Long Motif Predicted by MEME SUITE

Question

[fusion.slope]

Hello,

am wondering why MEME SUITE is predicting motif with the length as in the 1st in picture attached. Aren't motif in lenght 10-15 nucleotides? Thanks in advance for any reply.

[Santosh Anand]

??? I see the data of MEME and you are saying Homer? Also why did you include the results of MEME and DREME?

[fusion.slope]

yes sorry i said wrong. I correct immediately thanks.

Answer 1

[Biojl]

What are your input sequences for MEME? ChIP-seq peaks for a particular transcription (which one?) factor perhaps? Which is the species you are working with?

Motifs coming from transcription factors are typically 4 to 30 base pairs long, but longer motifs have been described in particular cases.

[fusion.slope]

The sequences are regions identified as hypermethylated after the Knock Down of a set of proteins involved in dna de-methylation. These regions are located mostly in LINE1 and ERVK repetitive elements. Is it possible that are promoters of retroviruses?

Thanks for your reply Biojl!

[Sinji]

The repetitive nature of the motif does indeed describe how you would imagine a LTR of a ERV would look like.

EDIT: Also looks like there may be some zinc finger binding proteins there, which as you probably know is indicative of a silenced ERV through the KAP1-SETDB1 complex.

[fusion.slope]

Thanks a lot but how have you realized this? Which of those proteins/TF is involved in such complexes?

[Sinji]

Just by looking at the known or similar motifs. You have a potential zinc finger protein. I work a lot on KAP1, who is historically known for it's repressive roles, and if you take a quick look at the literature you'll know that ERVs are silenced in Embryonic Stem Cells via this KAP1 binding to Zinc finger proteins, which recruit the SETDB1 protein and heterochromatinizes the ERV.

You may need to familiarize yourself with the literature more if you didn't immediately notice that since it's pretty common knowledge in the transposable element field.

[fusion.slope]

Thanks for the clarification but the question is:

why these regions are de-methylated by the proteins that am studying in regions that are usually silenced? A possible answer can be that many retrotransposones are activated in embryonic stem cells (the same cell type in which I have performed the experiments) as Anas Fadloun et al 2014 have described in their paper:

http://www.nature.com/nsmb/journal/v20/n3/abs/nsmb.2495.html

Also it is quite known that ERV elements are actively transcribed in mouse embryonic stem cells so i would not take a stable idea about transposones as repressed elements in the genome but actually they can regulate several processes during transcription as these guys have described:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3395470/

Thanks again.