Fig 1.
Flowchart illustrating the CAM with default parameters.
The workflow of CAM includes three components: data pre-processing (black box), analysis (grey box) and QC (white box). The QC metrics for all historical MNase-seq data were precompiled for unbiased judgment in the QC component.
Fig 2.
(A) Regions with higher sequencing coverage exhibit higher resolution for the detection of nucleosome positioning. The barplot shows the nucleosome positioning score in a simulated perfect positioning region and no positioning region. For low sequencing coverage, the positioning score is almost the same for the perfect and no positioning regions, and the difference increases when the coverage increases. (B) Distribution of sequencing coverage of all historical data.
Fig 3.
AA/TT/AT di-nucleotide periodicity.
(A) Distribution of the rotational scores of all historical data. A rotational score < 0.08 was determined as a “Fail” (blue) in this measurement. (B) Examples of “Pass” and “Fail” MNase-seq data in this measurement.
Fig 4.
Nucleosomal DNA length distribution.
(A) The MNase digestion level (concentration) is related to the nucleosomal DNA length: samples with a higher MNase concentration exhibite shorter nucleosomal DNA length, whereas lower concentration samples exhibite longer nucleosomal DNA length. The MNase-seq data used for the comparison were obtained from a previous study in mouse ESCs (GSM2083105, GSM2083106, GSM1083107, GSM1083108). (B) The distribution of the nucleosome length from all of the historical data. Nucleosome length < 140 or > 155 was determined as “Fail” (blue) in this measurement. (C) Examples of MNase-seq data with “Pass” and “Fail” in this measurement. The vertical line labels 147 bp.
Fig 5.
Nucleosome profiles on promoter regions and custom regions.
Nucleosome profiles were generated and plotted as aggregate plots and heatmaps on (A) promoters and (B) custom regions. By default, the promoter regions were defined as -1 kb to +2 kb from the TSS of all refseq genes, and the custom regions were extended to +/-1 kb (by default) from the center of the regions. Both the promoters and custom regions were sorted by the MNase-seq read counts within the regions. An MNase-seq data from a human lymphoblastoid cell line (GSM907784) was selected to plot the nucleosome profiles on both promoters and custom regions.
Fig 6.
Distribution of promoter NFR score and promoter positioning score.
(A) Distribution of the promoter NFR score of all historical data. A promoter NFR score < 0.4 was determined as a “Fail” (blue) in this measurement. (B) Examples of “Pass” and “Fail” MNase-seq data in promoter NFR score. (C) Distribution of the promoter positioning scores of all historical data. A promoter positioning score > 0.4 was determined as a “Fail” (blue) in this measurement. (D) Examples of “Pass” and “Fail” MNase-seq data in promoter positioning score.
Fig 7.
Well-positioned nucleosome array.
(A) An example of detected well-positioned nucleosome arrays. An MNase-seq data from a human lymphoblastoid cell line (GSM907784) was selected to plot the nucleosome profile for the example region. (B) Well-positioned nucleosome arrays are enriched in both the downstream promoters and the union DHS sites. The enrichment fold (y-axis) represents the fold change of the observed proportion of nucleosome arrays on the downstream promoters (or the union DHS sites) to the expected proportion (Method section). The nucleosome arrays were detected using MNase-seq data from human lymphoblastoid cell line (GSM907784).