CCDC134 与铁死亡相关基因相关性分析

数据读取

setwd("~/Documents/coploit/home")
#protein <- fread("protein/HCC_protein_after_scale.xls") %>% as.data.frame()
#rna <- fread("RNAseq/HCC_TPM_after_scale.xls") %>% as.data.frame()

#rna
all_TPM <- fread("SynologyDrive/RNAseq/All.TPM.exp.xls")

all_TPM <- all_TPM[,-c(208:219)]
all_TPM[1:4,202:207]

##    U00000828 U00000728 U00000880 U00000780 U00000821 U00000721
##        <num>     <num>     <num>     <num>     <num>     <num>
## 1:         0         0         0         0         0         0
## 2:         0         0         0         0         0         0
## 3:         0         0         0         0         0         0
## 4:         0         0         0         0         0         0

all_TPM = as.data.frame(all_TPM)
table(duplicated(all_TPM$AccID))

## 
## FALSE  TRUE 
## 56638  1711

new_all_TPM <- all_TPM %>% 
  dplyr :: distinct(AccID,.keep_all = TRUE) 
table(duplicated(new_all_TPM$AccID))

## 
## FALSE 
## 56638

new_all_TPM <- new_all_TPM %>%
  as_tibble() %>%
  column_to_rownames(var = "AccID")

id <- read.xlsx("SynologyDrive/RNAseq/肝癌RNA-seq样本编号.xlsx")  #该文件中有样本编号对应的住院号（seqn）

hcc_tpm <- new_all_TPM[,colnames(new_all_TPM)  %in% id$uid]  # 癌组织
para_tpm <- new_all_TPM[,colnames(new_all_TPM)  %in% id$pair]  #癌旁组织

hcc_rna<-log2(hcc_tpm + 1)
para_rna<-log2(para_tpm + 1)

#protein
all_protein <- fread("SynologyDrive/protein/dia_after_immpute_onehalf.xls") %>% as.data.frame()
row.names(all_protein)<-all_protein$V1; all_protein<-all_protein[,-1,drop=F]

id2 <- read.xlsx("SynologyDrive/protein/dia_sample_id.xlsx")  #该文件中有样本编号对应的住院号（seqn）
#table(id2$group)
hcc_pro<- all_protein[,colnames(all_protein)  %in% subset(id2,group=="hcc")$id,drop=F] # 癌组织
para_pro <- all_protein[,colnames(all_protein) %in% subset(id2,group=="para")$id]  #癌旁组织


##
fe <- c("CCDC134", "GPX4", "SAT1", "GCLM", "CP", "FTH1")

1. (肿瘤组织）查看目标基因

hcc_pro[row.names(hcc_pro) %in% fe, 1:2]

##              1875      1877
## CP      2402.3438 1673.9227
## FTH1     451.8825  470.6070
## GPX4     966.6021  477.3249
## GCLM     439.0198  356.1182
## CCDC134  304.5132  450.2108

hcc_rna[row.names(hcc_rna) %in% fe, 1:2]

##         U00000991 U00000990
## CP       7.960560  8.676991
## SAT1     9.017533  9.800277
## FTH1     9.366623  7.880857
## GCLM     5.481211  5.323010
## GPX4     7.759533  8.638239
## CCDC134  2.564475  1.460593

1. (in tumor) 数据整理

protein1 <- hcc_pro[row.names(hcc_pro) %in% fe, ]  %>% t()
rna1 <- hcc_rna[row.names(hcc_rna) %in% fe, ]%>% t()

CCDC134 与 Fe 基因（RNA）相关性分析

ccdc134_rna <- rna1[, "CCDC134"]
fe_genes <- setdiff(colnames(rna1), "CCDC134")
cor_stats <- sapply(fe_genes, function(gene) {
    test <- cor.test(ccdc134_rna, rna1[, gene], method = "pearson")
    c(cor = test$estimate, p = test$p.value)
})
cor_df <- data.frame(
    Gene = fe_genes,
    Correlation = cor_stats["cor.cor", ],
    P.value = cor_stats["p", ]
)
cor_df$Significant <- cor_df$P.value < 0.05
knitr::kable(cor_df, digits = 3)

	Gene	Correlation	P.value	Significant
CP	CP	0.052	0.601	FALSE
SAT1	SAT1	-0.110	0.269	FALSE
FTH1	FTH1	0.339	0.000	TRUE
GCLM	GCLM	0.439	0.000	TRUE
GPX4	GPX4	0.171	0.085	FALSE

柱状图展示相关性（RNA）

ggplot(cor_df, aes(x = Gene, y = Correlation, fill = Significant, label = sprintf("P=%.3g", P.value))) +
  geom_bar(stat = "identity", width = 0.6) +
  geom_text(vjust = ifelse(cor_df$Correlation >= 0, -0.5, 1.5), size = 3.5) +
  scale_fill_manual(values = c("FALSE" = "grey70", "TRUE" = "red")) +
  ylim(-0.5, 0.5) +
  theme_minimal() +
  labs(title = "Correlation of CCDC134 with Fe genes (RNA) in tumor", y = "Pearson Correlation", x = "Fe Gene") +
  theme(legend.position = "none")

散点图及拟合线（RNA）

for (gene in fe_genes) {
    df <- data.frame(
        CCDC134 = rna1[, "CCDC134"],
        FeGene = rna1[, gene]
    )
    test <- cor.test(df$CCDC134, df$FeGene, method = "pearson")
    pval <- signif(test$p.value, 3)
    corval <- signif(test$estimate, 3)
    gg <- ggplot(df, aes(x = CCDC134, y = FeGene)) +
        geom_point(size = 2, color = "steelblue") +
        geom_smooth(method = "lm", se = FALSE, color = "red") +
        labs(
            title = paste("CCDC134 vs", gene, "(RNA) in tumor"),
            subtitle = paste0("Pearson r = ", corval, ", P = ", pval),
            x = "CCDC134 (RNA)",
            y = paste(gene, "(RNA)")
        ) +
        theme_minimal()
    print(gg)
}

CCDC134 与 Fe 基因（Protein）相关性分析

ccdc134_protein <- protein1[, "CCDC134"]
fe_genes_protein <- setdiff(colnames(protein1), "CCDC134")
cor_stats_protein <- sapply(fe_genes_protein, function(gene) {
  test <- cor.test(ccdc134_protein, protein1[, gene], method = "pearson")
  c(cor = test$estimate, p = test$p.value)
})
cor_df_protein <- data.frame(
  Gene = fe_genes_protein,
  Correlation = cor_stats_protein["cor.cor", ],
  P.value = cor_stats_protein["p", ]
)
cor_df_protein$Significant <- cor_df_protein$P.value < 0.05
knitr::kable(cor_df_protein, digits = 3)

	Gene	Correlation	P.value	Significant
CP	CP	0.447	0.008	TRUE
FTH1	FTH1	-0.064	0.719	FALSE
GPX4	GPX4	-0.226	0.198	FALSE
GCLM	GCLM	-0.022	0.902	FALSE

柱状图展示相关性（Protein）

ggplot(cor_df_protein, aes(x = Gene, y = Correlation, fill = Significant, label = sprintf("P=%.3g", P.value))) +
  geom_bar(stat = "identity", width = 0.6) +
  geom_text(vjust = ifelse(cor_df_protein$Correlation >= 0, -0.5, 1.5), size = 3.5) +
  scale_fill_manual(values = c("FALSE" = "grey70", "TRUE" = "red")) +
  ylim(-0.5, 0.5) +
  theme_minimal() +
  labs(title = "Correlation of CCDC134 with Fe genes (Protein) in tumor", y = "Pearson Correlation", x = "Fe Gene") +
  theme(legend.position = "none")

散点图及拟合线（Protein）

for (gene in fe_genes_protein) {
  df <- data.frame(
    CCDC134 = protein1[, "CCDC134"],
    FeGene = protein1[, gene]
  )
  test <- cor.test(df$CCDC134, df$FeGene, method = "pearson")
  pval <- signif(test$p.value, 3)
  corval <- signif(test$estimate, 3)
  gg <- ggplot(df, aes(x = CCDC134, y = FeGene)) +
    geom_point(size = 2, color = "steelblue") +
    geom_smooth(method = "lm", se = FALSE, color = "red") +
    labs(
      title = paste("CCDC134 vs", gene, "(Protein) in tumor"),
      subtitle = paste0("Pearson r = ", corval, ", P = ", pval),
      x = "CCDC134 (Protein)",
      y = paste(gene, "(Protein)")
    ) +
    theme_minimal()
  print(gg)
}

2. (癌旁组织）查看目标基因

para_pro[row.names(para_pro) %in% fe, 1:2]

##               1876      1878
## CP      1013.98444 1787.4272
## FTH1     825.94214 1039.2148
## GPX4    2892.94434 1541.0046
## GCLM     363.01901  414.3443
## CCDC134   70.55958  150.4873

para_rna[row.names(para_rna) %in% fe, 1:2]

##         U00001020 U00001019
## CP       9.145246  8.530697
## SAT1     9.850314  9.525502
## FTH1     8.130989  8.184558
## GCLM     4.892074  4.889641
## GPX4     7.755636  8.123001
## CCDC134  1.729456  1.175767

2. (in para) 数据整理

protein1 <- para_pro[row.names(para_pro) %in% fe, ]  %>% t()
rna1 <- para_rna[row.names(para_rna) %in% fe, ]%>% t()

CCDC134 与 Fe 基因（RNA）相关性分析

ccdc134_rna <- rna1[, "CCDC134"]
fe_genes <- setdiff(colnames(rna1), "CCDC134")
cor_stats <- sapply(fe_genes, function(gene) {
    test <- cor.test(ccdc134_rna, rna1[, gene], method = "pearson")
    c(cor = test$estimate, p = test$p.value)
})
cor_df <- data.frame(
    Gene = fe_genes,
    Correlation = cor_stats["cor.cor", ],
    P.value = cor_stats["p", ]
)
cor_df$Significant <- cor_df$P.value < 0.05
knitr::kable(cor_df, digits = 3)

	Gene	Correlation	P.value	Significant
CP	CP	0.231	0.019	TRUE
SAT1	SAT1	0.068	0.497	FALSE
FTH1	FTH1	0.576	0.000	TRUE
GCLM	GCLM	0.419	0.000	TRUE
GPX4	GPX4	0.229	0.020	TRUE

柱状图展示相关性（RNA）

ggplot(cor_df, aes(x = Gene, y = Correlation, fill = Significant, label = sprintf("P=%.3g", P.value))) +
  geom_bar(stat = "identity", width = 0.6) +
  geom_text(vjust = ifelse(cor_df$Correlation >= 0, -0.5, 1.5), size = 3.5) +
  scale_fill_manual(values = c("FALSE" = "grey70", "TRUE" = "red")) +
  ylim(-0.5, 0.5) +
  theme_minimal() +
  labs(title = "Correlation of CCDC134 with Fe genes (RNA) in para", y = "Pearson Correlation", x = "Fe Gene") +
  theme(legend.position = "none")

散点图及拟合线（RNA）

for (gene in fe_genes) {
    df <- data.frame(
        CCDC134 = rna1[, "CCDC134"],
        FeGene = rna1[, gene]
    )
    test <- cor.test(df$CCDC134, df$FeGene, method = "pearson")
    pval <- signif(test$p.value, 3)
    corval <- signif(test$estimate, 3)
    gg <- ggplot(df, aes(x = CCDC134, y = FeGene)) +
        geom_point(size = 2, color = "steelblue") +
        geom_smooth(method = "lm", se = FALSE, color = "red") +
        labs(
            title = paste("CCDC134 vs", gene, "(RNA) in para"),
            subtitle = paste0("Pearson r = ", corval, ", P = ", pval),
            x = "CCDC134 (RNA)",
            y = paste(gene, "(RNA)")
        ) +
        theme_minimal()
    print(gg)
}

CCDC134 与 Fe 基因（Protein）相关性分析

ccdc134_protein <- protein1[, "CCDC134"]
fe_genes_protein <- setdiff(colnames(protein1), "CCDC134")
cor_stats_protein <- sapply(fe_genes_protein, function(gene) {
  test <- cor.test(ccdc134_protein, protein1[, gene], method = "pearson")
  c(cor = test$estimate, p = test$p.value)
})
cor_df_protein <- data.frame(
  Gene = fe_genes_protein,
  Correlation = cor_stats_protein["cor.cor", ],
  P.value = cor_stats_protein["p", ]
)
cor_df_protein$Significant <- cor_df_protein$P.value < 0.05
knitr::kable(cor_df_protein, digits = 3)

	Gene	Correlation	P.value	Significant
CP	CP	-0.225	0.200	FALSE
FTH1	FTH1	-0.135	0.445	FALSE
GPX4	GPX4	-0.498	0.003	TRUE
GCLM	GCLM	0.521	0.002	TRUE

柱状图展示相关性（Protein）

ggplot(cor_df_protein, aes(x = Gene, y = Correlation, fill = Significant, label = sprintf("P=%.3g", P.value))) +
  geom_bar(stat = "identity", width = 0.6) +
  geom_text(vjust = ifelse(cor_df_protein$Correlation >= 0, -0.5, 1.5), size = 3.5) +
  scale_fill_manual(values = c("FALSE" = "grey70", "TRUE" = "red")) +
  ylim(-0.5, 0.5) +
  theme_minimal() +
  labs(title = "Correlation of CCDC134 with Fe genes (Protein) in para", y = "Pearson Correlation", x = "Fe Gene") +
  theme(legend.position = "none")

散点图及拟合线（Protein）

for (gene in fe_genes_protein) {
  df <- data.frame(
    CCDC134 = protein1[, "CCDC134"],
    FeGene = protein1[, gene]
  )
  test <- cor.test(df$CCDC134, df$FeGene, method = "pearson")
  pval <- signif(test$p.value, 3)
  corval <- signif(test$estimate, 3)
  gg <- ggplot(df, aes(x = CCDC134, y = FeGene)) +
    geom_point(size = 2, color = "steelblue") +
    geom_smooth(method = "lm", se = FALSE, color = "red") +
    labs(
      title = paste("CCDC134 vs", gene, "(Protein) in para"),
      subtitle = paste0("Pearson r = ", corval, ", P = ", pval),
      x = "CCDC134 (Protein)",
      y = paste(gene, "(Protein)")
    ) +
    theme_minimal()
  print(gg)
}