Hi!
I am working on a project and I have microarray time-course data for a control and treatment group. I found a set of genes that are changing over time across the two groups using an R package (limma), then I used the enrichR package to find which transcription factors are enriched for this set of genes which gave about 180 TFs.
I was wondering if there is an R package or a method that I could use to determine at what point these transcription factors are turning on so that I could say something about the order that they turn on.
I thought about making a plot of the average expression of target genes of a specific TF but from that I'm not sure how to tell at what point they turn on.
Thanks!
I doubt that this can be predicted. From those 180TFs probably the bulk is not even expressed in your cells, and from the remaining ones you do not know whether they're even binding. Is this EnrichR analysis based on ChIP-seq data or just motifs?
the enrichR analysis was just done with the list of gene names and the only data I have is the expression levels for each gene at each time points between 0 and 24 hours
Yes, sure, I mean how does EnrichR does it, does it take actual ChIP/binding data or is this based on motifs?
I don't think it does either, this paper says it calculates an enrichment score using a combination of the Fisher exact test and a z-score they developed
Calculating statistics, be it Z- or whatever score, is still based on some kind of data, and I wonder whether it is motifs or actual binding data. Try to find that out, as understanding how resutls come to pass is important to judge whether this is reliable or not -- motifs for example would be very unspecific as most factors of a transcription factor family share the same motif -- e.g. ETS factors, they all have more or less the same motif but there are like 25 of these guys and only a fraction is properly expressed in a given celltype.