hi ankit, getting TFBS for eukaryotes its quit challenging. what people generally do is "fetch the TSS site of corresponding gene and extract 3KB sequences from upstream and DO the MEME analysis" if you have any further query then just let me know

Hi Raghvendra sir. Thanks for your answer. When we do TFBS prediction from ChIP seq data then we get 200 length of read on average. So finding the motifs from meme will not be that difficult. But to find a motif for the sequence which are upstream region of  3 Kb region from TSS will be difficult for meme. What do you think?  

Hi Ankit,

for ChIP-Seq I can recommend Homer :http://homer.salk.edu/homer/ for ChIP-Seq data.

I'm really happy with that (includes perl skripts) - installation seems to be easy for someone familiar to linux/shell skripts and the documentation website of homer is very helpful. It'll find you known motifs and even new motifs for your enriched DNA sequences (retrieved by ChIP-Seq). Furthermore, mapped sequences to the genome can be annotated no matter where the peak was called in the genome.

Good luck!

Claudia

Thanks Claudia. But the problem is that i do not have ChIP-Seq data. I have only RNA-Seq data. I need to somehow get the 3'UTR or promoter sequence. But these sequence length will be much larger than Chip-Seq read length. So, I was wondering, in that case meme,meme-chip or homer will work or not. Thanks. 

hey Ankit, There is another way. if are working in any model organism like C. alegans, Arabidopsis, Yeast or E. coli. then there is a genome annotation file called GFF3/GFT and few other synonyms.

It gives complete details in-terms of coordinates.for example

http://mblab.wustl.edu/GTF22.html you can follow this link.

specifically for C. elegans you can go through this

ftp://ftp.wormbase.org/pub/wormbase/releases/WS220/species/c_elegans/

I have the same question. I have RNAseq data and need to find putative transcriptional factors that could be responsible for the change in gene expressions under our treatment. It's possible?

Yes it is possible. You might need to look for ChIP-seq intervals which are closer to your RNAseq interval.