search

gifford-lab / GEM3

GEM software suite (GEM, GPS, KMAC, KSM, RMD, etc.)
10 stars 5 forks source link

comparing two KSMs #19

Open orionzhou opened 4 years ago

orionzhou commented 4 years ago

Thanks for developing this package I found KSM representation is really superior to the traditional PWMs. However, I wonder if there is a tool to compare two KSMs calculating their similarity? Previously when using PWMs, I can easily calculate a Pearson correlation efficient or distance score between any pair of PWMs, but with KSMs I'm just not sure how to do that ? Any thoughts?

yguoma commented 4 years ago

This is an interesting question. I am not aware of methods to do that. Here is my quick take. You might want to scan the two KSM motifs on same set of sequences, e.g. the cCRE regions from ENCODE, or all open chromatin regions from a cell type of interest. Then compare either the number of overlaps, or correlation of KSM scores.

On Wed, Oct 28, 2020 at 1:14 PM Peng Zhou notifications@github.com wrote:

Thanks for developing this package I found KSM representation is really superior to the traditional PWMs. However, I wonder if there is a tool to compare two KSMs calculating their similarity? Previously when using PWMs, I can easily calculate a Pearson correlation efficient or distance score between any pair of PWMs, but with KSMs I'm just not sure how to do that ? Any thoughts?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/gifford-lab/GEM3/issues/19, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACA4XBCN5U66YLBJFCPJYVLSNBGPHANCNFSM4TCTYY2Q .