AtacAlignmentEnrichmentQualityMetric

subsampled_reads

Number of reads subsampled from the sample for calculation of all cross-correlation metrics

fri_dhs

Fraction of reads in universal DHS regions

fri_exclusion_list

Fraction of reads in regions from exclusion list

fri_prom

Fraction of reads in promoter regions

fri_enh

Fraction of reads in enhancer regions

estimated_fragment_len

Fragment length/strandshift. This is the estimated fragment length/strand shift for each dataset as estimated by strand shift cross-correlation analysis

corr_estimated_fragment_len

Value of correlation function at estimated fragment length.

phantom_peak

Strand shift value at which the phantom (false) peak in cross-correlation is observed. This is typically the read length.

corr_phantom_peak

Value of cross-correlation function at strand shift corresponding to phantom_peak (which is typically the read length).

argmin_corr

Strand shift corresponding to the minimum value of cross-correlation (min_corr).

min_corr

Minimum cross-correlation over a sufficiently wide range of strand shifts (typically -100 bp to ~3 times the expected size of DNA fragments based on sonication and size selection protocols).

NSC

Normalized strand cross-correlation = FRAGLEN_CC / MIN_CC. Ratio of strand cross-correlation at estimated fragment length to the minimum cross-correlation over all shifts.

RSC

Relative cross-correlation coefficient. Ratio of strand cross-correlation at fragment length and at read length

auc

The “area under the curve”, with a maximum value of 0.5. Lower values generally indicate higher and more focal enrichment.

syn_auc

The expected area under the curve of a perfectly behaved input sample having the same mean sequencing depth of a given sample. This is useful to put the observed AUC into perspective.

x_intercept

The point (on the X-axis) at which the curve is 0. This is approximately the percentage of the genome that was not sequenced in a particular sample.

syn_x_intercept

The expected X-intercept of a perfectly behaved input sample having the same mean sequencing depth of a given sample. This is useful to put the observed X-intercept into perspective.

elbow_pt

The elbow point attempts to measure the position at which the line turns upward. In practice, this is the point at which the plotted line is furthest from the line from the lower-left to the upper-right corner of the graph (equivalent to a perfect input sample with infinite coverage). The point returned is the position on the X-axis of this elbow point and higher values indicate more enrichment.

syn_elbow_pt

The expected elbow point of a perfectly behaved input sample having the same mean sequencing depth of a given sample. This is useful to put the observed elbow point into perspective.

jsd

The Jensen-Shannon distance between the replicate and the control.

syn_jsd

The Jensen-Shannon distance between a given sample and a perfect input sample with the same coverage depth (i.e., the plot generated from the Poisson probability mass function with lambda equal to the mean coverage in the sample).

pct_genome_enrich

The approximate percentage of the genome enriched in signal (e.g., bound by a transcription factor or having a certain histone modification).

diff_enrich

The differential enrichment between a given sample and that indicated by --JSDsample at the elbow point.

ch_div

The CHANCE divergence between the replicate and the control.

tss_enrichment

Enrichment metric. Fold change of average ATAC-seq signal coverage at gene TSSs relative to background bins distal (typically 2Kb away) from TSS.

AtacAlignmentQualityMetric

total_reads

Number of total reads passing QC"

total_reads_qc_failed

Number of total reads failing QC

duplicate_reads

Number of reads with duplicates passing QC

duplicate_reads_qc_failed

Number of reads with duplicates failing QC

mapped_reads

Number of mapped reads passing QC

mapped_reads_qc_failed

Number of mapped reads failing QC

pct_mapped_reads

Percent of mapped reads passing QC

paired_reads

Number of paired reads passing QC

paired_reads_qc_failed

Number of paired reads failing QC

read1

Number of read1 reads passing QC

read1_qc_failed

Number of read1 reads failing QC

read2

Number of read2 reads passing QC

read2_qc_failed

Number of read2 reads failing QC

properly_paired_reads

Number of properly paired reads passing QC

properly_paired_reads_qc_failed

Number of properly paired reads failing QC

pct_properly_paired_reads

Percent of properly paired reads passing QC

with_itself

Number of reads with both itself & mate mapped passing QC

with_itself_qc_failed

Number of reads with both itself & mate mapped failing QC

singletons

Number of singletons (unpaired reads) passing QC

singletons_qc_failed

Number of singletons (unpaired reads) failing QC

pct_singletons

Percent of singletons (unpaired reads) passing QC

diff_chroms

Number of reads with mate mapped to different chromosomes passing QC

diff_chroms_qc_failed

Number of reads with mate mapped to different chromosomes failing QC

non_mito_reads

Number of non-mitochondrial reads

mito_reads

Number of mitochondrial reads

frac_mito_reads

Fraction of mitochondrial reads

frac_reads_in_nfr

Fraction of reads in nucleosome free region (NFR)

frac_reads_in_nfr_qc_pass

Whether NFR/mono-nuc reads passes QC

frac_reads_in_nfr_qc_reason

Reason for frac_reads_in_nfr_qc_pass

nfr_over_mono_nuc_reads

Ratio of reads in nucleosome free region (NFR) to reads in mononucleosome region (Mono-Nuc).

nfr_over_mono_nuc_reads_qc_pass

Whether NFR/mono-nuc reads passes QC

nfr_over_mono_nuc_reads_qc_reason

Reason for nfr_over_mono_nuc_reads_qc_pass

nfr_peak_exists

Whether NFR peak exists

mono_nuc_peak_exists

Whether mononucleosome peak exists

di_nuc_peak_exists

Whether there are peaks in dinucleosome regions (Di-Nuc).

usable_fragments

Usable fragments, based on the mapped_reads value.

AtacLibraryQualityMetric

unpaired_reads

Number of unpaired reads before filtering

paired_reads

Number of paired reads before duplicate filtering

unmapped_reads

Number of unmapped reads before duplicate filtering

unpaired_duplicate_reads

Number of unpaired duplicates before duplicate filtering

paired_duplicate_reads

Number of paired duplicates before duplicate filtering

paired_optical_duplicate_reads

Number of paired optical duplicates before duplicate filtering

pct_duplicate_reads

Percent of paired duplicates before duplicate filtering

total_fragments

Number of fragments before duplicate filtering

distinct_fragments

Number of distinct fragments

positions_with_one_read

Number of locations to which exactly one read (pair) maps

NRF

Non redundant fraction (indicates library complexity). Number of distinct unique mapping reads (i.e. after removing duplicates) / Total number of reads

PBC1

PCR Bottlenecking coefficient 1 = M1/M_DISTINCT where M1: number of genomic locations where exactly one read maps uniquely, M_DISTINCT: number of distinct genomic locations to which some read maps uniquely

PBC2

PCR Bottlenecking coefficient 2 (indicates library complexity) = M1/M2 where M1: number of genomic locations where only one read maps uniquely and M2: number of genomic locations where 2 reads map uniquely

AtacPeakEnrichmentQualityMetric

frip

Fraction of reads in the peak file

min_size

Smallest peak width

25_pct

25th percentile of peak widths

50_pct

50th percentile of peak widths

75_pct

75th percentile of peak widths

max_size

Largest peak width

mean

Mean peak width

AtacReplicationQualityMetric

reproducible_peaks

Number of peaks called from one replicate or pooled replicates passing reproducibility test by comparing self-pseudoreplicates or pooled pseudo-replicates.

idr_cutoff

Irreproducible Discovery Rate (IDR) threshold used to define reproducible peaks

rescue_ratio

max(Np,Nt) / min (Np,Nt); Np: Pooled-pseudoreplicate consistent peaks (comparing two pseudoreplicates generated by subsampling pooled reads from Rep1 and Rep2); Nt: True Replicate consistent peaks (comparing true replicates Rep1 vs Rep2)

self_consistency_ratio

max(N1,N2) / min (N1,N2); N1: Replicate 1 self-consistent peaks (comparing two pseudoreplicates generated by subsampling Rep1 reads); N2: same as N1 for Rep2

reproducibility

Reproducibility test result for this experiment (pass/fail)