YingMa0107
commented
4 years ago Hi,
Thanks for your interest in our package iDEA! Did you mean that you want to draw the idea@gsea results into a figure like bubble plot in the iDEA paper i.e. Figure 4e,5e,6e? If so, I've uploaded one example plot data here https://github.com/xzhoulab/iDEA-Analysis/tree/master/figures. You can follow the example code bubble_plot.R to generate the bubble plot.
Besides that, here is the example code for generating the plot data frame
-------------------------------------------------------------
create the plot data frame for bubble plot
-------------------------------------------------------------
Extract the category information for the human gene sets
library(iDEA)
you can download the data here https://github.com/xzhoulab/iDEA/tree/master/data
or extract it from the updated package.
data(humanGeneSetsInfo) Category = humanGeneSetsInfo[,c(4:5)] colnames(Category) = c("Category","annot_id") count = data.frame(annot_id = names(idea@annotation),Count = unlist(lapply(1:length(idea@annotation),function(i){length(idea@annotation [[i]])})))
merge the data with gsea results of iDEA
df = merge(idea@gsea,count,by = "annot_id") df = merge(df,Category,by = "annot_id")
create plotdata
plotdata = data.frame(IDNum = c(1:nrow(df)), ID = df$annot_id, Term = df$annot_id, Category = df$Category, Count = df$Count, Pvalue_Louis = idea@gsea$pvalue_louis, Log10_Pvalue_Louis = -log(idea@gsea$pvalue_louis,base = 10)) CateoryLevels = c("IMMUNOLOGIC SIGNATURES", "CHEMICAL AND GENETIC PERTURBATIONS", "GO BIOLOGICAL PROCESS", "GO MOLECULAR FUNCTION","GO CELLULAR COMPONENT","ONCOGENIC SIGNATURES","REACTOME","KEGG","PID","BIOCARTA") plotdata = plotdata[order(match(plotdata$Category, CateoryLevels)),]
The following code are just the same as the code in https://github.com/xzhoulab/iDEA-Analysis/blob/master/figures/bubble_plot.R
select the top biologically significant gene set in the bubbleplot to show
SigSelectTem = plotdata$Term[order(plotdata$Pvalue_Louis,decreasing = F)[1:5]] Sig = plotdata[which(plotdata$Term %in% SigSelectTem),]
you can modify your text label here .i.e. not capitalize the gene set term
Sig$Term = tolower(Sig$Term)
-------------------------------------------------------------
ggplot function for bubble plot
-------------------------------------------------------------
library(ggplot2) library("ggrepel") p4 = ggplot(plotdata, aes(x = IDNum, y = Log10_Pvalue_Louis,color =Category))+ geom_point(shape = 19,aes(fill = Category, size = Count,shape =19),alpha=0.8)+ scale_radius()+ labs(x = "", y =expression(paste(bold(-log[10]),bold("("),bolditalic(p),bold("-value)"))))+ theme(plot.margin = margin(1, 1, 1, 1, "cm"), axis.text.x=element_blank(), plot.title = element_text(lineheight=.8, face="bold"), axis.text = element_text(size = 30), axis.line = element_line(colour = 'black'), axis.ticks = element_line(colour = 'grey80'), axis.title = element_text(size = 40, face = 'bold'), axis.title.y = element_text(margin = margin(t = 0, r = 20, b = 0, l = 0)), legend.title=element_text(size=24,face = 'bold'), legend.text=element_text(size=24), panel.background = element_blank(), panel.grid.minor = element_blank(), panel.grid.major = element_line(colour = 'white'), plot.background = element_blank(), legend.key = element_rect(color = "transparent", fill = "transparent"))+ geom_hline(yintercept = 1.82, col = 'black',linetype = 2,size=2)+ guides(color = guide_legend(order = 1,override.aes = list(alpha =1,size=7)), size = guide_legend(order = 2,override.aes = list(alpha = 1,shape=21)),fill = FALSE)+ labs(size = "Gene set size")+ scale_color_manual(values=c("salmon","gold2","#42d4f4","#3cb44b","chocolate2","#4363d8","#bfef45","#911eb4","#f032e6","#a9a9a9"))+ scale_fill_manual(values = c("salmon","gold2","#42d4f4","#3cb44b","chocolate2","#4363d8","#bfef45","#911eb4","#f032e6","#a9a9a9"))+ theme(legend.direction = "vertical")+ theme(legend.position = c(0.30, 0.85))+ theme(legend.box = "horizontal")+ theme(legend.title.align = 0)+ geom_text_repel( data = Sig, aes(label = Term), force=1.0, point.padding=unit(1.1,'lines'), box.padding = unit(0.1, "lines"), vjust=-1.4, hjust = 0.2, size = 8, direction='y', nudge_x=0.2, nudge_y = 0.2, segment.size=0.2, col = "black")
-------------------------------------------------------------
Write out the .png file
-------------------------------------------------------------
png("~/Figure5e.png",width = 12500, height = 7500, res = 600) p4 dev.off()
You may have to change the legend.position according to the number of gene sets you tested and the width and height of the figure you set. And all the parameters in the theme can be modified by yourself. Hope it helps!
Best, Ying
On Sun, May 3, 2020 at 8:59 PM zh-Bian notifications@github.com wrote:
After we bulit the iDEA model uesd the data, there is gsea result in the iDEA@gsea. Could we draw them into a figure but not write in a table? I tried to used the other function but the structure of the object is different. So I download the program in the https://github.com/xzhoulab/iDEA-Analysis , but it did not tell us how to draw the gsea result into a figrue. So I turn to you for help, could you give us an example ahout how to make the result into the figure? Thank you.! @sqsun https://github.com/sqsun @YingMa1993 https://github.com/YingMa1993
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/xzhoulab/iDEA/issues/1, or unsubscribe https://github.com/notifications/unsubscribe-auth/AG4QA2ISK4NL3M45DPDRYE3RPYHPVANCNFSM4MYLXUZQ .
-- Ying Ma Ph.D. Candidate Department of Biostatistics,School of Public Health. University of Michigan, Ann Arbor.
zh-Bian
princy149
After we bulit the iDEA model uesd the data, there is gsea result in the iDEA@gsea. Could we draw them into a figure but not write in a table? I tried to used the other function but the structure of the object is different. So I download the program in the https://github.com/xzhoulab/iDEA-Analysis , but it did not tell us how to draw the gsea result into a figrue. So I turn to you for help, could you give us an example ahout how to make the result into the figure? Thank you.! @sqsun @YingMa1993