Extended Data Fig. 1: ChIP–seq data summary.
From: An atlas of dynamic chromatin landscapes in mouse fetal development

a, Summary of characteristic enrichment patterns for histone modifications surveyed here. Modifications are generally categorized as narrow or broad depending on the typical breadth of enrichment. H3K9me3 is further distinguished from other broad marks because it shows very few regions of enrichment in non-repetitive sequence in primary tissues and cells26. b, Metagene plot illustrating the typical patterns of histone modification enrichment at active genes (here defined as RPKM >10 in all tissue-stages surveyed). ChIP–seq data plotted are from embryonic heart at E15.5. c, Sequencing depth plotted for every library reported (n = 1,272 total, 552 narrow, 432 broad, 144 H3K9me3, 144 input). ENCODE ‘usable’ read depth standards (mapping quality scores (mapq) >30, and after PCR duplicate removal) are indicated to the right. Read depth standards changed part way through our study (increasing from 10M to 20M for narrow marks, 20M to 45M for broad marks, and 10M to 30M for input). All narrow mark libraries exceed the 10M minimal depth. Broad mark libraries exceed the 20M minimal depth with only four exceptions, all of which exceed 19M. Input libraries exceed the 10M minimal depth with only one exception, which exceeds 9.7M. The read depth standard for H3K9me3 is >45M mapped reads of any mapq (because H3k9me3 is enriched in repetitive sequence, Extended Data Fig. 10); all H3K9me3 libraries exceed this threshold. Box plots: horizontal line, median; box, IQR; whiskers, most extreme value within ±1.5 × IQR. d, Mapping quality plotted for every library, measured as the fraction of reads with mapq >30. Reads with lower mapq scores (that is, non-uniquely mapping reads) were eliminated from downstream analysis. e, Three metrics of library complexity are plotted (NRF, PBC1, PBC2). See ENCODE data standards90 for detailed descriptions and formulas. Tables below each plot show the percentage of libraries that exceed the thresholds indicated. f, Two measures of signal-to-noise ratio are plotted (NSC, RSC). Again, detailed descriptions are available in the ENCODE data standards descriptions. These metrics are not well calibrated for broad marks or input and thresholds apply only to narrow marks.