论文

Plasma proteome analyses in individuals of European and African ancestry identify cis-pQTLs and models for proteome-wide association studies

https://www.nature.com/articles/s41588-022-01051-w

本地pdf s41588-022-01051-w.pdf

代码链接

https://zenodo.org/record/6332981#.YroV0nZBzic

https://github.com/Jingning-Zhang/PlasmaProtein/tree/v1.2

今天的推文重复一下论文中的Figure1，涉及到5个图，分别是折线图，韦恩图，散点图，频率分布直方图，最后一个知识点是如何将这5个图组合到一起

image.png

首先是定义作图主题的内容

library(ggplot2)

My_Theme <- theme(
  panel.background = element_blank(), 
  title = element_text(size = 7),
  text = element_text(size = 6)
)

论文中提供的代码没有设置坐标轴的线，如果按照他的主题来做出图没有横纵坐标轴

第一个折线图的代码

library(readxl)

df.peer <- read_excel("data/20220627/Fig1.xlsx", sheet = "1a")
df.peer_highlight <- df.peer[c(9,21),]

p1 <- ggplot(data = df.peer, aes(x = NPEER, y=NpGene, col=Race)) + 
  geom_line() + geom_point() +
  geom_point(data=df.peer_highlight,aes(x = NPEER, y=NpGene, color=Race, fill=Race), size=4,shape=18) +
  My_Theme +
  labs(y = "# significant SOMAmers      ", x="# PEER factors", title= NULL) +
  scale_colour_manual(values=c("#238b45","#2171b5")) + 
  scale_fill_manual(values=c("#238b45","#2171b5"))+
  theme(axis.line = element_line())
p1+
  scale_y_continuous(labels = scales::label_comma())

这里有一个小知识点，坐标轴文本千分位用逗号分隔，论文中没有提供这个代码，可以参考链接 https://scales.r-lib.org/reference/label_number.html

https://bookdown.org/Maxine/ggplot2-maps/posts/2019-11-27-using-scales-package-to-modify-ggplot2-scale/

image.png

第二个韦恩图的代码

他这里的韦恩图是借助ggforce这个R包直接画了两个圆

df.venn <- read_excel("data/20220627/Fig1.xlsx", sheet = "1b")

library(ggforce)
p2 <-  ggplot(df.venn, aes(x0 = x, y0 = y, r = r, fill = labels)) +
  geom_circle(alpha = .4, size = 1, colour = NA) +
  theme_void() +
  theme(legend.position="none")+
  My_Theme+
  annotate("text", x = -0.34, y =0.4, label = "1,618 in AA", size=2)+
  annotate("text", x = 0, y = 0, label = "1,447 overlapping", size=2)+
  annotate("text", x = 0.4, y =0.43, label = "2,004 in EA", size=2)+
  coord_fixed()+
  scale_fill_manual(values=c("#238b45","#2171b5"))
p2

image.png

第三个图是散点图添加拟合曲线

df.effmaf <- read_excel("data/20220627/Fig1.xlsx", sheet = "1c")

p3 <- ggplot(data = df.effmaf, aes(x = maf, y = abs(BETA), col=Race)) +
  geom_point(alpha=0.5,size=0.5) +
  # geom_point(alpha=0.5,size=power) +  
  geom_smooth(method="lm", mapping = aes(weight = 1/power), col="#fd8d3c") +
  geom_smooth(method="lm", col="#525252") +
  scale_x_continuous(breaks=c(0, 0.1, 0.2)) +
  theme(legend.position="none")+
  My_Theme+
  facet_wrap(~Race,  ncol=2)+
  labs(x="MAF(1-MAF)", y = "Effect size", title = NULL)+
  scale_colour_manual(values=c("#238b45","#2171b5"))+
  theme(axis.line.y = element_line(),
        axis.line.x = element_line())
p3

分面以后两个x轴不能完全链接到一起，可能需要出图后再编辑

image.png

第四个散点图

df.efftss <- read_excel("data/20220627/Fig1.xlsx", sheet = "1d")

p4 <- ggplot(data = df.efftss, aes(x = dist, y = abs(BETA), col=Race)) + 
  geom_point(alpha=0.5,size=0.5) +  
  scale_x_continuous(breaks=c(-0.25*10^6,0.25*10^6),
                     labels = scales::label_comma()) +
  theme(legend.position="none")+
  My_Theme+
  facet_wrap(~Race,  ncol=2)+
  labs(x="Distance to TSS", y = "Effect size", title = NULL)+
  scale_colour_manual(values=c("#238b45","#2171b5"))+
  coord_cartesian(ylim = c(0.2,NA)) +
  theme(axis.line.y = element_line(),
        axis.line.x = element_line())
p4

image.png

第五个频率分布直方图

df.cond <- read_excel("data/20220627/Fig1.xlsx", sheet = "1e")

p5 <- ggplot(df.cond, aes(x = count)) + 
  geom_histogram(binwidth = 1, aes(fill=Race), col="white",alpha=0.8) +
  theme(legend.position="none")+
  My_Theme+
  labs(x = "# conditionally significant cis-SNPs", 
       y = "# significant SOMAmers",title=NULL) + 
  facet_grid(cols = vars(Race))+
  scale_fill_manual(values=c("#238b45","#2171b5"))+
  theme(axis.line.y = element_line(),
        axis.line.x = element_line())
p5

image.png

最后是拼图代码

library(ggpubr)
p <- ggarrange(ggarrange(p1, p2,
                         ncol = 2, labels = c("a", "b"),
                         widths = c(0.6,0.4)),
               ggarrange(p3, p4, p5,
                         ncol = 3, labels = c("c", "d","e"),
                         widths = c(0.35,0.35,0.3)),
               nrow = 2, heights = c(0.4,0.6))
p

image.png

示例数据和代码可以直接去论文中获取

欢迎大家关注我的公众号

小明的数据分析笔记本

小明的数据分析笔记本 公众号 主要分享：1、R语言和python做数据分析和数据可视化的简单小例子；2、园艺植物相关转录组学、基因组学、群体遗传学文献阅读笔记；3、生物信息学入门学习资料及自己的学习笔记！