Fig. 1

Determination of DSB end structures by CNCC. a Outline of the CNCC analysis for three types of breaks: blunt end (left), 3′ end overhang (middle) or 5′ end overhang (right). Each break produces two DNA ends and each of those are processed, 3′ overhangs trimmed and 5′ overhangs filled in (“Methods”), ligated to a sequencing adaptor (green rectangles), and sequenced. These reads result in distinctive patterns of coverage on both positive (blue) and negative strands (red) (fourth row). Conducting our genome-wide CNCC analysis between coverage on positive and negative strands then reveals a shift that is characteristic of the type of DSB end structure (fifth row). b CNCC analysis of DNA breaks caused by HaeIII, BanII and BbvI restriction enzyme cleavage (left, middle, and right column, respectively). The mean read coverage over the indicated enzyme motifs presents precise location of mapped reads (top two rows), and genome-wide CNCC spikes (bottom row, blue) at − 1 for HaeIII, − 5 for BanII, and + 3 for BbvI exactly reflect the expected end structures, with shuffled controls in gray. CNCC analysis was employed using all sequencing data, and was not limited to reads found at enzyme motifs