I used Homer to get the motif in the peaks of U2AF2 in the ENCODE eCLIP-seq dataset. But I can't get UUUUU motif, the motif I got is mainly CCAG motif.
Hi limeng - could you post some more detail on how you did the analysis? Did you restrict to the peaks that were enriched over input or did you use the entire set? Could you post the homer command you used? I ran homer on the K562 U2AF2 ENCODE peaks some time ago and recall getting something along the lines of UUUUUCAGG, which is more-or-less expected. Note that even though U2AF2 binds the PPT, the AG is expected to show up because of its close proximity (note that it is only the 5'-ends of the eCLIP reads that actually represent the crosslinking sites).
What Homer commands are you using? I've been trying to analyze some of the encode e-clip data, but I was using homer2 denovo instead of findmotif.pl. Did you use bed files or fasta files for your analysis?