刚上研一，复现blood（一）

2022-12-06 本文已影响0人生命数据科学

图1

Article name: A comprehensive transcriptome signature of murine hematopoietic stem cell aging

Journal: blood

Doi: 10.1182/blood.2020009729

IF: 23.629

Position: Figure 1C

图片

这是一张简单的条形图，用鼠标比着尺子画的话，10分钟就能画完，但是如果用R的话，用了俩小时

图2

不过学习了一些东西，也算傻人有傻福：

12个数据集的系统性规范化整理
对于建库过程中处理数据时的标准化问题
R语言中list的批量合并
管道运算
重复基因的分组与去重
去除空值
探针与基因中一对多的处理方法
ggplot2中如何按照y轴值的大小顺序绘制
绘制重叠条形图
如何选取好看的颜色
y轴坐标轴标签如何修改
如何去除背景刻度和背景颜色
y轴标签如何旋转一定角度并且紧贴坐标轴
标题，坐标轴标签如何修改
整个图片主题字体如何修改

图3

由于太晚了，就以两种方式进行分享：

1：可以在完整阅读文献后，下载原始数据，参考以下代码进行运行

library(ggplot2) 
library(dplyr)
library(tidyr)
library(reshape)
rm(list = ls())
setwd("./file")
Bersenev <- read.table("Bersenev_GSE39553.csv",sep = "\t",
                                              header = T,skip = 1,quote = "")%>%
  .[,c("Gene.symbol","logFC")]
colnames(Bersenev)<-c("genes","Bersenev")

Chambers <- read.table("Chambers_GSE6503.csv",sep = "\t",
                                              header = T,skip = 1,quote = "")%>%.[,c("Gene.symbol","logFC")]
colnames(Chambers)<-c("genes","Chambers")

Flach <- read.table("Flach_GSE48893.csv",sep = "\t",
                                        header = T,skip = 1,quote = "")%>%.[,c("Gene.symbol","logFC")]
colnames(Flach)<-c("genes","Flach")

Grover <- read.table("Grover_GSE70657.csv",sep = "\t",
                                          header = T,skip = 1,quote = "")%>%.[,c("Gene","avg_logFC")]
colnames(Grover)<-c("genes","Grover")

Kirschner <- read.table("Kirschner_GSE87631.csv",sep = "\t",
                                                header = T,skip = 1,quote = "")%>%.[,c("Gene","avg_logFC")]
colnames(Kirschner)<-c("genes","Kirschner")

Kowalczyk <- read.table("Kowalczyk_GSE59114.csv",sep = "\t",
                                                header = T,skip = 1,quote = "")%>%.[,c("Gene","avg_logFC")]
colnames(Kowalczyk)<-c("genes","Kowalczyk")

Lazare <- read.table("Lazare_GSE128050.csv",sep = "\t",
                                          header = T,skip = 1,quote = "")%>%.[,c("external_gene_name","logFC")]
colnames(Lazare)<-c("genes","Lazare")

Mann <- read.table("Mann_GSE1004426.csv",sep = "\t",
                                      header = T,skip = 1,quote = "")%>%.[,c("Gene","avg_logFC")]
colnames(Mann)<-c("genes","Mann")

Maryanovich <- read.table("Maryanovich_GSE109546.csv",sep = "\t",
                                                    header = T,skip = 1,quote = "")%>%.[,c("external_gene_name","logFC")]
colnames(Maryanovich)<-c("genes","Maryanovich")

Norddahl <- read.table("Norddahl_GSE27686.csv",sep = "\t",
                                              header = T,skip = 1,quote = "")%>%.[,c("Gene.symbol","logFC")]
colnames(Norddahl)<-c("genes","Norddahl")

Sun <- read.table("Sun_GSE47817.csv",sep = "\t",
                                    header = T,skip = 1,quote = "")%>%.[,c("external_gene_name","logFC")]
colnames(Sun)<-c("genes","Sun")

Wahlestedt <- read.table("Wahlestedt_GSE44923.csv",sep = "\t",
                                                  header = T,skip = 1,quote = "")%>%.[,c("Gene.symbol","logFC")]
colnames(Wahlestedt)<-c("genes","Wahlestedt")

all_data <- list(Bersenev,Chambers,Flach,Grover,Kirschner,
                                  Kowalczyk,Lazare, Mann,Maryanovich,Norddahl,  
                                  Sun,Wahlestedt)

all_pub <- purrr::reduce(.x = all_data,.f = full_join,by="genes")

geneMatrix <- all_pub %>% group_by(genes) %>% filter (!duplicated(genes))
geneMatrix<-geneMatrix[geneMatrix[,1]!="",]
blood_output<-separate_rows(geneMatrix,genes,sep = "///")
write.table(blood_output,"output.txt",sep = "\t",quote = F,row.names = F,col.names = T)

data <- blood_output[,2:13]

plot_matrix <- matrix(nrow = ncol(data),ncol = 3,dimnames = list(NULL,c("name","Upregulated","Downregulated")))
for (x in 1:ncol(data)) {
  data_name <- colnames(data)[x]
  non_na_data <- na.omit(data[,x])
  Upregulated <- length(non_na_data[non_na_data>0])
  Downregulated <- length(non_na_data[non_na_data<0])
  plot_matrix[x,] <- c(data_name,Upregulated,Downregulated)
}
plot_matrix <- as.data.frame(plot_matrix)
plot_matrix <- melt(plot_matrix,id.vars = c("name"))

plot_matrix$value <- as.numeric(plot_matrix$value) 

pl <- ggplot(data=plot_matrix, aes(x=reorder(name,-value), y=value)) +
  geom_bar(stat = "identity",aes(fill=variable))+
  scale_fill_manual(values=c("#005187","#e5082c"))+
  scale_y_continuous(breaks=c(1000,2000,3000),
                                        labels=c("1000", "2000", "3000"))+
  theme_bw()+
  theme(panel.grid=element_blank())+
  theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 1))+
  labs(x="",y="# of reported DE genes",title = "Reanalysis")+
  theme(text = element_text(family = "Arial",face = "bold"))

ggsave(pl, filename = "blood_figure_1c.pdf", device = cairo_pdf, 
              width = 8, height = 7, units = "in")

2：后台回复blood1c领取代码和数据，整个代码和文件将以project形式发送，也就是说，将文件解压后：

1. 双击blood_figure1.Rproj

图4

2.打开code文件夹中的code.R

图5

3.全选、运行即可

图6

4.结果将保存在file文件夹中，也会在Plots窗口展示

图7

刚上研一，复现blood（一）

猜你喜欢

热点阅读