← BioTransfer GEO Dataset Finder
GEO series

A Molecular Taxonomy of Urothelial Carcinoma

GSE32894 Homo sapiens (human) Expression microarray 308 samples Submitted 2012/05/24 Platform GPL6947
At a glance

Expression microarray · human · 308 samples

What it is
Expression microarrayGene activity read on a chip (the older technology that RNA-seq replaced).
Organism
Homo sapiens (human)
Topic
Bladder cancerThis series is part of our curated briefing on it.
Samples
308 samples: 308 samples, each with a distinct title — no replicated groupsGroups are read from the sample names, not from the paper. Check the list below.
Data you can get
A processed expression table (series matrix) and the array files, from NCBI.
Summary (written by the submitters)
Even though urothelial cancer (UC) is the fourth common tumor type among males, progress in treatment development has been deficient. Pathological assessment provides the urologists with only a broad classification, complicated by frequent disagreement among pathologist and the co-existence of different grading systems. Consequently, there is a great need for an objective, reproducible and biologically relevant classification system to make treatment more efficient.
Read the full summary ↓
In the present investigation we present a molecular taxonomy for UC stratification based on integrated genomics. We used gene expression profiles from 308 UC to define seven molecular subtypes using step-by-step partitions and a bootstrap approach. Results were validated in three independent and publically available data sets. The subtypes differ significantly with respect to expression of cell cycle genes, of receptor tyrosine kinases particularity FGFR3, ERBB2 (HER2), and EGFR, of an FGFR3 associated gene expression signature, of cytokaratins, and of cell adhesion genes. The subtypes also differ significantly with respect to FGFR3, PIK3CA, and TP53 mutations. The expression of key proteins was validated by IHC on TMA. A further inspection indicated that the subtypes could be reduced to four major types of UC; Urobasal/D-driven, Genomically unstable/E-driven, Evolved urobasal, and Basal/SCC like, with characteristic and highly divergent molecular phenotypes. We show that the molecular subtypes cut across pathological classification and that tumors classified as one subtype maintain their characteristic molecular phenotype irrespective of pathological stage and grade. Available data from the Drugbank database and the Cochrane central registry of controlled trials indicate that susceptibility to specific drugs is more likely to be associated with the molecular stratification than with pathological classification. The presented molecular taxonomy stratifies UC into subtypes with distinct molecular phenotypes and biological properties. We anticipate that the molecular taxonomy will be a useful tool in future clinical investigations.
This dataset
Open GSE32894 in finder → Find more human microarray datasets → Bladder cancer landscape →
Download · which file do I need?

Direct links to NCBI, no account and no request form. Pick by what you want to do: take the small processed table to look, the supplementary files to re-analyse, or the original raw data to start over.

1 · Just have a look

Processed table

The series matrix: one table of values per sample as the submitters processed it. The smallest download, and it opens in Excel or R. Start here to see what the data look like.

Series matrix ↓
2 · Re-analyse it

Supplementary files

The count tables and other files the authors uploaded. Take only the samples you need.

All supplementary files ↗
3 · The original raw files

Whole study · RAW.tar

Everything the authors deposited as one archive (GSE32894_RAW.tar), including the raw array or instrument files. Can be large; the picker in step 2 takes just the samples you need.

Whole study · RAW.tar ↓

What the files usually look like: For arrays the series matrix is normally the normalised expression table you want; the raw array files (CEL and similar) are in the RAW.tar.

Also filed as BioProject PRJNA146635. Searching any of these in the dataset finder brings you back here.

NCBI GEO page ↗ Paper (PMID 22553347) ↗
Study design
308 samples, each with a distinct title — no replicated groups

Any grouping that defines a comparison lives in the sample characteristics, which GEO records separately from the titles.

How we read this: Read from 308 sample titles: 308 distinct titles with little repetition. These groups are guessed from the sample names, not taken from the paper, so check them against the sample list below before relying on them.

Samples in this study

+ 268 more — browse all 308 samples with per-sample file links →

More Bladder cancer datasets

Search all human microarray datasets in GEO →

Share this dataset

Metadata from NCBI GEO, cached and refreshed periodically — the NCBI page above is authoritative. Downloads link straight to NCBI/ENA; nothing is proxied through BioTransfer.