comparison shm_csr.r @ 5:012a738edf5a draft

Uploaded

4:477e95b098fd

		transition2 = merge(transition2, base.order, by.x="id", by.y="base")

		transition2 = merge(transition2, base.order, by.x="variable", by.y="base")

		transition2[is.na(transition2$value),]$value = 0
		
		print(transition2)

		if(any(transition2$value != 0)){ #having rows of data but a transition table filled with 0 is bad
			print("Plotting stacked transition")
			png(filename=paste("transitions_stacked_", name, ".png", sep=""))
			p = ggplot(transition2, aes(factor(reorder(id, order.x)), y=value, fill=factor(reorder(variable, order.y)))) + geom_bar(position="fill", stat="identity", colour="black") #stacked bar
			p = p + xlab("From base") + ylab("To base") + ggtitle("Mutations frequency from base to base") + guides(fill=guide_legend(title=NULL))
			p = p + theme(panel.background = element_rect(fill = "white", colour="black"), text = element_text(size=13, colour="black")) + scale_fill_manual(values=c("A" = "blue4", "G" = "lightblue1", "C" = "olivedrab3", "T" = "olivedrab4"))
			#p = p + scale_colour_manual(values=c("A" = "black", "G" = "black", "C" = "black", "T" = "black"))
			print(p)
			dev.off()

			print("Plotting heatmap transition")

	genesForPlot$label = paste(genesForPlot$Gene, "-", genesForPlot$Freq)

	pc = ggplot(genesForPlot, aes(x = factor(1), y=Freq, fill=Gene))
	pc = pc + geom_bar(width = 1, stat = "identity") + scale_fill_manual(labels=genesForPlot$label, values=c("IGA1" = "lightblue1", "IGA2" = "blue4"))
	pc = pc + coord_polar(theta="y") + scale_y_continuous(breaks=NULL)
	pc = pc + theme(panel.background = element_rect(fill = "white", colour="black"), text = element_text(size=13, colour="black"))
	pc = pc + xlab(" ") + ylab(" ") + ggtitle(paste("IGA subclasses", "( n =", sum(genesForPlot$Freq), ")"))
	write.table(genesForPlot, "IGA_pie.txt", sep="\t",quote=F,row.names=F,col.names=T)

	png(filename="IGA.png")
	print(pc)

	genesForPlot$label = paste(genesForPlot$Gene, "-", genesForPlot$Freq)

	pc = ggplot(genesForPlot, aes(x = factor(1), y=Freq, fill=Gene))
	pc = pc + geom_bar(width = 1, stat = "identity") + scale_fill_manual(labels=genesForPlot$label, values=c("IGG1" = "olivedrab3", "IGG2" = "red", "IGG3" = "gold", "IGG4" = "darkred"))
	pc = pc + coord_polar(theta="y") + scale_y_continuous(breaks=NULL)
	pc = pc + theme(panel.background = element_rect(fill = "white", colour="black"), text = element_text(size=13, colour="black"))
	pc = pc + xlab(" ") + ylab(" ") + ggtitle(paste("IGG subclasses", "( n =", sum(genesForPlot$Freq), ")"))
	write.table(genesForPlot, "IGG_pie.txt", sep="\t",quote=F,row.names=F,col.names=T)

	png(filename="IGG.png")
	print(pc)

dat.clss = rbind(dat, dat.clss)

p = ggplot(dat.clss, aes(best_match, percentage_mutations))
p = p + geom_point(aes(colour=best_match), position="jitter") + geom_boxplot(aes(middle=mean(percentage_mutations)), alpha=0.1, outlier.shape = NA)
p = p + xlab("Subclass") + ylab("Frequency") + ggtitle("Frequency scatter plot") + theme(panel.background = element_rect(fill = "white", colour="black"), text = element_text(size=13, colour="black"))
p = p + scale_fill_manual(values=c("IGA" = "blue4", "IGA1" = "lightblue1", "IGA2" = "blue4", "IGG" = "olivedrab3", "IGG1" = "olivedrab3", "IGG2" = "red", "IGG3" = "gold", "IGG4" = "darkred", "IGM" = "darkviolet", "all" = "blue4"))
p = p + scale_colour_manual(values=c("IGA" = "blue4", "IGA1" = "lightblue1", "IGA2" = "blue4", "IGG" = "olivedrab3", "IGG1" = "olivedrab3", "IGG2" = "red", "IGG3" = "gold", "IGG4" = "darkred", "IGM" = "darkviolet", "all" = "blue4"))

png(filename="scatter.png")
print(p)

frequency_bins_data = merge(frequency_bins_data, frequency_bins_sum, by="best_match_class")

frequency_bins_data$frequency = round(frequency_bins_data$frequency_count / frequency_bins_data$class_sum * 100, 2)

p = ggplot(frequency_bins_data, aes(frequency_bins, frequency))
p = p + geom_bar(aes(fill=best_match_class), stat="identity", position="dodge") + theme(panel.background = element_rect(fill = "white", colour="black"), text = element_text(size=13, colour="black"))
p = p + xlab("Frequency ranges") + ylab("Frequency") + ggtitle("Mutation Frequencies by class") + scale_fill_manual(values=c("IGA" = "blue4", "IGG" = "olivedrab3", "IGM" = "black", "all" = "blue4"))

png(filename="frequency_ranges.png")
print(p)
dev.off()

5:012a738edf5a

		transition2 = merge(transition2, base.order, by.x="id", by.y="base")

		transition2 = merge(transition2, base.order, by.x="variable", by.y="base")

		transition2[is.na(transition2$value),]$value = 0

		if(any(transition2$value != 0)){ #having rows of data but a transition table filled with 0 is bad
			print("Plotting stacked transition")
			png(filename=paste("transitions_stacked_", name, ".png", sep=""))
			p = ggplot(transition2, aes(factor(reorder(id, order.x)), y=value, fill=factor(reorder(variable, order.y)))) + geom_bar(position="fill", stat="identity", colour="black") #stacked bar
			p = p + xlab("From base") + ylab("To base") + ggtitle("Mutations frequency from base to base") + guides(fill=guide_legend(title=NULL))
			p = p + theme(panel.background = element_rect(fill = "white", colour="black"), text = element_text(size=16, colour="black")) + scale_fill_manual(values=c("A" = "blue4", "G" = "lightblue1", "C" = "olivedrab3", "T" = "olivedrab4"))
			#p = p + scale_colour_manual(values=c("A" = "black", "G" = "black", "C" = "black", "T" = "black"))
			print(p)
			dev.off()

			print("Plotting heatmap transition")

	genesForPlot$label = paste(genesForPlot$Gene, "-", genesForPlot$Freq)

	pc = ggplot(genesForPlot, aes(x = factor(1), y=Freq, fill=Gene))
	pc = pc + geom_bar(width = 1, stat = "identity") + scale_fill_manual(labels=genesForPlot$label, values=c("IGA1" = "lightblue1", "IGA2" = "blue4"))
	pc = pc + coord_polar(theta="y") + scale_y_continuous(breaks=NULL)
	pc = pc + theme(panel.background = element_rect(fill = "white", colour="black"), text = element_text(size=16, colour="black"), axis.title=element_blank(), axis.text=element_blank(), axis.ticks=element_blank())
	pc = pc + xlab(" ") + ylab(" ") + ggtitle(paste("IGA subclasses", "( n =", sum(genesForPlot$Freq), ")"))
	write.table(genesForPlot, "IGA_pie.txt", sep="\t",quote=F,row.names=F,col.names=T)

	png(filename="IGA.png")
	print(pc)

	genesForPlot$label = paste(genesForPlot$Gene, "-", genesForPlot$Freq)

	pc = ggplot(genesForPlot, aes(x = factor(1), y=Freq, fill=Gene))
	pc = pc + geom_bar(width = 1, stat = "identity") + scale_fill_manual(labels=genesForPlot$label, values=c("IGG1" = "olivedrab3", "IGG2" = "red", "IGG3" = "gold", "IGG4" = "darkred"))
	pc = pc + coord_polar(theta="y") + scale_y_continuous(breaks=NULL)
	pc = pc + theme(panel.background = element_rect(fill = "white", colour="black"), text = element_text(size=16, colour="black"), axis.title=element_blank(), axis.text=element_blank(), axis.ticks=element_blank())
	pc = pc + xlab(" ") + ylab(" ") + ggtitle(paste("IGG subclasses", "( n =", sum(genesForPlot$Freq), ")"))
	write.table(genesForPlot, "IGG_pie.txt", sep="\t",quote=F,row.names=F,col.names=T)

	png(filename="IGG.png")
	print(pc)

dat.clss = rbind(dat, dat.clss)

p = ggplot(dat.clss, aes(best_match, percentage_mutations))
p = p + geom_point(aes(colour=best_match), position="jitter") + geom_boxplot(aes(middle=mean(percentage_mutations)), alpha=0.1, outlier.shape = NA)
p = p + xlab("Subclass") + ylab("Frequency") + ggtitle("Frequency scatter plot") + theme(panel.background = element_rect(fill = "white", colour="black"), text = element_text(size=16, colour="black"))
p = p + scale_fill_manual(values=c("IGA" = "blue4", "IGA1" = "lightblue1", "IGA2" = "blue4", "IGG" = "olivedrab3", "IGG1" = "olivedrab3", "IGG2" = "red", "IGG3" = "gold", "IGG4" = "darkred", "IGM" = "darkviolet", "all" = "blue4"))
p = p + scale_colour_manual(values=c("IGA" = "blue4", "IGA1" = "lightblue1", "IGA2" = "blue4", "IGG" = "olivedrab3", "IGG1" = "olivedrab3", "IGG2" = "red", "IGG3" = "gold", "IGG4" = "darkred", "IGM" = "darkviolet", "all" = "blue4"))

png(filename="scatter.png")
print(p)

frequency_bins_data = merge(frequency_bins_data, frequency_bins_sum, by="best_match_class")

frequency_bins_data$frequency = round(frequency_bins_data$frequency_count / frequency_bins_data$class_sum * 100, 2)

p = ggplot(frequency_bins_data, aes(frequency_bins, frequency))
p = p + geom_bar(aes(fill=best_match_class), stat="identity", position="dodge") + theme(panel.background = element_rect(fill = "white", colour="black"), text = element_text(size=16, colour="black"))
p = p + xlab("Frequency ranges") + ylab("Frequency") + ggtitle("Mutation Frequencies by class") + scale_fill_manual(values=c("IGA" = "blue4", "IGG" = "olivedrab3", "IGM" = "black", "all" = "blue4"))

png(filename="frequency_ranges.png")
print(p)
dev.off()