find transcription factor binding site with small sequence from UCSC

Question

[kdc15]

I am trying to identify whether a SNP will affect TF binding. Here is the DNA sequence that I have obtained:

>hg19_dna range=chr21:39819567-39819957 5'pad=0 3'pad=0 strand=+ repeatMasking=none
TAGACTTAGTCATGCTAATTAAGACAAAAATTAGACCTTATTAAAAAATT
TTTGCAAAACAGATACTCCCTTCCCATGGTGGCACTGGCTAGTGTGTTTT
ACAGGCTCAGAACAGGAAACAGAGCTGATGGCTGCTGCCTCTCTCCTTCC
TGGGCCTGAGACTCGACCAAGGCTCGCAGGGGAATAACACACTATGTAAT
GTTAGCTTTGGCTCACCCATGATGAGAAACATGAACAAATGTGTGATTTA
TATCAGAAATTGGCGAAAACAGGTTATATATATAAGAAACAGGCTAAATT
TAAAATTACAAACTAAAAAGGAAATGGCTGCAACAATAACAAACAAGCAA
TAACTTTAAAGTATTAGTATTTCTCCAAGATAAAGTAAGCT

Because it is such as short sequence, I am getting errors/non-reliable results from software such as HOMER and MEME. LASAGNA seems to have worked on the sequence, but I would like to view results in one of the other software mentioned above. Are there any suggestions on how to approach this?

Many Thanks

Answer 1

[ATpoint]

FIMO is the choice here as it scans for individual motif occurrence rather than checking for overrepresentation which is meaningless with a single sequence. fimo takes as input your fasta and a collection of transcription factor motifs which you could e.g. download from JASPAR. I typically find the nonredundant vertebrate core collection in MEME format helpful.

Answer 2

[GouthamAtla]

Does homer or meme tells if a SNP effects a TF binding ? You need to use tools like MotifbreakeR , unless I missed something from your statement "I am trying to identify whether a SNP will affect TF binding".

[kdc15]

Hi,

No you are correct, MotifbreakR will be needed to predict TF binding from the SNP. I wanted to first identfiy what motifs were present in this sequence, but wanted to provide context as to why I was searching such a small sequence.

[GouthamAtla]

Thanks for clarifying