Microarray And Epigenomic Data For Same Cancer Cell Line?

Question

[dirigible2012]

I am doing a short project looking into the causes of misregulation of gene expression in cancer. I would like to compare epigenomic datasets with microarray expression datasets to see how much epigenetic dysregulation contributes to dysregulation of expression.

The problem is, I can't find a cancer dataset that contains both whole genome epigenetic data AND microarray expression data.

Does anybody have any ideas as to where I could find such a dataset, or how to go about looking?

(I'm aware it's only possible to get a subset of the epigenetic marks - I'd even settle for only methylation! :-)

Thank you,
Stephanie

[Emily]

I don't think anybody does microarrays any more. You probably want RNASeq data instead.

[dirigible2012]

Good point, thanks. It's more finding the epigenomics data which is giving me trouble. I wonder if people simply aren't doing whole genomes.

Answer 1

[lkmklsmn]

My best guess would be The Cancer Genome Atlas (TCGA).

[brentp]

I agree, for example here you can see there is expression and mRNA expression for AML. You can go here and choose rows that have both.

[Chirag Nepal]

I clicked the second link you provided to download the data. I selected two RNAseq files which are Tumor-Normal pair. Screenshot of which files were selected can be seen here.

For each library selected, there are 6 files (gene expression level, isoform levels and so on) and altogether 12 files for two libraries.

Since this is TN pair, the gene expression measure from the result file represent for which RNA (tumor or normal). Am i understanding something wrong about TN pair?

The selected file (small) can be downloaded from here.

Any suggestions will be helpful.

[Neilfws]

Yes, TCGA has many such datasets.

[dirigible2012]

Just what I was looking for, thanks!

Answer 2

[dario.garvan]

There are histone methylation, DNA methylation, and gene expression datasets available for the prostate cancer cell line, LNCaP.

TCGA has a lot of biological replicates for each type of experiment, so it is better.