Figure 2

Characterization of reads located 1000 nt around the 5′ and 3′ ends.

(A) Enrichment of reads in the deletion region versus the random region calculated for the entire region or for the first and last 1000 nt. Reads were counted either in the entire region or in the first and last 1000 nt regions of deletions and in random regions. Read counts were normalized to the size of regions, and then the normalized read counts were divided by the number of deletions. Calculations were done separately for all 1114 deletions or only for 56 deletions. The enrichment score of reads was calculated as the ratio of the mean read count in a deletion to the mean read count in the random region. The P-value was calculated by the two-sample Wilcoxon test of the mean difference. (B) Distribution of unique and multimapped reads. The values represent the percentage of unique and multimapped ncRNA reads stemming from deletions overlapping all regions, the genic and intergenic regions. (C) Distribution of sizes of unique reads in the −/+1000 nt regions of the 5′- and 3′-ends of deletions (left) and random regions (right). The n indicates the total number of reads, and the mean indicates the mean size of reads. The sd means the standard deviation. The P-value of the two-sample Wilcoxon test with regard to the mean difference of read sizes between the deletion and random region is less than 2.2e-16. (D) Distribution of sizesof unique reads excluding 36 nt reads in the −/+1000 nt regions of the 5′- and 3′-ends of deletions (left) and random regions (right). The same as in C.