Best MEME tool: Identify TF binding sites for putative transcription factor

[Martin Gordon]

Hi everyone, 

A colleague is interesting in identifying regulatory targets of a potential transcription factor. Targets for the TF have been investigated previously by another lab using CHiP-seq, but unfortunately the data (including raw sequencing data) was not made publicly available. 

In absence of the data, we were wondering if it would be possible to generate a list of potential transcription factor binding sites in the human genome (hg38)  given the gene name/sequence of the transcription factor, using a tool within the MEME suite? 

Thanks in advance everyone,

Martin

[cegrant]

Hi Martin,

None of the MEME Suite tools will identify TF binding sites by the gene name or sequence for the TF. The MEME Suite does have multiple tools for searching for TF binding sites, but they all require that you have the binding motif for the TF expressed as a position weight matrix in MEME format. The closest tool to what you are looking for is probably T-Gene which will predict links between a list of genomic loci you provide, and the genes for the relevant organism.

You could first do de novo motif analysis, using STREME or MEME to find a candidate binding motif for your TF by analyzing a collection of sequences you think are binding sites. Once you have the motif in hand you could then use FIMO to search for other sites that are significant matches to that motif.

Alternatively you might look at a database of TF like JASPAR to see if your TF is already known. You might also checkout RSAT (Regulatory Sequence Analysis Tools). A quick google search also came up with hTFTarget. I've never used it, but it looks relevant to your question.

[Martin Gordon]

Thank you for your response and suggestions. Unfortunately outside of the earlier study the gene product haven't been linked to a TF role, so I couldn't find it in the public DBs.   While we don't have any data, we have a list of ~70 genes shown in the experiment to be regulated by the TF, so I think using STREME/MEME  to construct binding motifs and screening these against the TF sequence sounds like a sensible approach, or at least the best we can do w/o any of the original data.

Thanks again,

Martin