Pre-computed validation datasets: HETA readouts for gold-standard cohorts

Pre-computed HETA readouts for leading multimodal datasets, available to license. Each dataset includes clinical and molecular data to analyze alongside H&E features.

Dataset 01 · Based on TCGA

OpenTME

HETA profiles for The Cancer Genome Atlas (TCGA), a gold-standard, multimodal dataset.

4,691

Whole-slide Images

4.1B+

Cells Analyzed

30M+

Spatial Features Quantified

8

Tumor Types Covered

Current coverage

WSIs per indication

1,125

Breast

1,041

Lung & Bronchus

600

Colorectum

457

Bladder

448

Prostate

411

Liver & Bile Ducts

400

Stomach

209

Pancreas

Dataset 02 · Based on SPARK

PanCAN SPARK

HETA profiles for SPARK, one of the most comprehensive multimodal pancreatic cancer datasets assembled to date. SPARK is licensed by PanCAN; request access directly through the SPARK platform.

1,400+

Pancreatic Cancer Patients

1.1B+

Cells Analyzed

4

Linked Data Modalities

Details

Built from PanCAN's Know Your Tumor and Precision Promise initiatives, which include real-world treatment and outcome data

Patient-reported outcomes, clinical, omics, and imaging data from the same patients

Cohort scale that is rare in pancreatic cancer, a disease with a five-year survival rate near 13%

Analysis-ready in a secure cloud environment

License OpenTME

Fill out the form to request a guided walkthrough, sample readouts, and details on licensing OpenTME. For SPARK, request access through PanCAN.

External Dataset Inquiries
We are unable to respond to requests from personal email addresses.
Select an answer
Thank you for your submission!
Oops! Something went wrong while submitting the form.