and massive datasets, the public instance of cBioPortal is built on a cloud-native architecture using Amazon Web Services (AWS). Key components of its deployment include:

  • Amazon EKS (Kubernetes): Manages the deployment of frontend and backend containers, ensuring high availability and seamless updates.
  • Amazon EC2 & ELB: Provides scalable computing power and distributes incoming traffic via the Elastic Load Balancer to prevent site crashes during high-demand periods.
  • Amazon RDS: A managed database service (MySQL) that stores the structured genomic and clinical metadata essential for researcher queries.
  • Amazon S3: Often used as a “Storage for the Internet” to hold static assets and large data files securely.

Why the Cloud matters for biomedicine

By hosting cBioPortal in the cloud, the team at Memorial Sloan Kettering can provide a seamless experience where researchers don’t have to worry about hardware limitations. Cloud hosting allows for horizontal scaling—meaning as more researchers use the tool, the infrastructure automatically adds more capacity to keep the analysis fast and responsive.

References