Sample Dataset

Try PancrionDX with Sample Patient Data

Download curated genomic, transcriptomic, and radiomic inputs designed to reflect real pancreatic cancer biology.

All sample data is synthetic but biologically informed. No patient data is used.

Radiomic Features

Tumor diameter
Heterogeneity index
Necrosis ratio

Transcriptomic Data (.csv)

COL1A1 — high expression
FAP — elevated
KRAS pathway upregulated

Genomic Data (.vcf)

KRASG12D
TP53R175H
CDKN2ALoss

Data contents

What’s in the Sample?

Genomic Data

.vcf
  • Mutations in core PDAC genes (KRAS, TP53, CDKN2A)
  • Reflects realistic mutation frequencies and co-occurrence patterns
  • Formatted as a standard VCF with chromosome coordinates
  • Includes silent and functional variant examples

Transcriptomic Data

.csv
  • Expression values in TPM / log scale
  • Includes stromal and tumor genes (COL1A1, FAP)
  • Contains overlapping distributions and biological noise
  • Simulated to mirror real patient-to-patient variability

Radiomic Features

.csv
  • Tumor size, heterogeneity index, necrosis ratio
  • Simulated from known imaging trends in pancreatic tumors
  • Includes both early-stage and late-stage feature profiles
  • Normalized to match clinical imaging scale ranges

All sample data is synthetic but biologically informed. No patient data is used.


Quickstart guide

How to Use the Sample Data

01

Download dataset

Get the full .zip or individual files

02

Go to analysis page

Navigate to the PancrionDX tool

03

Upload your files

Drop in genomic, transcriptomic, radiomic inputs

04

View prediction

Review stage classification and interpretation

GeneHGVSpTypeVAFPatient
KRASp.G12VMissense0.48PAAD-001
TP53p.R248WMissense0.44PAAD-001
SMAD4p.R361*Nonsense0.39PAAD-001
CDKN2Ap.R58*Missense0.35PAAD-001
RNF43p.G659fsFrameshift Del0.21PAAD-001
BRAFp.V600EMissense0.51PAAD-002
GNASp.R201HMissense0.38PAAD-002
KRASp.G12DMissense0.52PAAD-003

Showing a preview of the sample dataset. Download the full file for complete records.



Clinical context

Why This Matters

Pancreatic ductal adenocarcinoma is often diagnosed late, when treatment options are limited. Tools that integrate multiple biological signals — genomic mutations, transcriptomic expression, and imaging features — may help identify risk patterns earlier and more accurately. This sample dataset is a starting point for exploring how those signals interact.

< 5%

Five-year survival rate for late-stage PDAC

Earlier detection window with multimodal integration

30

Somatic variant records across 4 sample patients