R/plots.R
In hanasusak/cDriver: Find best driver candidate genes
Defines functions
plotSamplesMut
plotMutChange
boxplotCCF.mutations
boxplotCCF.patients
plotStaircase
Documented in
boxplotCCF.mutations
boxplotCCF.patients
plotMutChange
plotSamplesMut
plotStaircase
######################################################################################
# CRG
# Hana SUSAK
#------------------------------------------------------------------------------------
# plotc for cDriver
######################################################################################
#' Plot mutations per sample distribution.
#' @description
#' \code{plot.samplesMut} Function to plot samples mutation counts.
#' @param sample.mutations data frame in MAF like format.
#' Columns (with exactly same names) which \code{sample.mutations} should have are:
#' \itemize{
#' \item Variant_Classification column specifed by MAF format, used to distinguish between silent and nonsilent SNVs
#' \item Hugo_Symbol column specifed by MAF format, which reports gene for each SNV.
#' \item Tumor_Sample_Barcode column specifed by MAF format, reporting for each SNV in wich patient was found.
#' \item Variant_Type columns pecifed by MAF format; is mutation SNV or InDel
#' }
#' @param indels a boolean value indicating should indels be counted. By default it is True.
#' @param silent a boolean value indicating should silent mutations be counted. By default it is True.
#' @param fill a boolean value indicating should plot only represent proportion between 3 types of mutations, not counts. By default it is False.
#' @param Tumor_Sample_Barcode (optional) integer/numeric value indicating column in \code{sample.mutations} which have sample ids for SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Hugo_Symbol (optional) integer/numeric value indicating column in \code{sample.mutations} having gene names for reported SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Classification (optional) integer/numeric value indicating column in \code{sample.mutations} which contain classification for SNV (Silent or not).
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Type (optional) integer/numeric value indicating column in \code{sample.mutations} which contains indormation if mutations is SNV or InDel .
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @return ggplot2 object
#' @examples
#' \donttest{
#' # plot sample's mutations , all of them
#' plotSamplesMut(sample.genes.mutect)
#' # plot proportion of silent and nonsilent, without indels
#' plotSamplesMut(sample.genes.mutect, indels=FALSE, fill=TRUE)
#' }
#' @export
plotSamplesMut <- function(sample.mutations, indels=TRUE, silent=TRUE, fill=FALSE, Tumor_Sample_Barcode=NULL, Hugo_Symbol=NULL, Variant_Classification=NULL, Variant_Type=NULL ) {
if (is.atomic(sample.mutations)) {
sample.mutations <- data.frame(x = sample.mutations)
}
if (!is.null(Variant_Classification)){
sample.mutations <- assign.columns(sample.mutations, Variant_Classification, "Variant_Classification")
}
if (!is.null(Variant_Type)){
sample.mutations <- assign.columns(sample.mutations, Variant_Type, "Variant_Type")
}
if (!is.null(Hugo_Symbol)){
sample.mutations <- assign.columns(sample.mutations, Hugo_Symbol, "Hugo_Symbol")
}
if (!is.null(Tumor_Sample_Barcode)){
sample.mutations <- assign.columns(sample.mutations, Tumor_Sample_Barcode, "Tumor_Sample_Barcode")
}
if(!is.factor(sample.mutations$Tumor_Sample_Barcode)){
sample.mutations$Tumor_Sample_Barcode <- factor(sample.mutations$Tumor_Sample_Barcode, levels=unique(sample.mutations$Tumor_Sample_Barcode))
}
##########
# make it not sensitive to lower/upper case in column names
original.col.names <- colnames(sample.mutations)
num.col <- ncol(sample.mutations)
colnames(sample.mutations) <- tolower(colnames(sample.mutations))
# take only exonic
suppressWarnings( sample.mutations <- exonic.only(sample.mutations))
if (indels){
#indel
sample.gene.indels <- sample.mutations[sample.mutations$variant_type %in% c('INS', 'DEL' ),]
sample.gene.indels$typeMut <- 'Indel'
ord.ind <- names(sort(table(sample.gene.indels$tumor_sample_barcode),decreasing=T))
} else {
sample.gene.indels <- c()
}
if (silent){
#silent
sample.gene.silent <- sample.mutations[sample.mutations$variant_type %in% c('SNP', 'DNP','TNP' ) & sample.mutations$variant_classification == "Silent",]
sample.gene.silent$typeMut <- 'Silent'
if (indels){
sample.gene.silent$tumor_sample_barcode <- factor(sample.gene.silent$tumor_sample_barcode, levels=ord.ind)
}
ord.sil <- names(sort(table(sample.gene.silent$tumor_sample_barcode),decreasing=F))
} else {
sample.gene.silent <- c()
}
#nonsilent
sample.gene.non.silent <- sample.mutations[sample.mutations$variant_type %in% c('SNP', 'DNP','TNP' ) &
sample.mutations$variant_classification %in% c('Frame_Shift_Del', 'Frame_Shift_Ins', 'In_Frame_Del',
'Missense_Mutation', 'Nonsense_Mutation', 'Splice_Site',
'Translation_Start_Site', 'Nonstop_Mutation'),]
sample.gene.non.silent$typeMut <- 'NonSilent'
#all
sample.gene.all <- rbind(sample.gene.non.silent, sample.gene.silent, sample.gene.indels)
if (silent){
sample.gene.all$tumor_sample_barcode <- factor(sample.gene.all$tumor_sample_barcode, levels=ord.sil)
} else if (indels) {
sample.gene.all$tumor_sample_barcode <- factor(sample.gene.all$tumor_sample_barcode, levels=ord.ind)
}
order <- names(sort(table(sample.gene.all$tumor_sample_barcode),decreasing=T))
sample.gene.all$tumor_sample_barcode <- factor(sample.gene.all$tumor_sample_barcode, levels=order)
max.y <- max(table(sample.gene.all$tumor_sample_barcode))
if (fill) {
p1 <- ggplot(data=sample.gene.all, aes_string('tumor_sample_barcode', fill='typeMut')) + geom_bar(position='fill') +
scale_y_continuous( name='Proportions', expand = c(0, 0))
} else {
p1 <- ggplot(data=sample.gene.all, aes_string('tumor_sample_barcode', fill='typeMut')) + geom_bar(position='stack') +
scale_y_continuous(name='Count', breaks=round(seq(0, max.y, length.out=10)), expand = c(0, 0))
}
p1 <- p1 + scale_x_discrete( expand = c(0, 0), drop=FALSE) +
theme_bw(base_size = 10) +
scale_fill_manual(values=c( "orange4",'orange1',"steelblue3")) +
theme( axis.text.x = element_text(size = 10 *0.5, angle = 90, hjust = 0.8, vjust=0.5, colour = "grey50")) +
theme(axis.title.x=element_text(angle=0, vjust=+0.5, size = 10 )) +
guides(fill = guide_legend(title = "Mutation type"))
#scale_fill_discrete(guide = guide_legend(title = "Mutation type"))
#theme( axis.text.x =element_blank()) +
#theme(axis.title.x=element_blank())
#return original names
colnames(sample.mutations)[1:num.col] <- original.col.names
p1 <- p1 + xlab('Sample ID')
p1
}
#' Plot mutations pattern frequncies.
#' @description
#' \code{plotMutChange} Function to plot frequency of mutation changes.
#' @param sample.mutations data frame in MAF like format.
#' Columns (with exactly same names) which \code{sample.mutations} should have are:
#' \itemize{
#' \item Tumor_Sample_Barcode column specifed by MAF format, reporting for each SNV in wich patient was found.
#' \item Variant_Type columns pecifed by MAF format; is mutation SNV or InDel
#' \item Variant_Classification column specifed by MAF format, used to distinguish between silent and nonsilent SNVs
#' \item Reference_Allele column specifed by MAF format, represents referent allele
#' \item Tumor_Seq_Allele2 column specifed by MAF format, represents alternative allele
#' }
#' @param silent a boolean value indicating should silent mutations be counted. By default it is True.
#' @param fill a boolean value indicating should plot only represent proportion between 3 types of mutations, not counts. By default it is True
#' @param Tumor_Sample_Barcode (optional) integer/numeric value indicating column in \code{sample.mutations} which have sample ids for SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Type (optional) integer/numeric value indicating column in \code{sample.mutations} which contains indormation if mutations is SNV or InDel .
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Classification (optional) integer/numeric value indicating column in \code{sample.mutations} which contain classification for SNV (Silent or not).
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Reference_Allele (optional) integer/numeric value indicating column in \code{sample.mutations} which contain reference alleles.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Tumor_Seq_Allele2 (optional) integer/numeric value indicating column in \code{sample.mutations} which contain alternative alleles.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @return ggplot2 object
#' @examples
#' \donttest{
#' # plot SNVs changes patterns
#' plotMutChange(sample.genes.mutect)
#' # plot SNVs changes patterns only for non silent mutaions
#' plotMutChange(sample.genes.mutect, silent=FALSE, fill=FALSE)
#' }
#' @export
plotMutChange <- function(sample.mutations, silent=TRUE, fill=TRUE, Tumor_Sample_Barcode=NULL, Variant_Type=NULL, Variant_Classification=NULL, Reference_Allele=NULL, Tumor_Seq_Allele2=NULL ) {
if (is.atomic(sample.mutations)) {
sample.mutations <- data.frame(x = sample.mutations)
}
if (!is.null(Tumor_Sample_Barcode)){
sample.mutations <- assign.columns(sample.mutations, Tumor_Sample_Barcode, "Tumor_Sample_Barcode")
}
if (!is.null(Variant_Type)){
sample.mutations <- assign.columns(sample.mutations, Variant_Type, "Variant_Type")
}
if (!is.null(Variant_Classification)){
sample.mutations <- assign.columns(sample.mutations, Variant_Classification, "Variant_Classification")
}
if (!is.null(Reference_Allele)){
sample.mutations <- assign.columns(sample.mutations, Reference_Allele, "Reference_Allele")
}
if (!is.null(Tumor_Seq_Allele2)){
sample.mutations <- assign.columns(sample.mutations, Tumor_Seq_Allele2, "Tumor_Seq_Allele2")
}
##########
# make it not sensitive to lower/upper case in column names
original.col.names <- colnames(sample.mutations)
num.col <- ncol(sample.mutations)
colnames(sample.mutations) <- tolower(colnames(sample.mutations))
# take only exonic
suppressWarnings( sample.mutations <- exonic.only(sample.mutations) )
if(!is.factor(sample.mutations$tumor_sample_barcode)){
sample.mutations$tumor_sample_barcode <- factor(sample.mutations$tumor_sample_barcode, levels=unique(sample.mutations$tumor_sample_barcode))
}
if(!silent){
sample.mutations <- sample.mutations[sample.mutations$variant_classification != 'Silent',]
}
sample.mutations <- sample.mutations[sample.mutations$variant_type %in% c('SNP', 'DNP','TNP' ),]
sample.mutations$Change <- base::apply(sample.mutations, 1, function(x)paste(x['reference_allele'], x['tumor_seq_allele2'], sep=';') )
sample.mutations[sample.mutations$Change == 'G;T','Change'] <- 'C;A'
sample.mutations[sample.mutations$Change == 'G;C','Change'] <- 'C;G'
sample.mutations[sample.mutations$Change == 'G;A','Change'] <- 'C;T'
sample.mutations[sample.mutations$Change == 'A;G','Change'] <- 'T;C'
sample.mutations[sample.mutations$Change == 'A;T','Change'] <- 'T;A'
sample.mutations[sample.mutations$Change == 'A;C','Change'] <- 'T;G'
sample.mutations$Change <- unlist(lapply(sample.mutations$Change, function(x) gsub(';','->',x)))
sample.mutations$Change <- factor((sample.mutations$Change), levels=c( 'C->T', 'T->C', 'C->A', 'C->G', 'T->A','T->G'))
sample.mutations$tumor_sample_barcode <- factor(sample.mutations$tumor_sample_barcode, levels=names(sort(table(sample.mutations$tumor_sample_barcode), decreasing=T) ))
p2 <- ggplot(sample.mutations, aes_string('tumor_sample_barcode', fill='Change')) + geom_bar(position=position_fill()) + theme_bw(base_size = 10) +
scale_y_continuous(name='Proportion',expand=c(0,0)) +
scale_x_discrete(name='Sample ID',drop=FALSE) + scale_fill_brewer(palette="Set2") +
theme( axis.text.x = element_text(size = 10 *0.5, angle = 90, hjust = 0.8, vjust=0.5, colour = "grey50")) +
theme(axis.title.x=element_text(angle=0, vjust=+0.5, size = 10 ))
#return original names
colnames(sample.mutations)[1:num.col] <- original.col.names
p2
}
#' Plot boxplot of mutations for top genes.
#' @description
#' \code{boxplotCCF.mutations} Function to plot CCF's of all mutations for top genes.
#' @param sample.mutations data frame in MAF like format.
#' Columns (with exactly same names) which \code{sample.mutations} should have are:
#' \itemize{
#' \item Tumor_Sample_Barcode column specifed by MAF format, reporting for each SNV in wich patient was found.
#' \item Hugo_Symbol column specifed by MAF format, which reports gene for each SNV.
#' \item CCF numeric column produce by \code{CCF} function.
#' \item Variant_Classification column specifed by MAF format, used to distinguish between silent and nonsilent SNVs
#' \item Variant_Type columns pecifed by MAF format; is mutation SNV or InDel
#' }
#' @param result.df a data frame with Hugo_Symbol column. Genes in this column should be ordered by importance.
#' @param topGenes a numeric/integer value; How meny top genes from results will be ploted.
#' @param silent a boolean value indicating should silent mutations be counted. By default it is True.
#' @param indels a boolean value indicating should indels be counted. By default it is True.
#' @param Tumor_Sample_Barcode (optional) integer/numeric value indicating column in \code{sample.mutations} which have sample ids for SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Hugo_Symbol (optional) integer/numeric value indicating column in \code{sample.mutations} having gene names for reported SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param CCF (optional) integer/numeric value indicating column in \code{sample.mutations} which have cancer cell fraction information for SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Classification (optional) integer/numeric value indicating column in \code{sample.mutations} which contain classification for SNV (Silent or not).
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Type (optional) integer/numeric value indicating column in \code{sample.mutations} which contains indormation if mutations is SNV or InDel .
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param color a numeric or character value indicating column to color the samples.
#' @param shape a numeric or character value indicating column to use as shape for the samples.
#' @return ggplot2 object
#' @examples
#' \donttest{
#' #get CCF column
#' sample.genes.mutect <- CCF(sample.genes.mutect)
#' # get background
#' bcgr.prob <- bcgr.combine(sample.genes.mutect, length.genes$Hugo_Symbol, length.genes$Coverd_len)
#' # get ranking
#' df1 <- bayes.risk(sample.genes.mutect, bcgr.prob, prior.sick = 0.00007)
#' df2 <- bayes.driver(sample.genes.mutect, bcgr.prob, prior.driver = 0.001)
#' df.final <- combine.ranking(list(df1, df2), min.mut = 2 )
#' # plot boxplot of CCF for top genes
#' boxplotCCF(sample.mutations=sample.genes.mutect, result.df=df.final)
#' }
#' @export
boxplotCCF.mutations <- function(sample.mutations, result.df, topGenes=20, silent=FALSE, indels=TRUE,
Tumor_Sample_Barcode=NULL, Hugo_Symbol=NULL, CCF=NULL, Variant_Classification=NULL, Variant_Type = NULL, color=NULL, shape=NULL ) {
if (is.atomic(sample.mutations)) {
sample.mutations <- data.frame(x = sample.mutations)
}
if (!is.null(Tumor_Sample_Barcode)){
sample.mutations <- assign.columns(sample.mutations, Tumor_Sample_Barcode, "Tumor_Sample_Barcode")
}
if (!is.null(Hugo_Symbol)){
sample.mutations <- assign.columns(sample.mutations, Hugo_Symbol, "Hugo_Symbol")
}
if (!is.null(CCF)){
sample.mutations <- assign.columns(sample.mutations, CCF, "CCF")
}
if (!is.null(Variant_Classification)){
sample.mutations <- assign.columns(sample.mutations, Variant_Classification, "Variant_Classification")
}
if (!is.null(Variant_Type)){
sample.mutations <- assign.columns(sample.mutations, Variant_Type, "Variant_Type")
}
##########
# make it not sensitive to lower/upper case in column names
original.col.names <- colnames(sample.mutations)
num.col <- ncol(sample.mutations)
colnames(sample.mutations) <- tolower(colnames(sample.mutations))
if (!is.null(color) & is.character(color)){
color <- tolower(color)
}
if (!is.null(shape) & is.character(shape)){
shape <- tolower(shape)
}
# take only exonic
suppressWarnings( sample.mutations <- exonic.only(sample.mutations) )
if(!'Hugo_Symbol' %in% colnames(result.df)){
stop('There must be Hugo_Symbol column in result.df data frame!')
}
if(!is.factor(sample.mutations$hugo_symbol)){
sample.mutations$hugo_symbol <- factor(sample.mutations$hugo_symbol, levels=unique(sample.mutations$hugo_symbol))
}
plotGenes <- as.character(result.df$Hugo_Symbol[1:min(topGenes,nrow(result.df))])
sample.mutations <- sample.mutations[sample.mutations$hugo_symbol %in% plotGenes, ]
if(!silent){
sample.mutations <- sample.mutations[sample.mutations$variant_classification != 'Silent',]
}
if(!indels){
sample.mutations <- sample.mutations[! sample.mutations$variant_type %in% c('DEL', 'INS' ),]
}
order.gene <- as.table(by(sample.mutations[,'ccf'],sample.mutations[,'hugo_symbol'],median))
sample.mutations$hugo_symbol <- factor(sample.mutations$hugo_symbol , levels=names(sort(order.gene, decreasing=T)))
if (is.null(color) & is.null(shape)){
p <- ggplot(sample.mutations, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(size=2.5, alpha=0.65, position = position_jitter(width = .3, height=0))
}
#just color
if (!is.null(color) & is.null(shape)){
if((is.numeric(color) | is.integer(color)) & color <= ncol(sample.mutations) ){
col.var <- colnames(sample.mutations)[color]
} else if (is.character(color) & color %in% colnames(sample.mutations)){
col.var <- color
} else {
stop('color need to be or numeric indicator of column in sample.mutations data frame or charater with column name.')
}
p <- ggplot(sample.mutations, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(aes_string(color=col.var),size=2.5, alpha=0.65, position = position_jitter(width = .3, height=0))
}
# just shape
if (!is.null(shape) & is.null(color)){
if((is.numeric(shape) | is.integer(shape)) & shape <= ncol(sample.mutations) ){
sh.var <- colnames(sample.mutations)[shape]
} else if (is.character(shape) & shape %in% colnames(sample.mutations)){
sh.var <- shape
} else {
stop('shape need to be or numeric indicator of column in sample.mutations data frame or charater with column name.')
}
n.sh <- length(unique(sample.mutations[,sh.var]))
if (n.sh <= 11 ) {
sh.seq <- 15:(n.sh+14)
} else {
sh.seq <- c(15:25,1:(n.sh-11))
}
p <- ggplot(sample.mutations, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(aes_string(shape=sh.var),size=2.5, alpha=0.65, position = position_jitter(width = .3, height=0)) +
scale_shape_manual(values=sample(sh.seq, replace=F) )
}
if (!is.null(shape) & !is.null(color)){
if((is.numeric(shape) | is.integer(shape)) & shape <= ncol(sample.mutations) ){
sh.var <- colnames(sample.mutations)[shape]
} else if (is.character(shape) & shape %in% colnames(sample.mutations)){
sh.var <- shape
} else {
stop('shape need to be or numeric indicator of column in sample.mutations data frame or charater with column name.')
}
n.sh <- length(unique(sample.mutations[,sh.var]))
if (n.sh <= 11 ) {
sh.seq <- 15:(n.sh+14)
} else {
sh.seq <- c(15:25,1:(n.sh-11))
}
if((is.numeric(color) | is.integer(color)) & color <= ncol(sample.mutations) ){
col.var <- colnames(sample.mutations)[color]
} else if (is.character(color) & color %in% colnames(sample.mutations)){
col.var <- color
} else {
stop('color need to be or numeric indicator of column in sample.mutations data frame or charater with column name.')
}
p <- ggplot(sample.mutations, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(aes_string(color=(col.var), shape=(sh.var)),size=2.5, alpha=0.65, position = position_jitter(width = .15, height=0)) +#, drop=FALSE
scale_shape_manual(values=sample(sh.seq, replace=F))
}
#return original names
colnames(sample.mutations)[1:num.col] <- original.col.names
return(p)
}
#' Plot boxplot of patients mutations CCF's for top genes.
#' @description
#' \code{boxplotCCF.patients} Function to plot CCF's of aggregated patient-gene mutatuons, for top genes.
#' @param sample.mutations data frame in MAF like format.
#' Columns (with exactly same names) which \code{sample.mutations} should have are:
#' \itemize{
#' \item Tumor_Sample_Barcode column specifed by MAF format, reporting for each SNV in wich patient was found.
#' \item Hugo_Symbol column specifed by MAF format, which reports gene for each SNV.
#' \item CCF numeric column produce by \code{CCF} function.
#' \item Variant_Classification column specifed by MAF format, used to distinguish between silent and nonsilent SNVs
#' \item Variant_Type columns pecifed by MAF format; is mutation SNV or InDel
#' }
#' @param result.df a data frame with Hugo_Symbol column. Genes in this column should be ordered by importance.
#' @param topGenes a numeric/integer value; How meny top genes from results will be ploted.
#' @param silent a boolean value indicating should silent mutations be counted. By default it is True.
#' @param indels a boolean value indicating should indels be counted. By default it is True.
#' @param mode a charechter value indicationg how to solve when in one gene-sample pair there are multiple mutations. Options are SUM, MAX and ADVANCE
#' @param epsilon a numeric value. If mode is ADVANCE, epsilone value will be threshold for CCF difference to decide if they are in same or different clone.
#' @param sample.gene.lim a numeric value specifying upper limit when summing is perfomed for gene-sample pair
#' @param Tumor_Sample_Barcode (optional) integer/numeric value indicating column in \code{sample.mutations} which have sample ids for SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Hugo_Symbol (optional) integer/numeric value indicating column in \code{sample.mutations} having gene names for reported SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param CCF (optional) integer/numeric value indicating column in \code{sample.mutations} which have cancer cell fraction information for SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Classification (optional) integer/numeric value indicating column in \code{sample.mutations} which contain classification for SNV (Silent or not).
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Type (optional) integer/numeric value indicating column in \code{sample.mutations} which contains indormation if mutations is SNV or InDel .
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param color (optional) charachter value indicating column in \code{sample.mutations} to used as color dots.
#' Default is NULL value and grey dots gonna be ploted. This variable should be unique by patients.
#' @param shape (optional) charachter value indicating column in \code{sample.mutations} to used as shape for dots .
#' Default is NULL value and grey dots gonna be ploted. This variable should be unique by patients.
#' @return ggplot2 object
#' @examples
#' \donttest{
#' #get CCF column
#' sample.genes.mutect <- CCF(sample.genes.mutect)
#' # get background
#' bcgr.prob <- bcgr.combine(sample.genes.mutect)
#' # get ranking
#' df1 <- bayes.risk(sample.genes.mutect, bcgr.prob, prior.sick = 0.00007)
#' df2 <- bayes.driver(sample.genes.mutect, bcgr.prob, prior.driver = 0.001)
#' df.final <- combine.ranking(list(df1, df2), min.mut = 2 )
#' # plot boxplot of CCF for top genes
#' boxplotCCF(sample.mutations=sample.genes.mutect, result.df=df.final)
#' }
#' @export
boxplotCCF.patients <- function(sample.mutations, result.df, topGenes=20, silent=FALSE, indels=TRUE, mode='MAX', epsilon=0.05, sample.gene.lim=1,
Tumor_Sample_Barcode=NULL, Hugo_Symbol=NULL, CCF=NULL, Variant_Classification=NULL, Variant_Type = NULL, color=NULL, shape=NULL) {
if (is.atomic(sample.mutations)) {
sample.mutations <- data.frame(x = sample.mutations)
}
mode <- toupper(mode)
if ( !mode %in% c('SUM', 'MAX', 'ADVANCE')) {
stop("mode mast be or SUM or MAX or ADVACE! ", call. = FALSE)
}
if (!is.null(Tumor_Sample_Barcode)){
sample.mutations <- assign.columns(sample.mutations, Tumor_Sample_Barcode, "Tumor_Sample_Barcode")
}
if (!is.null(Hugo_Symbol)){
sample.mutations <- assign.columns(sample.mutations, Hugo_Symbol, "Hugo_Symbol")
}
if (!is.null(CCF)){
sample.mutations <- assign.columns(sample.mutations, CCF, "CCF")
}
if (!is.null(Variant_Classification)){
sample.mutations <- assign.columns(sample.mutations, Variant_Classification, "Variant_Classification")
}
if (!is.null(Variant_Type)){
sample.mutations <- assign.columns(sample.mutations, Variant_Type, "Variant_Type")
}
##########
# make it not sensitive to lower/upper case in column names
original.col.names <- colnames(sample.mutations)
num.col <- ncol(sample.mutations)
colnames(sample.mutations) <- tolower(colnames(sample.mutations))
# take only exonic
suppressWarnings( sample.mutations <- exonic.only(sample.mutations ))
if ( (!is.null(color)) & is.character(color) ) {
if (tolower(color) %in% colnames(sample.mutations)){
color <- tolower(color)
} else {
stop('color variable must be one of the column names in sample.mutations data frame')
}
} else if (!is.null(color)){
stop('color variable must be character type')
}
if ( (!is.null(shape)) & is.character(shape) ) {
if (tolower(shape) %in% colnames(sample.mutations)){
shape <- tolower(shape)
} else {
stop('shape variable must be one of the column names in sample.mutations data frame')
} } else if (!is.null(shape)){
stop('shape variable must be character type')
}
if(!'Hugo_Symbol' %in% colnames(result.df)){
stop('There must be Hugo_Symbol column in result.df data frame!')
}
if(!is.factor(sample.mutations$hugo_symbol)){
sample.mutations$hugo_symbol <- factor(sample.mutations$hugo_symbol, levels=unique(sample.mutations$hugo_symbol))
}
advance.ccf.calc <- function(x, epsilon){
x <- sort(x, decreasing=T)
sum.ccf <- x[1]
if (length(x) >1){
for (i in 2:length(x)) {
if ((x[i-1] - x[i]) > epsilon){
sum.ccf <- sum.ccf + x[i]
}
}
}
return(sum.ccf)
}
plotGenes <- as.character(result.df$Hugo_Symbol[1:min(topGenes,nrow(result.df))])
sample.mutations <- sample.mutations[sample.mutations$hugo_symbol %in% plotGenes, ]
if(!silent){
sample.mutations <- sample.mutations[sample.mutations$variant_classification != 'Silent',]
}
if(!indels){
sample.mutations <- sample.mutations[! sample.mutations$variant_type %in% c('DEL', 'INS' ),]
}
# AGGREGATE SNPs which are in same sample-gene pair
if (mode == 'SUM' ){
# if we sum values of column CCF, for same sample-gene pairs
if (is.null(sample.gene.lim)){
sample.gene.lim <- Inf
}
f <- paste( "ccf ~ hugo_symbol + tumor_sample_barcode", collapse="")
sample.mutations2 <- aggregate(as.formula(f) , data=sample.mutations, function(x) min(sum(x),sample.gene.lim))
if (!is.null(color)){
df.temp <- sample.mutations[,c(color,'tumor_sample_barcode')]
df.temp <- df.temp[!duplicated(df.temp), ]
first.n <- nrow(sample.mutations2)
sample.mutations2 <- merge(x=sample.mutations2, y=df.temp, by='tumor_sample_barcode',all.x=T)
second.n <- nrow(sample.mutations2)
if(first.n!=second.n){
warning('You choose to color by paramether which is not unique by patients!')
}
}
if (!is.null(shape)){
df.temp <- sample.mutations[,c(shape,'tumor_sample_barcode')]
df.temp <- df.temp[!duplicated(df.temp), ]
first.n <- nrow(sample.mutations2)
sample.mutations2 <- merge(x=sample.mutations2, y=df.temp, by='tumor_sample_barcode',all.x=T)
second.n <- nrow(sample.mutations2)
if(first.n!=second.n){
warning('You choose to shape by paramether which is not unique by patients!')
}
}
} else if (mode == 'MAX' ) {
# if we take max value of column CCF, for same sample-gene pairs
if (is.null(sample.gene.lim)){
sample.gene.lim <- Inf
}
f <- paste("ccf ~ hugo_symbol + tumor_sample_barcode", collapse="")
sample.mutations2 <- aggregate(as.formula(f) , data=sample.mutations, function(x) min(max(x, na.rm=T),sample.gene.lim))
if (!is.null(color)){
df.temp <- sample.mutations[,c(color,'tumor_sample_barcode')]
df.temp <- df.temp[!duplicated(df.temp), ]
first.n <- nrow(sample.mutations2)
sample.mutations2 <- merge(x=sample.mutations2, y=df.temp, by='tumor_sample_barcode',all.x=T)
second.n <- nrow(sample.mutations2)
if(first.n!=second.n){
warning('You choose to color by paramether which is not unique by patients!')
}
}
if (!is.null(shape)){
df.temp <- sample.mutations[,c(shape,'tumor_sample_barcode')]
df.temp <- df.temp[!duplicated(df.temp), ]
first.n <- nrow(sample.mutations2)
sample.mutations2 <- merge(x=sample.mutations2, y=df.temp, by='tumor_sample_barcode',all.x=T)
second.n <- nrow(sample.mutations2)
if(first.n!=second.n){
warning('You choose to shape by paramether which is not unique by patients!')
}
}
} else if (mode == 'ADVANCE' ){
# if we take max or sum value of column CCF, for same sample-gene pairs, depending if in same or different clone
if (is.null(sample.gene.lim)){
sample.gene.lim <- Inf
}
#sample.gene.dt <- sample.gene.dt[order(sample.gene.dt$Hugo_Symbol, sample.gene.dt$Tumor_Sample_Barcode, -sample.gene.dt$CCF),]
f <- paste( "ccf ~ hugo_symbol + tumor_sample_barcode", collapse="")
sample.mutations2 <- aggregate(as.formula(f) , data=sample.mutations, function(x) min(advance.ccf.calc(x, epsilon),sample.gene.lim))
if (!is.null(color)){
df.temp <- sample.mutations[,c(color,'tumor_sample_barcode')]
df.temp <- df.temp[!duplicated(df.temp), ]
first.n <- nrow(sample.mutations2)
sample.mutations2 <- merge(x=sample.mutations2, y=df.temp, by='tumor_sample_barcode',all.x=T)
second.n <- nrow(sample.mutations2)
if(first.n!=second.n){
warning('You choose to color by paramether which is not unique by patients!')
}
}
if (!is.null(shape)){
df.temp <- sample.mutations[,c(shape,'tumor_sample_barcode')]
df.temp <- df.temp[!duplicated(df.temp), ]
first.n <- nrow(sample.mutations2)
sample.mutations2 <- merge(x=sample.mutations2, y=df.temp, by='tumor_sample_barcode',all.x=T)
second.n <- nrow(sample.mutations2)
if(first.n!=second.n){
warning('You choose to shape by paramether which is not unique by patients!')
}
}
}
order.gene <- as.table(by(sample.mutations2[,'ccf'],sample.mutations2[,'hugo_symbol'],median))
sample.mutations2$hugo_symbol <- factor(sample.mutations2$hugo_symbol , levels=names(sort(order.gene, decreasing=T)))
if ( (!is.null(color)) & (!is.null(shape))){
p <- ggplot(sample.mutations2, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(aes_string(color=(color), shape=(shape)),size=3, alpha=0.75, position = position_jitter(width = .3, height=0))
} else if ( !is.null(color) ){
p <- ggplot(sample.mutations2, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(aes_string(color=(color)),size=3, alpha=0.75, position = position_jitter(width = .3, height=0))
} else if ( !is.null(shape) ){
p <- ggplot(sample.mutations2, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(aes_string(shape=(shape)),size=3, alpha=0.75, position = position_jitter(width = .3, height=0))
} else {
p <- ggplot(sample.mutations2, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(size=3, alpha=0.75, position = position_jitter(width = .3, height=0))
}
#return original names
colnames(sample.mutations)[1:num.col] <- original.col.names
return(p)
}
#' Plot heatmap for genes-sample pairs .
#' @description
#' \code{plotStaircase} Function to plot heatmap for genes-sample pairs orderd by genes which explain most of the patients .
#' @param sample.mutations data frame in MAF like format.
#' Columns (with exactly same names) which \code{sample.mutations} should have are:
#' \itemize{
#' \item Tumor_Sample_Barcode column specifed by MAF format, reporting for each SNV in wich patient was found.
#' \item Hugo_Symbol column specifed by MAF format, which reports gene for each SNV.
#' \item Variant_Classification column specifed by MAF format, used to distinguish between silent and nonsilent SNVs
#' \item Variant_Type columns pecifed by MAF format; is mutation SNV or InDel
#' }
#' @param result.df a data frame with Hugo_Symbol column. Genes in this column should be ordered by importance.
#' @param topGenes a numeric/integer value; How meny top genes from results will be ploted.
#' @param silent a boolean value indicating should silent mutations be counted. By default it is True.
#' @param indels a boolean value indicating should indels be counted. By default it is True.
#' @param allSamples a boolean value indicating if all samples should be plotted.
#' @param Tumor_Sample_Barcode (optional) integer/numeric value indicating column in \code{sample.mutations} which have sample ids for SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Hugo_Symbol (optional) integer/numeric value indicating column in \code{sample.mutations} having gene names for reported SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Classification (optional) integer/numeric value indicating column in \code{sample.mutations} which contain classification for SNV (Silent or not).
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Type (optional) integer/numeric value indicating column in \code{sample.mutations} which contains indormation if mutations is SNV or InDel .
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param color a numeric or character value indicating column to color the samples.
#' @param groupMax a numeric value which will be taken as maximum value when grouping color column for samples and genes pairs.
#' @param order a character value which can be or \code{'frequency'} or \code{'CCF_sum'}.
#' @return ggplot2 object
#' @examples
#' \donttest{
#' #get CCF column
#' sample.genes.mutect <- CCF(sample.genes.mutect)
#' # get background
#' bcgr.prob <- bcgr.combine(sample.genes.mutect)
#' # get ranking
#' df1 <- bayes.risk(sample.genes.mutect, bcgr.prob, prior.sick = 0.00007)
#' df2 <- bayes.driver(sample.genes.mutect, bcgr.prob, prior.driver = 0.001)
#' df.final <- combine.ranking(list(df1, df2), min.mut = 2 )
#' # plot boxplot of CCF for top genes
#' plotStaircase(sample.mutations=sample.genes.mutect, result.df=df.final)
#' }
#' @export
plotStaircase <- function(sample.mutations, result.df, topGenes=40, silent=FALSE, indels=TRUE, allSamples= FALSE,
Tumor_Sample_Barcode=NULL, Hugo_Symbol=NULL, Variant_Classification=NULL, Variant_Type = NULL,
color='CCF' , groupMax=1, order='frequency') {
if (is.atomic(sample.mutations)) {
sample.mutations <- data.frame(x = sample.mutations)
}
if (!is.null(Tumor_Sample_Barcode)){
sample.mutations <- assign.columns(sample.mutations, Tumor_Sample_Barcode, "Tumor_Sample_Barcode")
}
if (!is.null(Hugo_Symbol)){
sample.mutations <- assign.columns(sample.mutations, Hugo_Symbol, "Hugo_Symbol")
}
if (!is.null(Variant_Classification)){
sample.mutations <- assign.columns(sample.mutations, Variant_Classification, "Variant_Classification")
}
if (!is.null(Variant_Type)){
sample.mutations <- assign.columns(sample.mutations, Variant_Type, "Variant_Type")
}
##########
# make it not sensitive to lower/upper case in column names
original.col.names <- colnames(sample.mutations)
num.col <- ncol(sample.mutations)
colnames(sample.mutations) <- tolower(colnames(sample.mutations))
# take only exonic
suppressWarnings( sample.mutations <- exonic.only(sample.mutations) )
if (!is.null(color) & is.character(color)){
color <- tolower(color)
}
if(!'Hugo_Symbol' %in% colnames(result.df)){
stop('There must be Hugo_Symbol column in result.df data frame!')
}
if(!is.factor(sample.mutations$tumor_sample_barcode)){
sample.mutations$tumor_sample_barcode <- factor(sample.mutations$tumor_sample_barcode, levels=unique(sample.mutations$tumor_sample_barcode))
}
plotGenes <- as.character(result.df$Hugo_Symbol[1:min(topGenes,nrow(result.df))])
sample.mutations <- sample.mutations[sample.mutations$hugo_symbol %in% plotGenes, ]
if(!silent){
sample.mutations <- sample.mutations[sample.mutations$variant_classification != 'Silent',]
}
if(!indels){
sample.mutations <- sample.mutations[! sample.mutations$variant_type %in% c('DEL', 'INS' ),]
}
if (order == 'CCF_sum'){
ccf.sums <- aggregate(ccf ~ hugo_symbol, data=sample.mutations, sum)
ccf.sums <- ccf.sums[order(ccf.sums$ccf, decreasing=F),]
sample.mutations$hugo_symbol <- factor(sample.mutations$hugo_symbol, levels=ccf.sums$hugo_symbol )
} else {
sample.mutations$hugo_symbol <- factor(sample.mutations$hugo_symbol, levels=names( sort(table(sample.mutations$hugo_symbol), decreasing=F )) )
}
if((is.numeric(color) | is.integer(color)) & color <= ncol(sample.mutations) ){
col.var <- colnames(sample.mutations)[color]
} else if (is.character(color) & color %in% colnames(sample.mutations)){
col.var <- color
} else {
stop('color need to be or numeric indicator of column in sample.mutations data frame or charater with column name (usually CCF column).')
}
formula<- paste(col.var,' ~ tumor_sample_barcode + hugo_symbol', sep='', collapse='')
sample.mutations <- aggregate(as.formula(formula), data = sample.mutations, function (x) min(sum(x),groupMax))
sample.mutations <- sample.mutations[order( sample.mutations$hugo_symbol, sample.mutations[,col.var], decreasing=T),]
if (allSamples) {
order.samples <- unique(c(as.character(sample.mutations$tumor_sample_barcode), levels(sample.mutations$tumor_sample_barcode)))
sample.mutations$tumor_sample_barcode <- factor( sample.mutations$tumor_sample_barcode, levels= order.samples)
} else {
sample.mutations$tumor_sample_barcode <- factor( sample.mutations$tumor_sample_barcode, levels=unique(as.character(sample.mutations$tumor_sample_barcode)))
}
p <- ggplot(sample.mutations, aes_string(x='tumor_sample_barcode', y='hugo_symbol', fill=col.var))
p <- p + geom_tile() + scale_fill_gradient(low="white", high="red") +
theme_bw() + theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust = 1, size=4)) +
scale_x_discrete(name='Samples',drop=FALSE) + ylab('Genes')
return(p)
}
hanasusak/cDriver documentation built on May 17, 2019, 2:27 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions plotSamplesMut plotMutChange boxplotCCF.mutations boxplotCCF.patients plotStaircase
Documented in boxplotCCF.mutations boxplotCCF.patients plotMutChange plotSamplesMut plotStaircase
######################################################################################
# CRG
# Hana SUSAK
#------------------------------------------------------------------------------------
# plotc for cDriver
######################################################################################
#' Plot mutations per sample distribution.
#' @description
#' \code{plot.samplesMut} Function to plot samples mutation counts.
#' @param sample.mutations data frame in MAF like format.
#' Columns (with exactly same names) which \code{sample.mutations} should have are:
#' \itemize{
#' \item Variant_Classification column specifed by MAF format, used to distinguish between silent and nonsilent SNVs
#' \item Hugo_Symbol column specifed by MAF format, which reports gene for each SNV.
#' \item Tumor_Sample_Barcode column specifed by MAF format, reporting for each SNV in wich patient was found.
#' \item Variant_Type columns pecifed by MAF format; is mutation SNV or InDel
#' }
#' @param indels a boolean value indicating should indels be counted. By default it is True.
#' @param silent a boolean value indicating should silent mutations be counted. By default it is True.
#' @param fill a boolean value indicating should plot only represent proportion between 3 types of mutations, not counts. By default it is False.
#' @param Tumor_Sample_Barcode (optional) integer/numeric value indicating column in \code{sample.mutations} which have sample ids for SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Hugo_Symbol (optional) integer/numeric value indicating column in \code{sample.mutations} having gene names for reported SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Classification (optional) integer/numeric value indicating column in \code{sample.mutations} which contain classification for SNV (Silent or not).
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Type (optional) integer/numeric value indicating column in \code{sample.mutations} which contains indormation if mutations is SNV or InDel .
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @return ggplot2 object
#' @examples
#' \donttest{
#' # plot sample's mutations , all of them
#' plotSamplesMut(sample.genes.mutect)
#' # plot proportion of silent and nonsilent, without indels
#' plotSamplesMut(sample.genes.mutect, indels=FALSE, fill=TRUE)
#' }
#' @export
plotSamplesMut <- function(sample.mutations, indels=TRUE, silent=TRUE, fill=FALSE, Tumor_Sample_Barcode=NULL, Hugo_Symbol=NULL, Variant_Classification=NULL, Variant_Type=NULL ) {
if (is.atomic(sample.mutations)) {
sample.mutations <- data.frame(x = sample.mutations)
}
if (!is.null(Variant_Classification)){
sample.mutations <- assign.columns(sample.mutations, Variant_Classification, "Variant_Classification")
}
if (!is.null(Variant_Type)){
sample.mutations <- assign.columns(sample.mutations, Variant_Type, "Variant_Type")
}
if (!is.null(Hugo_Symbol)){
sample.mutations <- assign.columns(sample.mutations, Hugo_Symbol, "Hugo_Symbol")
}
if (!is.null(Tumor_Sample_Barcode)){
sample.mutations <- assign.columns(sample.mutations, Tumor_Sample_Barcode, "Tumor_Sample_Barcode")
}
if(!is.factor(sample.mutations$Tumor_Sample_Barcode)){
sample.mutations$Tumor_Sample_Barcode <- factor(sample.mutations$Tumor_Sample_Barcode, levels=unique(sample.mutations$Tumor_Sample_Barcode))
}
##########
# make it not sensitive to lower/upper case in column names
original.col.names <- colnames(sample.mutations)
num.col <- ncol(sample.mutations)
colnames(sample.mutations) <- tolower(colnames(sample.mutations))
# take only exonic
suppressWarnings( sample.mutations <- exonic.only(sample.mutations))
if (indels){
#indel
sample.gene.indels <- sample.mutations[sample.mutations$variant_type %in% c('INS', 'DEL' ),]
sample.gene.indels$typeMut <- 'Indel'
ord.ind <- names(sort(table(sample.gene.indels$tumor_sample_barcode),decreasing=T))
} else {
sample.gene.indels <- c()
}
if (silent){
#silent
sample.gene.silent <- sample.mutations[sample.mutations$variant_type %in% c('SNP', 'DNP','TNP' ) & sample.mutations$variant_classification == "Silent",]
sample.gene.silent$typeMut <- 'Silent'
if (indels){
sample.gene.silent$tumor_sample_barcode <- factor(sample.gene.silent$tumor_sample_barcode, levels=ord.ind)
}
ord.sil <- names(sort(table(sample.gene.silent$tumor_sample_barcode),decreasing=F))
} else {
sample.gene.silent <- c()
}
#nonsilent
sample.gene.non.silent <- sample.mutations[sample.mutations$variant_type %in% c('SNP', 'DNP','TNP' ) &
sample.mutations$variant_classification %in% c('Frame_Shift_Del', 'Frame_Shift_Ins', 'In_Frame_Del',
'Missense_Mutation', 'Nonsense_Mutation', 'Splice_Site',
'Translation_Start_Site', 'Nonstop_Mutation'),]
sample.gene.non.silent$typeMut <- 'NonSilent'
#all
sample.gene.all <- rbind(sample.gene.non.silent, sample.gene.silent, sample.gene.indels)
if (silent){
sample.gene.all$tumor_sample_barcode <- factor(sample.gene.all$tumor_sample_barcode, levels=ord.sil)
} else if (indels) {
sample.gene.all$tumor_sample_barcode <- factor(sample.gene.all$tumor_sample_barcode, levels=ord.ind)
}
order <- names(sort(table(sample.gene.all$tumor_sample_barcode),decreasing=T))
sample.gene.all$tumor_sample_barcode <- factor(sample.gene.all$tumor_sample_barcode, levels=order)
max.y <- max(table(sample.gene.all$tumor_sample_barcode))
if (fill) {
p1 <- ggplot(data=sample.gene.all, aes_string('tumor_sample_barcode', fill='typeMut')) + geom_bar(position='fill') +
scale_y_continuous( name='Proportions', expand = c(0, 0))
} else {
p1 <- ggplot(data=sample.gene.all, aes_string('tumor_sample_barcode', fill='typeMut')) + geom_bar(position='stack') +
scale_y_continuous(name='Count', breaks=round(seq(0, max.y, length.out=10)), expand = c(0, 0))
}
p1 <- p1 + scale_x_discrete( expand = c(0, 0), drop=FALSE) +
theme_bw(base_size = 10) +
scale_fill_manual(values=c( "orange4",'orange1',"steelblue3")) +
theme( axis.text.x = element_text(size = 10 *0.5, angle = 90, hjust = 0.8, vjust=0.5, colour = "grey50")) +
theme(axis.title.x=element_text(angle=0, vjust=+0.5, size = 10 )) +
guides(fill = guide_legend(title = "Mutation type"))
#scale_fill_discrete(guide = guide_legend(title = "Mutation type"))
#theme( axis.text.x =element_blank()) +
#theme(axis.title.x=element_blank())
#return original names
colnames(sample.mutations)[1:num.col] <- original.col.names
p1 <- p1 + xlab('Sample ID')
p1
}
#' Plot mutations pattern frequncies.
#' @description
#' \code{plotMutChange} Function to plot frequency of mutation changes.
#' @param sample.mutations data frame in MAF like format.
#' Columns (with exactly same names) which \code{sample.mutations} should have are:
#' \itemize{
#' \item Tumor_Sample_Barcode column specifed by MAF format, reporting for each SNV in wich patient was found.
#' \item Variant_Type columns pecifed by MAF format; is mutation SNV or InDel
#' \item Variant_Classification column specifed by MAF format, used to distinguish between silent and nonsilent SNVs
#' \item Reference_Allele column specifed by MAF format, represents referent allele
#' \item Tumor_Seq_Allele2 column specifed by MAF format, represents alternative allele
#' }
#' @param silent a boolean value indicating should silent mutations be counted. By default it is True.
#' @param fill a boolean value indicating should plot only represent proportion between 3 types of mutations, not counts. By default it is True
#' @param Tumor_Sample_Barcode (optional) integer/numeric value indicating column in \code{sample.mutations} which have sample ids for SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Type (optional) integer/numeric value indicating column in \code{sample.mutations} which contains indormation if mutations is SNV or InDel .
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Classification (optional) integer/numeric value indicating column in \code{sample.mutations} which contain classification for SNV (Silent or not).
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Reference_Allele (optional) integer/numeric value indicating column in \code{sample.mutations} which contain reference alleles.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Tumor_Seq_Allele2 (optional) integer/numeric value indicating column in \code{sample.mutations} which contain alternative alleles.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @return ggplot2 object
#' @examples
#' \donttest{
#' # plot SNVs changes patterns
#' plotMutChange(sample.genes.mutect)
#' # plot SNVs changes patterns only for non silent mutaions
#' plotMutChange(sample.genes.mutect, silent=FALSE, fill=FALSE)
#' }
#' @export
plotMutChange <- function(sample.mutations, silent=TRUE, fill=TRUE, Tumor_Sample_Barcode=NULL, Variant_Type=NULL, Variant_Classification=NULL, Reference_Allele=NULL, Tumor_Seq_Allele2=NULL ) {
if (is.atomic(sample.mutations)) {
sample.mutations <- data.frame(x = sample.mutations)
}
if (!is.null(Tumor_Sample_Barcode)){
sample.mutations <- assign.columns(sample.mutations, Tumor_Sample_Barcode, "Tumor_Sample_Barcode")
}
if (!is.null(Variant_Type)){
sample.mutations <- assign.columns(sample.mutations, Variant_Type, "Variant_Type")
}
if (!is.null(Variant_Classification)){
sample.mutations <- assign.columns(sample.mutations, Variant_Classification, "Variant_Classification")
}
if (!is.null(Reference_Allele)){
sample.mutations <- assign.columns(sample.mutations, Reference_Allele, "Reference_Allele")
}
if (!is.null(Tumor_Seq_Allele2)){
sample.mutations <- assign.columns(sample.mutations, Tumor_Seq_Allele2, "Tumor_Seq_Allele2")
}
##########
# make it not sensitive to lower/upper case in column names
original.col.names <- colnames(sample.mutations)
num.col <- ncol(sample.mutations)
colnames(sample.mutations) <- tolower(colnames(sample.mutations))
# take only exonic
suppressWarnings( sample.mutations <- exonic.only(sample.mutations) )
if(!is.factor(sample.mutations$tumor_sample_barcode)){
sample.mutations$tumor_sample_barcode <- factor(sample.mutations$tumor_sample_barcode, levels=unique(sample.mutations$tumor_sample_barcode))
}
if(!silent){
sample.mutations <- sample.mutations[sample.mutations$variant_classification != 'Silent',]
}
sample.mutations <- sample.mutations[sample.mutations$variant_type %in% c('SNP', 'DNP','TNP' ),]
sample.mutations$Change <- base::apply(sample.mutations, 1, function(x)paste(x['reference_allele'], x['tumor_seq_allele2'], sep=';') )
sample.mutations[sample.mutations$Change == 'G;T','Change'] <- 'C;A'
sample.mutations[sample.mutations$Change == 'G;C','Change'] <- 'C;G'
sample.mutations[sample.mutations$Change == 'G;A','Change'] <- 'C;T'
sample.mutations[sample.mutations$Change == 'A;G','Change'] <- 'T;C'
sample.mutations[sample.mutations$Change == 'A;T','Change'] <- 'T;A'
sample.mutations[sample.mutations$Change == 'A;C','Change'] <- 'T;G'
sample.mutations$Change <- unlist(lapply(sample.mutations$Change, function(x) gsub(';','->',x)))
sample.mutations$Change <- factor((sample.mutations$Change), levels=c( 'C->T', 'T->C', 'C->A', 'C->G', 'T->A','T->G'))
sample.mutations$tumor_sample_barcode <- factor(sample.mutations$tumor_sample_barcode, levels=names(sort(table(sample.mutations$tumor_sample_barcode), decreasing=T) ))
p2 <- ggplot(sample.mutations, aes_string('tumor_sample_barcode', fill='Change')) + geom_bar(position=position_fill()) + theme_bw(base_size = 10) +
scale_y_continuous(name='Proportion',expand=c(0,0)) +
scale_x_discrete(name='Sample ID',drop=FALSE) + scale_fill_brewer(palette="Set2") +
theme( axis.text.x = element_text(size = 10 *0.5, angle = 90, hjust = 0.8, vjust=0.5, colour = "grey50")) +
theme(axis.title.x=element_text(angle=0, vjust=+0.5, size = 10 ))
#return original names
colnames(sample.mutations)[1:num.col] <- original.col.names
p2
}
#' Plot boxplot of mutations for top genes.
#' @description
#' \code{boxplotCCF.mutations} Function to plot CCF's of all mutations for top genes.
#' @param sample.mutations data frame in MAF like format.
#' Columns (with exactly same names) which \code{sample.mutations} should have are:
#' \itemize{
#' \item Tumor_Sample_Barcode column specifed by MAF format, reporting for each SNV in wich patient was found.
#' \item Hugo_Symbol column specifed by MAF format, which reports gene for each SNV.
#' \item CCF numeric column produce by \code{CCF} function.
#' \item Variant_Classification column specifed by MAF format, used to distinguish between silent and nonsilent SNVs
#' \item Variant_Type columns pecifed by MAF format; is mutation SNV or InDel
#' }
#' @param result.df a data frame with Hugo_Symbol column. Genes in this column should be ordered by importance.
#' @param topGenes a numeric/integer value; How meny top genes from results will be ploted.
#' @param silent a boolean value indicating should silent mutations be counted. By default it is True.
#' @param indels a boolean value indicating should indels be counted. By default it is True.
#' @param Tumor_Sample_Barcode (optional) integer/numeric value indicating column in \code{sample.mutations} which have sample ids for SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Hugo_Symbol (optional) integer/numeric value indicating column in \code{sample.mutations} having gene names for reported SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param CCF (optional) integer/numeric value indicating column in \code{sample.mutations} which have cancer cell fraction information for SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Classification (optional) integer/numeric value indicating column in \code{sample.mutations} which contain classification for SNV (Silent or not).
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Type (optional) integer/numeric value indicating column in \code{sample.mutations} which contains indormation if mutations is SNV or InDel .
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param color a numeric or character value indicating column to color the samples.
#' @param shape a numeric or character value indicating column to use as shape for the samples.
#' @return ggplot2 object
#' @examples
#' \donttest{
#' #get CCF column
#' sample.genes.mutect <- CCF(sample.genes.mutect)
#' # get background
#' bcgr.prob <- bcgr.combine(sample.genes.mutect, length.genes$Hugo_Symbol, length.genes$Coverd_len)
#' # get ranking
#' df1 <- bayes.risk(sample.genes.mutect, bcgr.prob, prior.sick = 0.00007)
#' df2 <- bayes.driver(sample.genes.mutect, bcgr.prob, prior.driver = 0.001)
#' df.final <- combine.ranking(list(df1, df2), min.mut = 2 )
#' # plot boxplot of CCF for top genes
#' boxplotCCF(sample.mutations=sample.genes.mutect, result.df=df.final)
#' }
#' @export
boxplotCCF.mutations <- function(sample.mutations, result.df, topGenes=20, silent=FALSE, indels=TRUE,
Tumor_Sample_Barcode=NULL, Hugo_Symbol=NULL, CCF=NULL, Variant_Classification=NULL, Variant_Type = NULL, color=NULL, shape=NULL ) {
if (is.atomic(sample.mutations)) {
sample.mutations <- data.frame(x = sample.mutations)
}
if (!is.null(Tumor_Sample_Barcode)){
sample.mutations <- assign.columns(sample.mutations, Tumor_Sample_Barcode, "Tumor_Sample_Barcode")
}
if (!is.null(Hugo_Symbol)){
sample.mutations <- assign.columns(sample.mutations, Hugo_Symbol, "Hugo_Symbol")
}
if (!is.null(CCF)){
sample.mutations <- assign.columns(sample.mutations, CCF, "CCF")
}
if (!is.null(Variant_Classification)){
sample.mutations <- assign.columns(sample.mutations, Variant_Classification, "Variant_Classification")
}
if (!is.null(Variant_Type)){
sample.mutations <- assign.columns(sample.mutations, Variant_Type, "Variant_Type")
}
##########
# make it not sensitive to lower/upper case in column names
original.col.names <- colnames(sample.mutations)
num.col <- ncol(sample.mutations)
colnames(sample.mutations) <- tolower(colnames(sample.mutations))
if (!is.null(color) & is.character(color)){
color <- tolower(color)
}
if (!is.null(shape) & is.character(shape)){
shape <- tolower(shape)
}
# take only exonic
suppressWarnings( sample.mutations <- exonic.only(sample.mutations) )
if(!'Hugo_Symbol' %in% colnames(result.df)){
stop('There must be Hugo_Symbol column in result.df data frame!')
}
if(!is.factor(sample.mutations$hugo_symbol)){
sample.mutations$hugo_symbol <- factor(sample.mutations$hugo_symbol, levels=unique(sample.mutations$hugo_symbol))
}
plotGenes <- as.character(result.df$Hugo_Symbol[1:min(topGenes,nrow(result.df))])
sample.mutations <- sample.mutations[sample.mutations$hugo_symbol %in% plotGenes, ]
if(!silent){
sample.mutations <- sample.mutations[sample.mutations$variant_classification != 'Silent',]
}
if(!indels){
sample.mutations <- sample.mutations[! sample.mutations$variant_type %in% c('DEL', 'INS' ),]
}
order.gene <- as.table(by(sample.mutations[,'ccf'],sample.mutations[,'hugo_symbol'],median))
sample.mutations$hugo_symbol <- factor(sample.mutations$hugo_symbol , levels=names(sort(order.gene, decreasing=T)))
if (is.null(color) & is.null(shape)){
p <- ggplot(sample.mutations, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(size=2.5, alpha=0.65, position = position_jitter(width = .3, height=0))
}
#just color
if (!is.null(color) & is.null(shape)){
if((is.numeric(color) | is.integer(color)) & color <= ncol(sample.mutations) ){
col.var <- colnames(sample.mutations)[color]
} else if (is.character(color) & color %in% colnames(sample.mutations)){
col.var <- color
} else {
stop('color need to be or numeric indicator of column in sample.mutations data frame or charater with column name.')
}
p <- ggplot(sample.mutations, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(aes_string(color=col.var),size=2.5, alpha=0.65, position = position_jitter(width = .3, height=0))
}
# just shape
if (!is.null(shape) & is.null(color)){
if((is.numeric(shape) | is.integer(shape)) & shape <= ncol(sample.mutations) ){
sh.var <- colnames(sample.mutations)[shape]
} else if (is.character(shape) & shape %in% colnames(sample.mutations)){
sh.var <- shape
} else {
stop('shape need to be or numeric indicator of column in sample.mutations data frame or charater with column name.')
}
n.sh <- length(unique(sample.mutations[,sh.var]))
if (n.sh <= 11 ) {
sh.seq <- 15:(n.sh+14)
} else {
sh.seq <- c(15:25,1:(n.sh-11))
}
p <- ggplot(sample.mutations, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(aes_string(shape=sh.var),size=2.5, alpha=0.65, position = position_jitter(width = .3, height=0)) +
scale_shape_manual(values=sample(sh.seq, replace=F) )
}
if (!is.null(shape) & !is.null(color)){
if((is.numeric(shape) | is.integer(shape)) & shape <= ncol(sample.mutations) ){
sh.var <- colnames(sample.mutations)[shape]
} else if (is.character(shape) & shape %in% colnames(sample.mutations)){
sh.var <- shape
} else {
stop('shape need to be or numeric indicator of column in sample.mutations data frame or charater with column name.')
}
n.sh <- length(unique(sample.mutations[,sh.var]))
if (n.sh <= 11 ) {
sh.seq <- 15:(n.sh+14)
} else {
sh.seq <- c(15:25,1:(n.sh-11))
}
if((is.numeric(color) | is.integer(color)) & color <= ncol(sample.mutations) ){
col.var <- colnames(sample.mutations)[color]
} else if (is.character(color) & color %in% colnames(sample.mutations)){
col.var <- color
} else {
stop('color need to be or numeric indicator of column in sample.mutations data frame or charater with column name.')
}
p <- ggplot(sample.mutations, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(aes_string(color=(col.var), shape=(sh.var)),size=2.5, alpha=0.65, position = position_jitter(width = .15, height=0)) +#, drop=FALSE
scale_shape_manual(values=sample(sh.seq, replace=F))
}
#return original names
colnames(sample.mutations)[1:num.col] <- original.col.names
return(p)
}
#' Plot boxplot of patients mutations CCF's for top genes.
#' @description
#' \code{boxplotCCF.patients} Function to plot CCF's of aggregated patient-gene mutatuons, for top genes.
#' @param sample.mutations data frame in MAF like format.
#' Columns (with exactly same names) which \code{sample.mutations} should have are:
#' \itemize{
#' \item Tumor_Sample_Barcode column specifed by MAF format, reporting for each SNV in wich patient was found.
#' \item Hugo_Symbol column specifed by MAF format, which reports gene for each SNV.
#' \item CCF numeric column produce by \code{CCF} function.
#' \item Variant_Classification column specifed by MAF format, used to distinguish between silent and nonsilent SNVs
#' \item Variant_Type columns pecifed by MAF format; is mutation SNV or InDel
#' }
#' @param result.df a data frame with Hugo_Symbol column. Genes in this column should be ordered by importance.
#' @param topGenes a numeric/integer value; How meny top genes from results will be ploted.
#' @param silent a boolean value indicating should silent mutations be counted. By default it is True.
#' @param indels a boolean value indicating should indels be counted. By default it is True.
#' @param mode a charechter value indicationg how to solve when in one gene-sample pair there are multiple mutations. Options are SUM, MAX and ADVANCE
#' @param epsilon a numeric value. If mode is ADVANCE, epsilone value will be threshold for CCF difference to decide if they are in same or different clone.
#' @param sample.gene.lim a numeric value specifying upper limit when summing is perfomed for gene-sample pair
#' @param Tumor_Sample_Barcode (optional) integer/numeric value indicating column in \code{sample.mutations} which have sample ids for SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Hugo_Symbol (optional) integer/numeric value indicating column in \code{sample.mutations} having gene names for reported SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param CCF (optional) integer/numeric value indicating column in \code{sample.mutations} which have cancer cell fraction information for SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Classification (optional) integer/numeric value indicating column in \code{sample.mutations} which contain classification for SNV (Silent or not).
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Type (optional) integer/numeric value indicating column in \code{sample.mutations} which contains indormation if mutations is SNV or InDel .
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param color (optional) charachter value indicating column in \code{sample.mutations} to used as color dots.
#' Default is NULL value and grey dots gonna be ploted. This variable should be unique by patients.
#' @param shape (optional) charachter value indicating column in \code{sample.mutations} to used as shape for dots .
#' Default is NULL value and grey dots gonna be ploted. This variable should be unique by patients.
#' @return ggplot2 object
#' @examples
#' \donttest{
#' #get CCF column
#' sample.genes.mutect <- CCF(sample.genes.mutect)
#' # get background
#' bcgr.prob <- bcgr.combine(sample.genes.mutect)
#' # get ranking
#' df1 <- bayes.risk(sample.genes.mutect, bcgr.prob, prior.sick = 0.00007)
#' df2 <- bayes.driver(sample.genes.mutect, bcgr.prob, prior.driver = 0.001)
#' df.final <- combine.ranking(list(df1, df2), min.mut = 2 )
#' # plot boxplot of CCF for top genes
#' boxplotCCF(sample.mutations=sample.genes.mutect, result.df=df.final)
#' }
#' @export
boxplotCCF.patients <- function(sample.mutations, result.df, topGenes=20, silent=FALSE, indels=TRUE, mode='MAX', epsilon=0.05, sample.gene.lim=1,
Tumor_Sample_Barcode=NULL, Hugo_Symbol=NULL, CCF=NULL, Variant_Classification=NULL, Variant_Type = NULL, color=NULL, shape=NULL) {
if (is.atomic(sample.mutations)) {
sample.mutations <- data.frame(x = sample.mutations)
}
mode <- toupper(mode)
if ( !mode %in% c('SUM', 'MAX', 'ADVANCE')) {
stop("mode mast be or SUM or MAX or ADVACE! ", call. = FALSE)
}
if (!is.null(Tumor_Sample_Barcode)){
sample.mutations <- assign.columns(sample.mutations, Tumor_Sample_Barcode, "Tumor_Sample_Barcode")
}
if (!is.null(Hugo_Symbol)){
sample.mutations <- assign.columns(sample.mutations, Hugo_Symbol, "Hugo_Symbol")
}
if (!is.null(CCF)){
sample.mutations <- assign.columns(sample.mutations, CCF, "CCF")
}
if (!is.null(Variant_Classification)){
sample.mutations <- assign.columns(sample.mutations, Variant_Classification, "Variant_Classification")
}
if (!is.null(Variant_Type)){
sample.mutations <- assign.columns(sample.mutations, Variant_Type, "Variant_Type")
}
##########
# make it not sensitive to lower/upper case in column names
original.col.names <- colnames(sample.mutations)
num.col <- ncol(sample.mutations)
colnames(sample.mutations) <- tolower(colnames(sample.mutations))
# take only exonic
suppressWarnings( sample.mutations <- exonic.only(sample.mutations ))
if ( (!is.null(color)) & is.character(color) ) {
if (tolower(color) %in% colnames(sample.mutations)){
color <- tolower(color)
} else {
stop('color variable must be one of the column names in sample.mutations data frame')
}
} else if (!is.null(color)){
stop('color variable must be character type')
}
if ( (!is.null(shape)) & is.character(shape) ) {
if (tolower(shape) %in% colnames(sample.mutations)){
shape <- tolower(shape)
} else {
stop('shape variable must be one of the column names in sample.mutations data frame')
} } else if (!is.null(shape)){
stop('shape variable must be character type')
}
if(!'Hugo_Symbol' %in% colnames(result.df)){
stop('There must be Hugo_Symbol column in result.df data frame!')
}
if(!is.factor(sample.mutations$hugo_symbol)){
sample.mutations$hugo_symbol <- factor(sample.mutations$hugo_symbol, levels=unique(sample.mutations$hugo_symbol))
}
advance.ccf.calc <- function(x, epsilon){
x <- sort(x, decreasing=T)
sum.ccf <- x[1]
if (length(x) >1){
for (i in 2:length(x)) {
if ((x[i-1] - x[i]) > epsilon){
sum.ccf <- sum.ccf + x[i]
}
}
}
return(sum.ccf)
}
plotGenes <- as.character(result.df$Hugo_Symbol[1:min(topGenes,nrow(result.df))])
sample.mutations <- sample.mutations[sample.mutations$hugo_symbol %in% plotGenes, ]
if(!silent){
sample.mutations <- sample.mutations[sample.mutations$variant_classification != 'Silent',]
}
if(!indels){
sample.mutations <- sample.mutations[! sample.mutations$variant_type %in% c('DEL', 'INS' ),]
}
# AGGREGATE SNPs which are in same sample-gene pair
if (mode == 'SUM' ){
# if we sum values of column CCF, for same sample-gene pairs
if (is.null(sample.gene.lim)){
sample.gene.lim <- Inf
}
f <- paste( "ccf ~ hugo_symbol + tumor_sample_barcode", collapse="")
sample.mutations2 <- aggregate(as.formula(f) , data=sample.mutations, function(x) min(sum(x),sample.gene.lim))
if (!is.null(color)){
df.temp <- sample.mutations[,c(color,'tumor_sample_barcode')]
df.temp <- df.temp[!duplicated(df.temp), ]
first.n <- nrow(sample.mutations2)
sample.mutations2 <- merge(x=sample.mutations2, y=df.temp, by='tumor_sample_barcode',all.x=T)
second.n <- nrow(sample.mutations2)
if(first.n!=second.n){
warning('You choose to color by paramether which is not unique by patients!')
}
}
if (!is.null(shape)){
df.temp <- sample.mutations[,c(shape,'tumor_sample_barcode')]
df.temp <- df.temp[!duplicated(df.temp), ]
first.n <- nrow(sample.mutations2)
sample.mutations2 <- merge(x=sample.mutations2, y=df.temp, by='tumor_sample_barcode',all.x=T)
second.n <- nrow(sample.mutations2)
if(first.n!=second.n){
warning('You choose to shape by paramether which is not unique by patients!')
}
}
} else if (mode == 'MAX' ) {
# if we take max value of column CCF, for same sample-gene pairs
if (is.null(sample.gene.lim)){
sample.gene.lim <- Inf
}
f <- paste("ccf ~ hugo_symbol + tumor_sample_barcode", collapse="")
sample.mutations2 <- aggregate(as.formula(f) , data=sample.mutations, function(x) min(max(x, na.rm=T),sample.gene.lim))
if (!is.null(color)){
df.temp <- sample.mutations[,c(color,'tumor_sample_barcode')]
df.temp <- df.temp[!duplicated(df.temp), ]
first.n <- nrow(sample.mutations2)
sample.mutations2 <- merge(x=sample.mutations2, y=df.temp, by='tumor_sample_barcode',all.x=T)
second.n <- nrow(sample.mutations2)
if(first.n!=second.n){
warning('You choose to color by paramether which is not unique by patients!')
}
}
if (!is.null(shape)){
df.temp <- sample.mutations[,c(shape,'tumor_sample_barcode')]
df.temp <- df.temp[!duplicated(df.temp), ]
first.n <- nrow(sample.mutations2)
sample.mutations2 <- merge(x=sample.mutations2, y=df.temp, by='tumor_sample_barcode',all.x=T)
second.n <- nrow(sample.mutations2)
if(first.n!=second.n){
warning('You choose to shape by paramether which is not unique by patients!')
}
}
} else if (mode == 'ADVANCE' ){
# if we take max or sum value of column CCF, for same sample-gene pairs, depending if in same or different clone
if (is.null(sample.gene.lim)){
sample.gene.lim <- Inf
}
#sample.gene.dt <- sample.gene.dt[order(sample.gene.dt$Hugo_Symbol, sample.gene.dt$Tumor_Sample_Barcode, -sample.gene.dt$CCF),]
f <- paste( "ccf ~ hugo_symbol + tumor_sample_barcode", collapse="")
sample.mutations2 <- aggregate(as.formula(f) , data=sample.mutations, function(x) min(advance.ccf.calc(x, epsilon),sample.gene.lim))
if (!is.null(color)){
df.temp <- sample.mutations[,c(color,'tumor_sample_barcode')]
df.temp <- df.temp[!duplicated(df.temp), ]
first.n <- nrow(sample.mutations2)
sample.mutations2 <- merge(x=sample.mutations2, y=df.temp, by='tumor_sample_barcode',all.x=T)
second.n <- nrow(sample.mutations2)
if(first.n!=second.n){
warning('You choose to color by paramether which is not unique by patients!')
}
}
if (!is.null(shape)){
df.temp <- sample.mutations[,c(shape,'tumor_sample_barcode')]
df.temp <- df.temp[!duplicated(df.temp), ]
first.n <- nrow(sample.mutations2)
sample.mutations2 <- merge(x=sample.mutations2, y=df.temp, by='tumor_sample_barcode',all.x=T)
second.n <- nrow(sample.mutations2)
if(first.n!=second.n){
warning('You choose to shape by paramether which is not unique by patients!')
}
}
}
order.gene <- as.table(by(sample.mutations2[,'ccf'],sample.mutations2[,'hugo_symbol'],median))
sample.mutations2$hugo_symbol <- factor(sample.mutations2$hugo_symbol , levels=names(sort(order.gene, decreasing=T)))
if ( (!is.null(color)) & (!is.null(shape))){
p <- ggplot(sample.mutations2, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(aes_string(color=(color), shape=(shape)),size=3, alpha=0.75, position = position_jitter(width = .3, height=0))
} else if ( !is.null(color) ){
p <- ggplot(sample.mutations2, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(aes_string(color=(color)),size=3, alpha=0.75, position = position_jitter(width = .3, height=0))
} else if ( !is.null(shape) ){
p <- ggplot(sample.mutations2, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(aes_string(shape=(shape)),size=3, alpha=0.75, position = position_jitter(width = .3, height=0))
} else {
p <- ggplot(sample.mutations2, aes_string('hugo_symbol','ccf', label='tumor_sample_barcode')) + geom_boxplot(outlier.size=0) + theme_bw() +
xlab('Top Genes') + theme(axis.text.x = element_text(angle = -30, hjust = 0, vjust=1)) + ylab('Cancer Cell Fraction') +
geom_jitter(size=3, alpha=0.75, position = position_jitter(width = .3, height=0))
}
#return original names
colnames(sample.mutations)[1:num.col] <- original.col.names
return(p)
}
#' Plot heatmap for genes-sample pairs .
#' @description
#' \code{plotStaircase} Function to plot heatmap for genes-sample pairs orderd by genes which explain most of the patients .
#' @param sample.mutations data frame in MAF like format.
#' Columns (with exactly same names) which \code{sample.mutations} should have are:
#' \itemize{
#' \item Tumor_Sample_Barcode column specifed by MAF format, reporting for each SNV in wich patient was found.
#' \item Hugo_Symbol column specifed by MAF format, which reports gene for each SNV.
#' \item Variant_Classification column specifed by MAF format, used to distinguish between silent and nonsilent SNVs
#' \item Variant_Type columns pecifed by MAF format; is mutation SNV or InDel
#' }
#' @param result.df a data frame with Hugo_Symbol column. Genes in this column should be ordered by importance.
#' @param topGenes a numeric/integer value; How meny top genes from results will be ploted.
#' @param silent a boolean value indicating should silent mutations be counted. By default it is True.
#' @param indels a boolean value indicating should indels be counted. By default it is True.
#' @param allSamples a boolean value indicating if all samples should be plotted.
#' @param Tumor_Sample_Barcode (optional) integer/numeric value indicating column in \code{sample.mutations} which have sample ids for SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Hugo_Symbol (optional) integer/numeric value indicating column in \code{sample.mutations} having gene names for reported SNVs/Indels.
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Classification (optional) integer/numeric value indicating column in \code{sample.mutations} which contain classification for SNV (Silent or not).
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param Variant_Type (optional) integer/numeric value indicating column in \code{sample.mutations} which contains indormation if mutations is SNV or InDel .
#' Default is NULL value (in this case \code{sample.mutations} should already have this column)
#' @param color a numeric or character value indicating column to color the samples.
#' @param groupMax a numeric value which will be taken as maximum value when grouping color column for samples and genes pairs.
#' @param order a character value which can be or \code{'frequency'} or \code{'CCF_sum'}.
#' @return ggplot2 object
#' @examples
#' \donttest{
#' #get CCF column
#' sample.genes.mutect <- CCF(sample.genes.mutect)
#' # get background
#' bcgr.prob <- bcgr.combine(sample.genes.mutect)
#' # get ranking
#' df1 <- bayes.risk(sample.genes.mutect, bcgr.prob, prior.sick = 0.00007)
#' df2 <- bayes.driver(sample.genes.mutect, bcgr.prob, prior.driver = 0.001)
#' df.final <- combine.ranking(list(df1, df2), min.mut = 2 )
#' # plot boxplot of CCF for top genes
#' plotStaircase(sample.mutations=sample.genes.mutect, result.df=df.final)
#' }
#' @export
plotStaircase <- function(sample.mutations, result.df, topGenes=40, silent=FALSE, indels=TRUE, allSamples= FALSE,
Tumor_Sample_Barcode=NULL, Hugo_Symbol=NULL, Variant_Classification=NULL, Variant_Type = NULL,
color='CCF' , groupMax=1, order='frequency') {
if (is.atomic(sample.mutations)) {
sample.mutations <- data.frame(x = sample.mutations)
}
if (!is.null(Tumor_Sample_Barcode)){
sample.mutations <- assign.columns(sample.mutations, Tumor_Sample_Barcode, "Tumor_Sample_Barcode")
}
if (!is.null(Hugo_Symbol)){
sample.mutations <- assign.columns(sample.mutations, Hugo_Symbol, "Hugo_Symbol")
}
if (!is.null(Variant_Classification)){
sample.mutations <- assign.columns(sample.mutations, Variant_Classification, "Variant_Classification")
}
if (!is.null(Variant_Type)){
sample.mutations <- assign.columns(sample.mutations, Variant_Type, "Variant_Type")
}
##########
# make it not sensitive to lower/upper case in column names
original.col.names <- colnames(sample.mutations)
num.col <- ncol(sample.mutations)
colnames(sample.mutations) <- tolower(colnames(sample.mutations))
# take only exonic
suppressWarnings( sample.mutations <- exonic.only(sample.mutations) )
if (!is.null(color) & is.character(color)){
color <- tolower(color)
}
if(!'Hugo_Symbol' %in% colnames(result.df)){
stop('There must be Hugo_Symbol column in result.df data frame!')
}
if(!is.factor(sample.mutations$tumor_sample_barcode)){
sample.mutations$tumor_sample_barcode <- factor(sample.mutations$tumor_sample_barcode, levels=unique(sample.mutations$tumor_sample_barcode))
}
plotGenes <- as.character(result.df$Hugo_Symbol[1:min(topGenes,nrow(result.df))])
sample.mutations <- sample.mutations[sample.mutations$hugo_symbol %in% plotGenes, ]
if(!silent){
sample.mutations <- sample.mutations[sample.mutations$variant_classification != 'Silent',]
}
if(!indels){
sample.mutations <- sample.mutations[! sample.mutations$variant_type %in% c('DEL', 'INS' ),]
}
if (order == 'CCF_sum'){
ccf.sums <- aggregate(ccf ~ hugo_symbol, data=sample.mutations, sum)
ccf.sums <- ccf.sums[order(ccf.sums$ccf, decreasing=F),]
sample.mutations$hugo_symbol <- factor(sample.mutations$hugo_symbol, levels=ccf.sums$hugo_symbol )
} else {
sample.mutations$hugo_symbol <- factor(sample.mutations$hugo_symbol, levels=names( sort(table(sample.mutations$hugo_symbol), decreasing=F )) )
}
if((is.numeric(color) | is.integer(color)) & color <= ncol(sample.mutations) ){
col.var <- colnames(sample.mutations)[color]
} else if (is.character(color) & color %in% colnames(sample.mutations)){
col.var <- color
} else {
stop('color need to be or numeric indicator of column in sample.mutations data frame or charater with column name (usually CCF column).')
}
formula<- paste(col.var,' ~ tumor_sample_barcode + hugo_symbol', sep='', collapse='')
sample.mutations <- aggregate(as.formula(formula), data = sample.mutations, function (x) min(sum(x),groupMax))
sample.mutations <- sample.mutations[order( sample.mutations$hugo_symbol, sample.mutations[,col.var], decreasing=T),]
if (allSamples) {
order.samples <- unique(c(as.character(sample.mutations$tumor_sample_barcode), levels(sample.mutations$tumor_sample_barcode)))
sample.mutations$tumor_sample_barcode <- factor( sample.mutations$tumor_sample_barcode, levels= order.samples)
} else {
sample.mutations$tumor_sample_barcode <- factor( sample.mutations$tumor_sample_barcode, levels=unique(as.character(sample.mutations$tumor_sample_barcode)))
}
p <- ggplot(sample.mutations, aes_string(x='tumor_sample_barcode', y='hugo_symbol', fill=col.var))
p <- p + geom_tile() + scale_fill_gradient(low="white", high="red") +
theme_bw() + theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust = 1, size=4)) +
scale_x_discrete(name='Samples',drop=FALSE) + ylab('Genes')
return(p)
}
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.