exPATT.R is an R script to detect DNA binding motifs (e.g. OCT4) and change the genomic coordinates of each matching position at its center. The script filters genomic regions, recorded in an input BED file, that match a given DNA pattern depicted as a position weight matrix (PWM). These pattern matrices are matched in genomic regions using a minimum threshold given as a percentage of the highest possible score (e.g. 80%). Then, genomic regions with a minimum number of hit(s) are filtered and centered around the pattern to generate a new BEDGRAPH file.
Hereafter we consider that R version 4.3+ is installed on your computer.
Dependencies requieremnts can be installed using the setup.R script.
If Rscript is in your PATH you just type:
> ./setup.R
Otherwise you need to source this file from your R IDE:
R> source("setup.R")
During this installation process various packages will be installed. You can customize the list of packages (especially regarding Genomes annotations from Bioconductor) in setup.R.
The easiest way to execute exPATT.R is directly from your shell command line (given Rscript is in your PATH).
A command line helper is shown if no arguments are provided:
> exPATT.R
+-----------------------------------------------+
| (c) 2024 - CEA, CNRS, Paris-Saclay University |
+-----------------------------------------------+
$Id: 4309a63ab943ed917080c85d1d33af116620a9d6 $
======================================================================
This script is used to filter a BED file given a set of patterns.
USAGE:
%prog [-G <genome>] -b <input-bed-file> -p <pattern> [-m <min-score>]
[-l <min-hits>] [-h <max-hits>] [-w <bed-track-width>]
-o <output-bed-file>
or
%prog -L
options (regular mode):
-G : genome, as a Bioconductor annotation package
(default: Mmusculus.UCSC.mm9)
-b : input bed file to be filtered
-p : the DNA pattern(s), either as a filename or one of the following.
presets: OCT4,SOX2 (default: OCT4)
-m : pattern, minimum score (given as percentage of the maximum possible
score) for counting a match (default: 80)
-l : lowest/minimum number of hits per BED entries (default:1)
-h : highest/maximum number of hits per BED entries (default:none)
-w : width (in bp) of the track in the output file (default: 2000)
-o : the filename, including the '.bedgraph' extension to use for
storing results (default: <input-file>_out.bed)
options (genome list mode):
-L : list all available Bioconductor genomes
======================================================================
This program can be executed in two different modes:
- regular or query mode
- genomes list, using the
-Loption.
This execution mode allows to select regions in an input BED file that match a motif provided as a PWM matrix.
By default, exPATT search the Mmusculus.UCSC.mm9 genomic regions defined in the input BED file (here sample.bed) for the OCT4 motif with a minimum score of 80% and at least one hit. The output is a BEDGRAPH file of 2kb centered around each hit. The motif is searched on both strands. The execution output is given below:
> exPATT.R -b example/sample.bed
+-----------------------------------------------+
| (c) 2024 - CEA, CNRS, Paris-Saclay University |
+-----------------------------------------------+
$Id: 907226b92e8f6069a12ab8da81c2ef171dca4bef $
- Load Genome: done
- Import BED file: #8432 records
- Retrieve DNA sequences: done
- Search Pattern with min-score=80% (strands '+'/'-'):
|======================================================================| 100%
|======================================================================| 100%
- Found 1701 hit(s) on the '+' strand and 1677 on the '-' strand
- Select #hits in the range [1,+Inf]: 1156 on the '+' strand, 1132 on the '-' strand and 545 on both strands
- Filter & Process hit(s) (strands '+'/'-'):
|======================================================================| 100%
|======================================================================| 100%
exPATT includes two default motifs, OCT4 and SOX2. Still, you can use your own motifs using regular tabulated text files to describe the associated PWM matrix. As an example, the PWM matrix of OCT4 is given in the file example/oct4.pwm.
The -L gives a list of all available anotations packages in Bioconductor. Packages already installed in your R environment are indicated with a star character (*). For example, the genome annotations Mmusculus.UCSC.mm8 is installed as shown below.
> ./exPATT.R -L
+-----------------------------------------------+
| (c) 2024 - CEA, CNRS, Paris-Saclay University |
+-----------------------------------------------+
$Id: 907226b92e8f6069a12ab8da81c2ef171dca4bef $
Genome : OS Prov. Ver. Inst.
Alyrata.JGI.v1 : Alyrata JGI v1
Amellifera.BeeBase.assembly4 : Amellifera BeeBase assembly4
Amellifera.NCBI.AmelHAv3.1 : Amellifera NCBI AmelHAv3.1
...
Mmusculus.UCSC.mm39 : Mmusculus UCSC mm39
Mmusculus.UCSC.mm8 : Mmusculus UCSC mm8 *
Mmusculus.UCSC.mm8.masked : Mmusculus UCSC mm8
...
Vvinifera.URGI.IGGP12Xv0 : Vvinifera URGI IGGP12Xv0
Vvinifera.URGI.IGGP12Xv2 : Vvinifera URGI IGGP12Xv2
Vvinifera.URGI.IGGP8X : Vvinifera URGI IGGP8X
To install new annotations package you must do it from an R session. For example, if you want to install the Mmusculus.UCSC.mm39 version you must use the following command on the R console.
R> BiocManager::install("BSgenome.Mmusculus.UCSC.mm39")
Notice the BSgenome prefix that is not shown on the genomes list to optimize the display.
BRG1 generates subnucleosomes that expand OCT4 binding and function beyond DNA motifs at enhancers.
Marina C. Nocente, Anida Mesihovic Karamitsos, Emilie Drouineau, Waad Albawardi, Ccile Dulary, Florence Ribierre, Hlne Picaud, Olivier Alibert, Jol Acker, Jean-Christophe Aude, Nick Gilbert, Franoise Ochsenbein, Sophie Chantalat, Matthieu Grard
nature structural & molecular biology, 2024 - doi:10.1038/s41594-024-01344-0