Discovering Motifs in Ranked Lists of DNA Sequences

Paper by Eran Eden, Doron Lipson, Sivan Yogev and Zohar Yakhini
Talk by Yonit Halperin

I will present the above recently published work with an emphasis on its theoretical parts. The paper introduces a tool, DRIM, which discovers motifs that are enriched at the top (bottom) of a given ranked list of sequences, without prior partitioning of the data into to target and background sets. The authors use a score called mHG (minimum hypergeometric) to rank the motifs.
I will present the score and discuss some theoretic and algorithmic issues related to mHG and to the calculation of its exact p-value. Finally, I will also show some of the results that are presented in the paper.