tcga中配对样本的可视化

tcga癌旁数据较少，所以有时候看配对数据的对比，更能代表基因在癌和癌旁的区别。

dd = fread("COAD_Portal_RNA_FPKM.txt", h = T, sep = "\t", check.names = F) %>% column_to_rownames("ensembl_gene_id")
text <- dd[1:10,1:10]

table(!duplicated(colnames(dd)))

dd = dd[,!duplicated(colnames(dd))]
#去重复测序两次的标本

tcga_panmRNA_expr1 <- dd

IGHG1 <- tcga_panmRNA_expr1['ENSG00000211896',]
IGHG2 <- tcga_panmRNA_expr1['ENSG00000211893',] 
IGHG3 <- tcga_panmRNA_expr1['ENSG00000211897',]
IGHG4 <- tcga_panmRNA_expr1['ENSG00000211892',]
IGHA1 <- tcga_panmRNA_expr1['ENSG00000211895',]
IGHA2 <- tcga_panmRNA_expr1['ENSG00000211890',]
JCHAIN <- tcga_panmRNA_expr1['ENSG00000132465',]


tcga_panmRNA_expr2 <- rbind(IGHG1,IGHG2,IGHG3,IGHG4,IGHA1,IGHA2,JCHAIN)
rownames(tcga_panmRNA_expr2) <- c('IGHG1','IGHG2','IGHG3','IGHG4','IGHA1','IGHA2','JCHAIN')
T=tcga_panmRNA_expr2[,str_sub(colnames(tcga_panmRNA_expr2),14,16)=="01A"]#肿瘤样本
N=tcga_panmRNA_expr2[,str_sub(colnames(tcga_panmRNA_expr2),14,16)=="11A"]
colnames(T) <- str_sub(colnames(T),1,12)
colnames(N) <- str_sub(colnames(N),1,12)
T1 <- as.data.frame(t(T))
N1 <- as.data.frame(t(N))

T1$sampel <- 'tumor'
N1$sampel <- 'mormal'

index <- intersect(rownames(T1),rownames(N1))
exprSet_pair <- data.frame(ID=rep(seq(1,length(index)),2),
                           rbind(T1[index,],N1[index,]))

data <- exprSet_pair
library(ggpubr)
C41 <- ggpaired(data, x = "sampel", y = "IGHA1",
         color = "black", 
         fill = c("#E11E24","#FBB96F"),
         line.color = "gray", line.size = 0.4,
         ylab = "expression of IGHA1",
         palette = "npg")+
                    stat_compare_means(paired = TRUE)