Estimating enrichment of repetitive elements from high-throughput sequence data

Table 1 Combining reads mapping to canonical and instance sequences improves ability to resolve enrichment levels

	Canonical	Instance	Combined
Total number of input reads utilized	150,377	1,270,497	1,498,796
Fraction of repeat types with 0 input reads	0.57	0.05	0.04
Fraction of repeat types with less than 10 input reads	0.71	0.20	0.20
Fraction of repeat types with significant enrichment or depletion	0.20	0.69	0.69
Fraction of repeat types with CI less than 1	0.14	0.57	0.57
Median size of the CI	14.67	0.63	0.58
Mean size of the CI	9.25	1.87	1.77

Using mES H3K9me3 data, the table shows that incorporating reads mapping to the repeat instances allows enrichment estimates to be provided for a large number of repeat types for which canonical sequence alone is not sufficient. CI refers to a 95% confidence interval.

ISSN: 1474-760X