Identification of functionally relevant regions from ChIP-seq experiments

Question

[Dataminer]

Hello!

We all know that through ChIP-seq experiments a lot of binding sites are identified for a transcription factor but the major question which remains(or atleast for me) is that, how many of these ChIP-seq sites are functionally important?

So for instance if you have a transcription factor binding sites identified from a ChIP-seq experiment and this TF binds mostly in promoter (as the sites identifiedd are present in promoter regions). How can one identify the functionally important sites for this transcription factor out of the number of sites identified through ChIP-seq experiment. Also, if one has the availability of RNA-seq data how this data can be efficiently be coupled with ChIP-seq to aid the identification of potential functional regions.

Please share your experience and logic that you would choose for the given set of data and also if you have across some articles please share.

Thank you

Answer 1

[Ming Tommy Tang]

Actually, people have demonstrated that only around 10% of the binding sites are functional:

After knock down it, expression of only 10% of the nearby genes are affected.

The Functional Consequences of Variation in Transcription Factor Binding

On average, 14.7% of genes bound by a factor were differentially expressed following the knockdown of that factor, suggesting that most interactions between TF and chromatin do not result in measurable changes in gene expression levels of putative target genes.

see several other papers:

Extensive Divergence of Transcription Factor Binding in Drosophila Embryos with Highly Conserved Gene Expression

Transcription Factors Bind Thousands of Active and Inactive Regions in the Drosophila Blastoderm

To assign functional binding sites to their target genes, one may consider to use BETA developed in Shirely Liu's lab. It integrates ChIP-seq data and gene expression data to infer the TF target genes. You may be also interested in this paper:

Assessing Computational Methods for Transcription Factor Target Gene Identification Based on ChIP-seq Data

Answer 2

[Sukhi Singh]

This really depends how you define what is functional.

Protein binding can regulate the gene expression (activation and repression) but sometimes its can just be sitting at a specific locus with no effect. From your list of TF binding sites, you can subset them using expression profiling

1. promoters which loose expression when that TF or other interactor is removed (knocked out/down)
2. promoters which gain expression when that TF or other interactor is removed (knocked out/down)
3. promoters which have no effect when that TF or other interactor is removed (knocked out/down)

ChIP-Sequencing for a chemical modification signature (methylation, acetylation etc) or an interactor, when TF or other interactor is removed

1. promoters gaining / losing the signature mark 2, elongation, constriction or total loss/gain of TF or an interactor on a specific locus or loci.

All these analysis can reveal some specific information regarding what you are interested in.

With expression profiling, you can also try co-relating the protein binding with the alternative splicing and the transcripts produced, plus the effect in the absence or presence of that specific TF.

Even if you don't have your own expression profiling, you can still use public wild type datasets for your cell lines to generate a list of gene expression values and answer what kind of genes my TF binds, high, low, medium expressed etc.

You can also find some tools where you can integrate your datasets with others available.

Target analysis by integration of transcriptome and ChIP-seq data with BETA