文献配套GitHub发表级别绘图04-相关性散点图

rm(list = ls())
library(reshape2)
library(ggplot2)
library(RColorBrewer)
data <- read.delim('../data/Figure 2B input.txt', header = T, row.names = 1, check.names = F)
head(data)
idx <- c(2, 4, 6, 8, 24, 20, 22)
celltypes <- c('CD3 T cells (except for Tregs)', 'CD8 T cells', 'Memory T cells', 
               'Macrophages', 'Macrophages M2', 'Plasma cells', 'Neutrophils')

celltypes

idx <- c(2, 4, 6, 8, 24, 20, 22)
data[1,idx]
data[1,idx+1]

# 准备数据
i=1
tmpdata <- data[, c(1, idx[i], idx[i]+1)]
head(tmpdata)

colnames(tmpdata) <- c('Group', 'IHC', 'CIBERSORT')
tmpdata$log10IHC <- log10(tmpdata$IHC + 1)
tmpdata <- na.omit(tmpdata) # 删除缺失值
head(tmpdata)

cor.test(x, y,
         alternative = c("two.sided", "less", "greater"),
         method = c("pearson", "kendall", "spearman"),
         exact = NULL, conf.level = 0.95, continuity = FALSE, ...)
# 其中x,y是供检验的样本；alternative指定是双侧检验还是单侧检验；method为检验的方法；conf.level为检验的置信水平
# 参考：http://www.sthda.com/english/wiki/correlation-test-between-two-variables-in-r

spearman <- cor.test(tmpdata$IHC, tmpdata$CIBERSORT, method = 'spearman') 
str(spearman)

pval <- round(spearman$p.value, 3) # 设置小数点数量
coef <- data.frame(spearman$estimate) # 相关性
coef <- coef$spearman.estimate
coef <- round(coef, 3)
coef
# 设置标签,paste连接字符，collapse设置分隔符
text <- paste(c("Spearman Rho ", coef, '\n', 'P-value ', pval), collapse = '')
text

plt <- ggplot(tmpdata, aes(CIBERSORT, log10IHC)) + 
    geom_point(size=1) + # 改变shape形状, color线条颜色, fill填充颜色, size填充大小，stroke线条粗细
    geom_smooth(method = 'lm', se = FALSE, col = 'grey70') + # 拟合线，method：统计算法(lm\glm\gam\loess\rlm等),se：误差范围(就是围绕着拟合直线的颜色带),col:颜色
    labs(title=celltypes[i], y = 'log10 # of cell/mm2', x = 'CIBERSORT absolute score') + # 大标题，xy轴标签
    theme_bw(base_size = 7) + # 黑白主题：白色背景，灰色网格线;base_size控制字体大小
    theme(axis.text = element_text(colour = 'black'), # 轴刻度值
          axis.ticks = element_line(colour = 'black'), # 轴刻度线
          plot.title = element_text(hjust = 0.5), # 标题 hjust介于0,1之间,调节标题的横向位置
          panel.grid = element_blank()) + # 空白背景
    annotate("text", x = (max(tmpdata$CIBERSORT)+min(tmpdata$CIBERSORT))/2, 
             y = max(tmpdata$log10IHC)*0.95, label = text, size=2) + # 添加注释，"text"：指文本；xy 指定标签的位置；label：内容；size：大小
    xlim(0,max(tmpdata$CIBERSORT)) + ylim(0,max(tmpdata$log10IHC)) # xlim，ylim设置xy轴范围
plt

library(ggplot2)
data <- read.delim('../data/Figure 2B input.txt', header = T, row.names = 1, check.names = F)
colnames(data)
idx <- c(2, 4, 6, 8, 24, 20, 22)
celltypes <- c('CD3 T cells (except for Tregs)', 'CD8 T cells', 'Memory T cells', 
               'Macrophages', 'Macrophages M2', 'Plasma cells', 'Neutrophils')

celltypes
for (i in c(1:length(idx))) {
  tmpdata <- data[, c(1, idx[i], idx[i]+1)]
  head(tmpdata)
  colnames(tmpdata) <- c('Group', 'IHC', 'CIBERSORT')
  tmpdata$log10IHC <- log10(tmpdata$IHC + 1)
  tmpdata <- na.omit(tmpdata)
  head(tmpdata)
  
  spearman <- cor.test(tmpdata$IHC, tmpdata$CIBERSORT, method = 'spearman') 
  pval <- round(spearman$p.value, 3)
  coef <- data.frame(spearman$estimate)
  coef <- coef$spearman.estimate
  coef <- round(coef, 3)
  text <- paste(c("Spearman Rho ", coef, '\n', 'P-value ', pval), collapse = '')
  text
  
  plt <- ggplot(tmpdata, aes(CIBERSORT, log10IHC)) + 
    geom_point(size=1) +
    geom_smooth(method = 'lm', se = FALSE, col = 'grey70') + 
    labs(title=celltypes[i], y = 'log10 # of cell/mm2', x = 'CIBERSORT absolute score') + 
    theme_bw(base_size = 7) +
    theme(axis.text = element_text(colour = 'black'),
          axis.ticks = element_line(colour = 'black'),
          plot.title = element_text(hjust = 0.5),
          panel.grid = element_blank()) +
    annotate("text", x = (max(tmpdata$CIBERSORT)+min(tmpdata$CIBERSORT))/2, 
             y = max(tmpdata$log10IHC)*0.95, label = text, size=2) + 
    xlim(0,max(tmpdata$CIBERSORT)) + ylim(0,max(tmpdata$log10IHC)); plt
  
  outputPdf <- paste(c('../results/corrPlot.', paste(unlist(strsplit(celltypes[i], split = ' ')), collapse=''), 
                       '.small.log10trans.pdf'), collapse = '')
  ggsave(outputPdf, plt, units = 'cm', height = 5, width = 4.5)
}

df <- mtcars[, c("mpg", "cyl", "wt")]
df$cyl <- as.factor(df$cyl)
head(df)

library(ggplot2)
# Basic scatter plot
ggplot(df, aes(x=wt, y=mpg)) +
  geom_point()
# Change the point shape
ggplot(df, aes(x=wt, y=mpg)) +
  geom_point(shape=18)
# change shape, color, fill, size
ggplot(df, aes(x=wt, y=mpg)) +
  geom_point(shape=23, fill="blue", color="darkred", size=3)

library(ggplot2)
# Scatter plot with multiple groups
# shape depends on cyl
ggplot(df, aes(x=wt, y=mpg, group=cyl)) +
  geom_point(aes(shape=cyl))
# Change point shapes and colors
ggplot(df, aes(x=wt, y=mpg, group=cyl)) +
  geom_point(aes(shape=cyl, color=cyl))
# change point shapes,  colors and sizes
ggplot(df, aes(x=wt, y=mpg, group=cyl)) +
  geom_point(aes(shape=cyl, color=cyl, size=cyl))

# Change colors and shapes manually
ggplot(df, aes(x=wt, y=mpg, group=cyl)) +
  geom_point(aes(shape=cyl, color=cyl), size=2)+
  scale_shape_manual(values=c(3, 16, 17))+
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  theme(legend.position="top") # 注释位置
# Change the point size manually
ggplot(df, aes(x=wt, y=mpg, group=cyl)) +
  geom_point(aes(shape=cyl, color=cyl, size=cyl))+
  scale_shape_manual(values=c(3, 16, 17))+
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  scale_size_manual(values=c(2,3,4))+
  theme(legend.position="top")