Define genes containing new found motif

Question

[Morris_Chair]

Hello guys,

I want to find out which of about 500 genes of my interest have a defined consensus sequence in the promoter, I want thinking to create a bed file and look the peaks with IGV but that's gonna take forever, is there any faster way ?

thank you :)

Answer 1

[ATpoint]

If you have it as PFM/PWM then you can use fimo from the MEME suite to check whether you can match the motif to each of the promoters.

[Morris_Chair]

Hi ATpoint,

I do have the PFM but how can I extract the promoter of all the genes? or how can I tell to FIMO to scan the genes of my interest?

thanks

[ATpoint]

You could get the TSS coordinates for the relevant genes from a GTF file and then simply take a window upstream of it, say 500bp which is often use as a proxy for the promoter. The coordinates then can be converted to fasta with bedtools getfasta, which then can go to fimo.

[Morris_Chair]

Hi ATpoint, I used this used defined function for getting the the TSS , it seems pretty much straightforward to use

tss <- getTSSinUCSCFormatFromEnsemblGTF("Homo_sapiens.GRCh38.95.gtf.gz")

I assume I got many TSS for single gene which should be the IRanges, is it correct?

    GRanges object with 6 ranges and 20 metadata columns:
      seqnames    ranges strand |   source     type     score     phase           gene_id     transcript_id   gene_type gene_status   gene_name
         <Rle> <IRanges>  <Rle> | <factor> <factor> <numeric> <integer>       <character>       <character> <character> <character> <character>
  [1]     chr1     11869      + |   HAVANA     exon      <NA>      <NA> ENSG00000223972.4 ENST00000456328.2  pseudogene       KNOWN     DDX11L1
  [2]     chr1     11872      + |  ENSEMBL     exon      <NA>      <NA> ENSG00000223972.4 ENST00000515242.2  pseudogene       KNOWN     DDX11L1

doing this promoter(), I can have a list of the promoters range, I think I can use this ranges directly without making TSS windows, what do you think?

thanks