Command: call¶
Call peaks in a qBED file
usage: blockify call [-h] -i INPUT [--prior PRIOR | --p0 P0]
[--method {OP,PELT}] [-r REGIONS] -bg BACKGROUND
[--intermediate INTERMEDIATE]
(-a ALPHA | -p PVALUECUTOFF) [--correction CORRECTION]
[-d DISTANCE] [--min MIN] [--max MAX] [-t | -s]
[-c PSEUDOCOUNT] [--measure {enrichment,depletion}]
output
Named Arguments¶
- -i, --input
Input file
- output
Output file (BED format); default: stdout
- --prior
Explicit prior on the number of blocks (not recommended for general use)
- --p0
Empirical prior based on a specified false-positive rate; must be between 0 and 1 (default: 0.05)
- --method
Possible choices: OP, PELT
Segment using the optimal partitioning (OP) or pruned exact linear time (PELT) algorithm (default: “PELT”)
- -r, --regions
Regions over which to normalize event counts; should be supplied as a BED file. If not provided, the input file will be segmented using Bayesian blocks.
- -bg, --background
Background qBED file
- --intermediate
Intermediate file to write verbose output (CSV format)
- -a, --alpha
Alpha for multiple hypothesis correction (must be between 0 and 1)
- -p, --pValueCutoff
p-value cutoff (NOTE: This is a straight cutoff and will not take into account multiple hypothesis correction!)
- --correction
If alpha provided, need to specificity method of multiple hypothesis correction. See statsmodels.stats.multitest for a complete list of choices (default: “bonferroni”)
- -d, --distance
Merge features closer than this distance (bp)
- --min
Report peaks larger than this cutoff (bp)
- --max
Report peaks smaller than this cutoff (bp)
- -t, --tight
Shrink peak boundaries to overlap data points
- -s, --summit
Return peak summits
- -c, --pseudocount
Pseudocount for background regions (default: 1)
- --measure
Possible choices: enrichment, depletion
Perform a one-tailed test for either enrichment or depletion relative to the background file (default: “enrichment”)