Abstract
Background
DNA double-stranded breaks (DSBs) are potentially deleterious events in a cell. The end structures (blunt, 3′- and 5′-overhangs) at DSB sites contribute to the fate of their repair and provide critical information concerning the consequences of the damage. Therefore, there has been a recent eruption of DNA break mapping and sequencing methods that aim to map at single-nucleotide resolution where breaks are generated genome-wide. These methods provide high resolution data for the location of DSBs, which can encode the type of end-structure present at these breaks. However, genome-wide analysis of the resulting end structures has not been investigated following these sequencing methods.
Results
To address this analysis gap, we develop the use of a coverage-normalized cross correlation analysis (CNCC) to process the high-precision genome-wide break mapping data, and determine genome-wide break end structure distributions at single-nucleotide resolution. We take advantage of the single-nucleotide position and the knowledge of strandness from every mapped break to analyze the relative shifts between positive and negative strand encoded break nucleotides. By applying CNCC we can identify the most abundant end structures captured by a break mapping technique, and further can make comparisons between different samples and treatments. We validate our analysis with restriction enzyme digestions of genomic DNA and establish the sensitivity of the analysis using end structures that only exist as a minor fraction of total breaks. Finally, we demonstrate the versatility of our analysis by applying CNCC to the breaks resulting after treatment with etoposide and study the variety of resulting end structures.
Conclusion
For the first time, on a genome-wide scale, our analysis revealed the increase in the 5′ to 3′ end resection following etoposide treatment, and the global progression of the resection. Furthermore, our method distinguished the change in the pattern of DSB end structure with increasing doses of the drug. The ability of this method to determine DNA break end structures without a priori knowledge of break sequences or genomic position should have broad applications in understanding genome instability.
FICC-Seq: a method for enzyme-specified profiling of methyl-5-uridine in cellular RNA
