histrader 1.0.2

A tool to identify nucleosome free regions from ChIP-Seq of histone modifications

HisTrader

Introduction

HisTrader is a tool designed to identify Nucleosome-Free Regions (NFRs) from ChIP-Seq data targeting histone modifications. It takes a signal file (bedGraph or bigWig) and a broad peak file as input, and outputs BED files identifying NFRs and nucleosome-occupied regions (NORs) within those peaks.

NFRs are detected using a combination of moving averages and second-order differencing to identify valley-like patterns within peaks.

For more details, please refer to our publication: https://doi.org/10.1101/2020.03.12.989228

---

Installation

Requires Python 3.8+ and the following packages:

pip install histrader

# or

pip install numpy pandas pyranges pyBigWig pytest
pip install .

---

Usage

python histrader.py --signal ChIP.bedGraph/bigWig --peaks ChIP.bed

Required Parameters

Flag	Description
--signal	ChIP-Seq signal file in bedGraph or bigWig format
--peaks	Broad peak file in BED format

Optional Parameters

Flag	Description	Default
--genome	Genome FASTA file (used to extract DNA sequences from NFRs)	—
--trim	Trim extracted FASTA sequences (requires --genome and --trimSize)	off
--trimSize	Length (bp) of trimmed sequences, centred on each NFR	100
--out	Output file prefix	Histrader
--method	NFR detection method: MA, DIFF, or BOTH	BOTH
--step	Fixed step size (bp) for the converted signal	25
--minSize	Minimum peak size (bp) for NFR calling	500
--nucSize	Estimated nucleosome size (bp); should be divisible by --step	150
--mergeMulti	Step multiplier for merging (merge distance = mergeMulti × step)	3
--maMulti	Nucleosome multiplier for the slow moving average	3
--pMax	Fraction of peak max signal used as a zeroing threshold (0 = off)	0.0
--filter	Discard NFRs wider than this value (bp)	1000
--maxValley	Keep only the NFR at the max peak region per peak	off
--useDifferential	With --maxValley, use differential-based NFR selection	off
--randValley	Keep only one random NFR per peak	off
--seed	Random seed for reproducible --randValley output	—
--outBG	Also output the fixed-step signal within peaks (bedGraph)	off

---

Output Files

File	Contents
<prefix>.nfr.bed	Nucleosome-Free Regions
<prefix>.nuc.bed	Nucleosome-Occupied Regions
<prefix>.missing.bed	Peaks where no NFR was detected
<prefix>.nfr.fa	NFR sequences (only with --genome)
<prefix>.nuc.fa	Nucleosome sequences (only with --genome)
<prefix>.bedGraph	Fixed-step signal track (only with --outBG)

---

Example

python histrader.py \
  --signal  ChIP.bedGraph \
  --peaks   ChIP.bed \
  --method  BOTH \
  --step    25 \
  --nucSize 150 \
  --pMax    0.1 \
  --out     output_prefix

---

Test Data

Test data for a single H3K27ac peak on chr11 is provided in TEST_DATA/:

python histrader.py \
  --signal TEST_DATA/test.region.histrader.chr11_12286115_12289133.H3K27AC.bdg \
  --peaks  TEST_DATA/test.region.histrader.chr11_12286115_12289133.H3K27AC.broadPeak \
  --out    test_output

Tests

pytest test_histrader.py -v

---

Citation

If you use HisTrader in your research, please cite:

HisTrader: A Tool to Identify Nucleosome Free Regions from ChIP-Seq of Histone Post-Translational Modifications Eftyhios Kirbizakis, Yifei Yan, Ansley Gnanapragasam, Xiaoyang Zhang, and Swneke Bailey bioRxiv, 2020 https://doi.org/10.1101/2020.03.12.989228