Fast-NR

Name	Fast-NR
Author	Na He
Version	1.0

Fast-NR is a software which applied on STARR-seq data to identify the negative regulatory elements, like silencer or insulator.

Workflow

The workflow of Fast-NR as following:

Installation

You can install Fast-NR either with using pip or git repository.

1. Using git repository

git clone https://github.com/Na-He/Fast-NR.git

unzip Fast-NR-main.zip

cd Fast-NR-main

python setup.py install

2. Using git repository

   pip install git+https://github.com/Na-He/Fast-NR.git

Dependencies

Fast-NR requires

• numpy
• pysam
• scipy

Command

Fast-NR -t <controlFile> -c <treatmentFile> -g <genomeFile> -o <outputName> [options]

Parameters

requires:

parameter	type	Description
-t	file	Control file path and name, usually the STARR-seq plasmid library. Only .bam and .bed file accpected.
-c	file	Treatment file path and name, usually the STARR-seq cDNA library. Only .bam and 3 column or 6 column .bed file accpected. But this program will not consider the strand information.
-g	file	Genome chromosome size file, which can download from the UCSC. File include two column, chromosome, length of chromosome.
-o	file	Output path and prefix name of the final result peaks.

options:

parameter	type	Description
-p	value	The cut-off of p value. Default 10-5.
-cp	string	The correct method of p value. Include "BH", "Bonferroni". Default "Bonferroni".
-cm	string	The method used to calculate the curve similarity. Include 'Cosine', 'Pearson', 'Euclidean', and 'Gradiente'. Default 'Cosine'.
-ct	value	The percent cut-off of similarity distance. From 0 to 1. Default 0.9.
-ws	value	The size of window, used to find differential coverage region. Default 600 bp.
-l	value	Extend the fragment length to this fixed bp length. Default 0.
--remove	logical	Remove the slop or not. Set to "TRUE" if want to remove the results in slop regions. Default "FALSE".

Output Format

The final output file is a .bed file, include 9 columns.

• column 1 | chromosome. The chromosome name.
• column 2 | start. The start position of peak region.
• column 3 | end. The end position of peak region.
• column 4 | summit. Summit of peak region, also the mid position of peak.
• column 5 | tCount. Number of fragments in treatment library in peak region.
• column 6 | cCount. Number of fragments in control library in peak region.
• column 7 | FC. The fold change of fragment counts in peak region compared with treatment count and control count.
• column 8 | P. P value of peak region.
• column 9 | cP. Corrected p value.

Example:

chr2L 103500 104100 103800 120 224 0.535714 2.198757e-13 3.493824e-10

chr2L 114264 114864 114564 8 112 0.071429 1.877090e-36 2.982696e-33

chr2L 143773 144373 144073 260 440 0.590909 7.688970e-19 1.221777e-15

chr2L 148969 149569 149269 139 329 0.422492 1.780870e-30 2.829802e-27

---

Citation:

Once you used this program, please cite the following paper:

(https://github.com/Na-He/Fast-NR)