How to identify transcription factors (TFs) from expression data comparison?

Question

[XC]

Dear All,

I have a dataset of gene expression which consists two kinds of cells: 5 replicates of testing vs 5 replicates of controlling. Is there any tools / web-based platforms which can generate TFs based on the differentially expressed genes between testing group and controlling group? For example, if I have a differentially expressed gene list between testing group and controlling group, may I find reliable TFs which result in the behavior of testing group? Then I can validate them experimentally.

Thank you very much.

[Michael]

I don't think this experiment design is suitable for detecting TF's. You will likely get more reliable predictions based on sequence similarity.

[XC]

Hi Michael, thank you very much for your suggestions. Right, based on sequence data we can get more reliable predictions, but now I only have microarray data with oligo probes, that's why I have to find solution based on the differentially expressed gene list.

Answer 1

[5utr]

Are these human cell lines? If you are interested in identifying TFs that might regulate a differentially expressed gene you can calculate the correlation between the expression profile of your gene and a list of TFs. Considering that the expression profile only includes 10 samples you probably would then want to confirm a similar strong correlation( or anticorrelation if TF is a negative regulator) in other expression dataset and look for predicted TFBS close to the genomic coordinates of your gene of interest.

[XC]

Hi Gian, thank you very much. Right, these are human cell lines. Your suggestion is great and meaningful.