Datasets
What data you can explore in Data Insight today, and what’s on the way.
How datasets work
Each Data Insight conversation is scoped to one dataset - the queries and charts draw only from the dataset you opened, and you can’t join across datasets in a single chat. Pick the dataset that holds the data you need, then ask your questions.
During beta, the catalogue shows every dataset PAICON offers, but only the public ones are queryable; the rest are visible as locked or upcoming entries.
Available now (public)
TCGA - The Cancer Genome Atlas
Pan-cancer clinical + genomic data. ~11K cases across 26 disease types and 56 primary sites.
Rich clinical coverage (cases, diagnoses, treatment, follow-up, exposure, and curated PanCanAtlas survival endpoints) alongside genomic layers - somatic mutations, copy number, DNA methylation, and miRNA expression. Good for cohort demographics, stage and survival analysis, and mutation-frequency questions.
CPTAC - Clinical Proteomic Tumor Analysis Consortium
Proteogenomic cancer data. ~1K+ cases across multiple study cohorts.
Mass-spectrometry proteome and phosphoproteome abundance paired with RNA-seq, copy number, somatic mutations, and clinical data, organised by study cohort (e.g. discovery vs confirmatory). Good for protein/phosphosite abundance comparisons, tumour-vs-normal contrasts, and multi-omic exploration.
Visible but not yet available
These appear in the catalogue but can’t be opened for questions during beta - they’re either PAICON-exclusive (access on request) or not yet live:
| Dataset | Type | Status |
|---|---|---|
| PAI_101 - Pathology Dataset (IND) | Histopathology & whole-slide images | PAICON Exclusive |
| PAI_102 - Pathology Dataset (POL) | Histopathology & whole-slide images | PAICON Exclusive |
| PAI_103 - Pathology Dataset (CHN) | Histopathology & whole-slide images | Coming soon |
| PAI_104 - Pathology & Radiology (MULTI) | Histopathology, radiology & clinical imaging | Coming soon |
For a locked dataset, use the catalogue’s contact option to enquire about access.
Can’t find the data you need?
If the cohort you’re after isn’t in an available dataset, switch to Data Request - describe what you need and the assistant will build a request to source it.