Fig. 1

De novo ChIP-seq analysis pipeline. Top: Reads from the TF experiment are assembled using a de novo assembly method (SEECER or Velvet) leading to longer ChIPtigs, each of which is based on several (often hundreds or thousands) of assembled reads. Bottom right: Each of the assembled ChIPtigs is scored to determine its enrichment for experiment vs. control reads. ChIPtigs are ranked based on their enrichment scores. Bottom left: Top ranking ChIPtigs are used as input for a motif discovery method resulting in a set of motifs for the experiments