SamBuckberry

#' Get the signal from bigwig file for defined regions.
#' 
#' @param bigwig Path to a bigwig file. Also accpets http paths. 
#' @param gr A genomic ranges object of regions to obtain signal. All ranges should be the same width.
#' @param range The distance from the centre of gr to flank. Default=100. 
#' @param log Logical. Is the signal in the bigwig file in log space? Default=FALSE.
#' If log = TRUE, the scores in the bigwig file are transformed by e^scores.
#' @return A numeric vector of aggregate signal if aggregate = TRUE, which is the default. 
#' If aggregate = FALSE a data.frame of signal of all regions is returned.
#' @export

SamBuckberry

library(magrittr)
library(GenomicRanges)
library(BSgenome.Mmusculus.UCSC.mm10)
library(rtracklayer)
library(reshape2)
library(ggplot2)
library(dplyr)
library(JASPAR2016)
library(TFBSTools)
library(stringr)

SamBuckberry

library(magrittr)
library(GenomicRanges)
library(BSgenome.Mmusculus.UCSC.mm10)
library(rtracklayer)
library(reshape2)
library(ggplot2)
library(dplyr)
library(JASPAR2016)
library(TFBSTools)
library(stringr)

SamBuckberry

### Get Tn5 bias track for regions

bias_bigwig <- "~/Desktop/mm10_bias_chr19.Scores.bigwig"
atac_regions <- "~/polo_iPSC/ATACseq/processed_data/atac_peaks/atac_all_peaks_union.bed"
atac_bam <- "~/Desktop/atac_iPSC_combined_replicates_chr19.bam"
ctcf <- read.table("~/R_packages/RunATAC/inst/exdata/ctcf.pwm") %>% as.matrix()

#' Read BED formatted file to GRanges object
#' 
#' @param bed_file Path Path to BED formatted file.

SamBuckberry

library(data.table)
library(R.utils)

aggregateStrandsCG <- function(CGmap){
        
    # Do not output scientific notation
    options(scipen=999)
        
    # Create the temp directory for uncompressed CGmap file
    tmpFile <- tempfile(tmpdir = tempdir(), fileext = ".CGmap.tmp")

SamBuckberry

# See examples at http://sambuckberry.github.io/2015/07/15/quickCorPlots/

plotAnnotatedScatter <- function(x, y, pointCol=rgb(0,0,0,0.7), 
                                 legendPos="topleft", legendCex=1, ... ){
        # Generate a linear model summary
        fit <- lm(y ~ x)
        fitSum <- summary(fit)
        r2 <- fitSum$r.squared
        pVal <- fitSum$coefficients[2,4]
        

SamBuckberry

#source("http://bioconductor.org/biocLite.R")
#biocLite(c("rtracklayer", "Rsamtools", "GenomicRanges"))

library(rtracklayer)
library(Rsamtools)
library(GenomicRanges)
library(GenomicAlignments)

#' SummarizeOverlapsByGene
#' 

SamBuckberry

## Write a bash script for looping through the bam files


#! /bin/bash
for i in *accepted_hits.bam
do
	samtools index ${i}
done

## Save this files as indexBamFiles.sh

SamBuckberry

# install the Rsamtools package if necessary
source("http://bioconductor.org/biocLite.R")
biocLite("Rsamtools")
 
# load the library
library(Rsamtools)

# specify the bam file you want to import
bamFile <- "test.bam"