Encode Genome Browser: How To Correlate Dnasei Hypersensitivity With Rna-Seq?

Question

[khandarius]

Hi! I'm interested in exploring the relationship of certain genes' expression and DNaseI hypersensitivity of their promoter regions. I would like to correlate hypersensitivity data with measures of gene activity such as RNA-seq data. I'm using the UCSC genome browser with ENCODE data but I have some problems. I can't seem to find data with comparable signal values in the Table Browser - the formats are such that the correlate -function can't be used. I can get signal peak values for DNaseI hypersensitivity but the RNA-seq tables just describe sequence reads. I'd be grateful for any advice on how to proceed.

-dk

[Asaf]

ENCODE and correlation are antonyms. I've heard from several people that the ENCODE data cannot be compared between different experiments.

Answer 1

[Alex Reynolds]

Perhaps you could bin your RNAseq reads, creating windowed signal that you can correlate with the DNaseI data. You probably want to match the DNaseI window step and width parameters, so that you are correlating apples with apples.

[khandarius]

Thanks for the tip - especially pointing me to bedops, the suite is very useful. I need peak signal values for transcript abundance, though. I think I'll try to extract the FPKM values of a transcript for different cell lines and correlate those with the hypersensitivity peaks of corresponding transcript TSS.