qqboxplot--实现Q-Q plot和箱型图的整合！

BiocManager::install("qqboxplot")library(qqboxplot)

library(ggplot2)library(dplyr)#设置统一文本大小eltext <- 12#q-q boxplotqqbox <- expression_data %>%  ggplot(aes(specimen, log_count)) + geom_qqboxplot(varwidth = TRUE, notch = TRUE) + ylab('logged normalized expression') + ggtitle("c) q-q boxplot") + theme(plot.title=element_text(size=eltext, face="plain", hjust=0.5), axis.title.x = element_text(size=eltext), axis.title.y = element_text(size=eltext), panel.border = element_blank(), panel.background = element_rect(fill="white"), panel.grid.major = element_line(colour = "grey70"), panel.grid.minor = element_line(colour="grey80"))
#常规箱型图box <- expression_data %>% ggplot(aes(specimen, log_count)) + geom_boxplot(varwidth = TRUE, notch = TRUE) + ylab('logged normalized expression') + ggtitle('a) boxplot') + theme(plot.title=element_text(size=eltext, face="plain", hjust=0.5), axis.title.x = element_text(size=eltext), axis.title.y = element_text(size=eltext), panel.border = element_blank(), panel.background = element_rect(fill="white"), panel.grid.major = element_line(colour = "grey70"), panel.grid.minor = element_line(colour="grey80"))
override.shape <- c(16, 17)#q-q plotqq <- expression_data %>% ggplot(aes(sample=log_count)) + geom_qq(aes(color=specimen, shape=specimen)) + xlab('theoretical normal distribution') + ylab('logged normalized expression') + ggtitle('b) q-q plot') + labs(color="specimen") + guides(color = guide_legend(override.aes = list(shape=override.shape)), shape=FALSE) + theme(plot.title=element_text(size=eltext, face="plain", hjust=0.5), axis.title.x = element_text(size=eltext), axis.title.y = element_text(size=eltext), panel.border = element_blank(), panel.background = element_rect(fill="white"), panel.grid.major = element_line(colour = "grey70"), panel.grid.minor = element_line(colour="grey80"), legend.position = c(0.8, 0.2))
library(gridExtra)#合并图片gridExtra::grid.arrange(box, qq, qqbox, ncol=3)

tibble(y=c(rnorm(1000, mean=2), rt(1000, 16), rt(500, 4),  rt(1000, 8), rt(1000, 32)), group=c(rep("normal, mean=2", 1000),  rep("t distribution, df=16", 1000),  rep("t distribution, df=4", 500),  rep("t distribution, df=8", 1000),  rep("t distribution, df=32", 1000)))

simulated_data %>% ggplot(aes(factor(group, levels=c("normal, mean=2", "t distribution, df=32", "t distribution, df=16", "t distribution, df=8", "t distribution, df=4")), y=y)) + geom_boxplot(notch=TRUE, varwidth = TRUE) + xlab(NULL) + ylab(NULL) + theme(axis.text.x = element_text(angle = 23, size = 15), axis.title.y = element_text(size=15), panel.border = element_blank(), panel.background = element_rect(fill="white"),        panel.grid = element_line(colour = "grey70"))

override.shape <- c(16, 17, 15, 3, 7)
simulated_data %>% ggplot(aes(sample=y, color=factor(group, levels=c("normal, mean=2", "t distribution, df=32", "t distribution, df=16", "t distribution, df=8", "t distribution, df=4")), shape=factor(group, levels=c("normal, mean=2", "t distribution, df=32", "t distribution, df=16", "t distribution, df=8", "t distribution, df=4")))) + geom_qq() + geom_qq_line() + labs(color="distribution") + xlab("Normal Distribution") + ylab("Simulated Datasets") + guides(color = guide_legend(override.aes = list(shape=override.shape)), shape=FALSE) + theme(axis.title.x = element_text(size=15), axis.title.y = element_text(size=15), panel.border = element_blank(), panel.background = element_rect(fill="white"),        panel.grid = element_line(colour = "grey70"))

simulated_data %>% ggplot(aes(factor(group, levels=c("normal, mean=2", "t distribution, df=32", "t distribution, df=16", "t distribution, df=8", "t distribution, df=4")), y=y)) + geom_qqboxplot(notch=TRUE, varwidth = TRUE, reference_dist="norm") + xlab("reference: normal distribution") + ylab(NULL) + guides(color=FALSE) + theme(axis.text.x = element_text(angle = 23, size = 15), axis.title.y = element_text(size=15), axis.title.x = element_text(size=15), panel.border = element_blank(), panel.background = element_rect(fill="white"),        panel.grid = element_line(colour = "grey70"))

comparison_data <- indicators %>% filter(year==2008 & `Series Code`=="SL.TLF.ACTI.1524.MA.NE.ZS")
indicators %>%  #将series名称中的标签更改为较短的标题  mutate(`Series Name`= ifelse( `Series Name`=="Labor force participation rate for ages 15-24, male (%) (national estimate)",  "Male ages 15-24",  "Female ages 15-24")) %>% ggplot(aes(as.factor(year), y=indicator))+ geom_qqboxplot(notch=TRUE, varwidth = TRUE, compdata=comparison_data$indicator) + xlab("Year") + ylab("Country level labor force\nparticipation rate (%)") + facet_wrap(~factor(`Series Name`, levels = c("Male ages 15-24", "Female ages 15-24"))) + theme(strip.text = element_text(size=12), axis.text.x = element_text(size = 15), axis.title.x = element_text(size=15), axis.title.y = element_text(size=12), panel.border = element_blank(), panel.background = element_rect(fill="white"),        panel.grid = element_line(colour = "grey70"))

spike_data %>% filter(region=="V1") %>% ggplot(aes(factor(orientation), nspikes)) + geom_qqboxplot(notch=TRUE, varwidth = TRUE, reference_dist="norm") + xlab("orientation") + ylab("spike count") + theme(axis.text.x = element_text(size = 15), axis.text.y = element_text(size=14), axis.title.x = element_text(size=15), axis.title.y = element_text(size=15), panel.border = element_blank(), panel.background = element_rect(fill="white"),        panel.grid = element_line(colour = "grey70"))

spike_data %>% filter(region=="V1") %>% ggplot()+ geom_violin(aes(x=factor(orientation),y=nspikes),fill='grey',trim=F, draw_quantiles = c(.25, .5, .75))+ geom_segment(aes( x=match(factor(orientation),levels(factor(orientation)))-0.1, xend=match(factor(orientation),levels(factor(orientation)))+0.1, y=nspikes,yend=nspikes), col='black' ) + xlab("orientation") + ylab("spike count") + theme(axis.text.x = element_text(size = 15), axis.text.y = element_text(size=14), axis.title.x = element_text(size=15), axis.title.y = element_text(size=15), panel.border = element_blank(), panel.background = element_rect(fill="white"), panel.grid = element_line(colour = "grey70"))