signature 生存分析 p<0.05 的癌症：ACC、KIRP、LGG、LUAD、MESO、SARC、THYM

PCA将样本分成四组

sigs <- t(expr[c("CYP4B1","ADH1B","IL33","CD1D","CR1","CABCOCO1",
               "TCEAL2","ID1","CDC45","ANLN","CDCA2",
               "PRC1","TOP2A"),])

一套流程

rm(list = ls())
### expr
expr <- read.table('TCGA-DLBC.htseq_counts.tsv/TCGA-DLBC.htseq_counts.tsv',header = T,sep='\t')
tail(expr[,1])
expr <- expr[1:(nrow(expr)-5),]
expr[,1] <- as.character(expr[,1])
for (i in 1:nrow(expr)) {
  expr[i,1] <- strsplit(expr[i,1],"\\.")[[1]][1]
}
symb <- AnnotationDbi::select(org.Hs.eg.db,expr[,1],"SYMBOL","ENSEMBL")
symbtrim <- symb[match(unique(symb$SYMBOL),symb$SYMBOL),] %>% na.omit()
row.names(expr) <- expr[,1]
expr <- expr[,-1]
expr <-  expr[symbtrim$ENSEMBL,]
row.names(expr) <- symbtrim$SYMBOL
expr <- log2(edgeR::cpm(expr)+1) 
# kp <- substring(colnames(expr),14,15)=='01';table(kp)
# expr <- expr[,kp]
dim(expr)
### phe
phe <- read.table('./TCGA-DLBC.survival.tsv/TCGA-DLBC.survival.tsv',header = T,sep='\t',
                  fill = T,quote = "")
phe$sample = gsub('-','.',phe$sample)
row.names(phe) <- phe[,1]
# phe <- phe[substring(phe$sample,14,15)=='01',]
id <- intersect(colnames(expr),rownames(phe))
expr <- expr[,id]
phe <- phe[id,]
dim(phe)
dim(expr)
sigs <- t(expr[c("CYP4B1","ADH1B","IL33","CD1D","CR1","CABCOCO1",
               "TCEAL2","ID1","CDC45","ANLN","CDCA2",
               "PRC1","TOP2A"),])
## 这里如果不做转置，pca会根据13个基因来分组
library("FactoMineR") 
library("factoextra") 
dat.pca <- PCA(sigs, graph = FALSE) 
fviz_pca_ind(dat.pca,
             geom.ind = "point" 
)
table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
phe$pc1=dat.pca$ind$coord[,1]>0
phe$pc2=dat.pca$ind$coord[,2]>0
phe$gene=paste0(phe$pc1,phe$pc2)
library(survival)
library(survminer)
fit <- survfit(Surv(OS.time, OS) ~ gene,
               data = phe)
ggsurvplot(
  fit,                     # survfit object with calculated statistics.
  risk.table = TRUE,       # show risk table.
  pval = TRUE,             # show p-value of log-rank test.
  conf.int = TRUE,         # show confidence intervals for 
  # point estimaes of survival curves.
  xlim = c(0,200),        # present narrower X axis, but not affect
  # survival estimates. 
  risk.table.y.text.col = T, # colour risk table text annotations.
  risk.table.y.text = FALSE # show bars instead of names in text annotations
  # in legend of risk table
)

Ovarian Cancer (OV)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    89   88
#   TRUE     76  121

Head and Neck Cancer (HNSC)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE   105  148
#   TRUE    115  132

Acute Myeloid Leukemia (LAML)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    29   32
#   TRUE     30   41

Adrenocortical Cancer (ACC) (p=0.0014)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    15   21
#   TRUE     22   21

Bile Duct Cancer (CHOL)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE     8    6
#   TRUE     17   14

Bladder Cancer (BLCA) (p=0.034)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    83  104
#   TRUE    122  115

Breast Cancer (BRCA)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE   253  274
#   TRUE    285  382

Cervical Cancer (CESC)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    54   85
#   TRUE     81   76

Colon Cancer (COAD)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    93  122
#   TRUE    134  138

Endometrioid Cancer (UCEC)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    88  149
#   TRUE    163  167

Esophageal Cancer (ESCA)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#  FALSE    31   44
#   TRUE     50   47

Glioblastoma (GBM)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    31   43
#   TRUE     41   52

Kidney Chromophobe (KICH) (14 datasets)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    23   22
#   TRUE     18   24

Kidney Clear Cell Carcinoma (KIRC)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    97  112
#   TRUE    159  234

Kidney Papillary Cell Carcinoma (KIRP) (p=4e-04)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    69   88
#   TRUE     74   87

Large B-cell Lymphoma (DLBC)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    11   11
#   TRUE     13   12

Liver Cancer (LIHC)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    62  103
#   TRUE    123  130

Lower Grade Glioma (LGG) (p<0.0001)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE   118  158
#   TRUE    117  131

Lung Adenocarcinoma (LUAD) (p=0.0026)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE   133  133
#   TRUE    126  180

Lung Squamous Cell Carcinoma (LUSC)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    69  136
#   TRUE    150  187

Melanoma (SKCM)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#       Dim.2
# Dim.1   FALSE TRUE
#   FALSE    94  128
#   TRUE    106  130

不知道为什么不显示p值…

Mesothelioma (MESO) (p<0.0001)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    16   19
#   TRUE     23   26

Ocular melanomas (UVM)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    13   24
#   TRUE     20   23

Pancreatic Cancer (PAAD)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    25   60
#   TRUE     53   43

Pheochromocytoma & Paraganglioma (PCPG)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    46   43
#   TRUE     43   54

Prostate Cancer (PRAD)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE   125  152
#   TRUE    123  151

Rectal Cancer (READ)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    29   46
#   TRUE     43   49

Sarcoma (SARC) (p=0.0073)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    42   69
#   TRUE     73   78

Stomach Cancer (STAD)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    63  117
#   TRUE    108   93

Testicular Cancer (TGCT)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    63  117
#   TRUE    108   93

Thymoma (THYM) (p=0.023)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    25   26
#   TRUE     36   33

Thyroid Cancer (THCA)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE   108  139
#   TRUE    147  173

Uterine Carcinosarcoma (UCS)

table(as.data.frame(dat.pca$ind$coord[,1:2]>0))
#        Dim.2
# Dim.1   FALSE TRUE
#   FALSE    17   13
#   TRUE     11   13