一、Spearman analysis
#RGS1(cg10718027)
setwd("G:/A/H/CO/CO2")
spss <- read_csv("spss2.csv") ##read data
library(ggplot2) ## library R packages
library(ggpubr)
p <- ggplot(data = spss, aes(x = cg10718027, y = RGS1,  color = cyl)) + 
  geom_point() + geom_smooth(method = lm) +
  scale_color_manual(values = c('#FF7400', '#009999'))  +
  labs (title = 'iris') + 
  theme_bw() + theme(plot.title = element_text(hjust = 0.5)) + 
  stat_cor(method = 'spearman', aes(x = cg10718027, y = RGS1,  color = cyl),label.x = 0.6)
#Set the title, coordinate axes
p<-p+ggtitle("RGS1(cg10718027)") +xlab ("Methylation level") + ylab ("Gene expression(mRNA)") + theme (title = element_text (family="sans", size=12, color="black",face="plain",hjust=0.5))
p<-p+theme (axis.text.x=element_text(family="sans", size=12,color="black",face="plain"),axis.text.y=element_text(family="sans",size=12,color="black",face="plain"))
p<-p+theme(plot.title = element_text(family="sans",size = 12,hjust=0.5)) 
p<-p+theme(legend.title  = element_text(family="sans",size = 12,hjust=0.5))
p
#store the image
ggsave("RGS1(cg10718027).png", width=5.5, height=4) 




















二、GSEA analysis
setwd("G:/A/H/GSEA")
rm (list = ls())
###1. Install Packages
library(devtools)
install_local("G:/A/H/GSEA/Xinpeng-Zhou-msigdf-master.zip")
###2. library R packages
library(ggplot2)
library(clusterProfiler)
library(tidyverse)
library(msigdf)
library(org.Hs.eg.db)
library(enrichplot)
# ##3. data reduction
all_gene <- read_csv("G:/A/H/GSEA/GSEA3.csv", na = c("", "#NAME?", "Inf")) %>% dplyr::select(gene_id, log2FoldChange) %>% filter(!is.na(log2FoldChange))
# FC_4 <- read_csv("./data/RA-vs-OA-diff-pval-0.05-FC-4_gene.csv", na = c("","#NAME?"))
# FC_2 <- read_csv("./data/RA-vs-OA-diff-pval-0.05-FC-2_gene.csv", na = c("","#NAME?"))
id <- bitr(all_gene$gene_id, "SYMBOL", "ENTREZID", "org.Hs.eg.db") %>% mutate(gene_id = SYMBOL) %>% dplyr::select(gene_id, ENTREZID) %>% inner_join(all_gene) %>% dplyr::select(ENTREZID, log2FoldChange) %>% arrange(desc(log2FoldChange))
genelist <- id$log2FoldChange
names(genelist) <- id$ENTREZID
# 4. GSEA
## Molecular Signatures Database (MSigDB)
## C5 GO gene set
c5 <- msigdf.human %>% filter(collection == "c5") %>% dplyr::select(geneset, entrez) %>% as.data.frame()
gsea_c5 <- GSEA (genelist, TERM2GENE = c5)
head(gsea_c5)
write.csv (gsea_c5, file ="gsea_c5.csv")  
ridgeplot(gsea_c5, showCategory = 3, label_format = 30)
ggsave ("ridgeplot.png", width=10, height=5)
gseaplot (gsea_c5, 12, title = gsea_c5@result[["ID"]][12])
gseaplot2(gsea_c5, 12, title = gsea_c5@result[["ID"]][12], base_size=11)
ggsave ("gsea_c5_12.png", width=7, height=5)



三、GO and KEGG analysis
rm(list = ls())
setwd("G:/GEO")
if (!require("BiocManager", quietly = TRUE))
  install.packages("BiocManager")
BiocManager::install(version = "3.15")
BiocManager::install("clusterProfiler")
BiocManager::install("org.Hs.eg.db")
BiocManager::install("KEGG.db")
library(stringr)###library R packages
library(clusterProfiler)
library(enrichplot)
library(org.Hs.eg.db)
library(KEGG.db)
library(forcats)
library(ggplot2)
read.csv("G:/GEO/DE_genes")
geneList <- DE_genes$logFC
DE_genes$symbol <- rownames(DE_genes)
names(geneList) <- DE_genes$symbol
geneList <- sort(geneList, decreasing = TRUE);head(geneList)
entreid <- bitr(names(geneList), fromType="SYMBOL", toType="ENTREZID", OrgDb="org.Hs.eg.db")
names(geneList) <- entreid$ENTREZID
gene <- names(geneList)[abs(geneList)>2]
head(gene)
kk <- enrichKEGG(gene = names(geneList),
                 organism = 'hsa',
                 pvalueCutoff = 0.05,
                 qvalueCutoff = 0.05,
                 use_internal_data = T)
head(kk)
nrow(kk)
kegg_data <- kk@result
p = ggplot(kegg_data,aes(-log(pvalue),Description))
p=p + geom_point() 
p=p + geom_point(aes(size=Count))
pbubble = p+ geom_point(aes(size=Count,color=-log10(pvalue)))
pr = pbubble+scale_color_gradient(low="#FDE725FF",high = "#2C738EFF")
pr = pr+labs(color=expression(-log(pvalue)),size="Count",  
             x="-log10(pvalue)",y="Description",title="KEGG pathway analysis")+
  scale_y_discrete(labels = function(x) str_wrap(x, width = 45) )
pr<-pr + theme_bw()
pr<-pr+theme(axis.title.x =element_text(family="sans",size=27,color="black",face="plain"), axis.title.y=element_text(family="sans",size=27,color="black",face="plain"))
pr<-pr+theme(legend.title = element_text(family="sans",size = 27,color="black",face="plain",hjust=0.5))
pr<-pr+theme(legend.text = element_text(family="sans",size = 25,color="black",face="plain",hjust=0.5))
pr<-pr+theme(plot.title = element_text(family="sans",face="bold",size =27,hjust=0.5))
print(pr)
ggsave("KEGG1.png",width=14.5,height=14.5)

ego <- enrichGO(gene = names(geneList),
                OrgDb = org.Hs.eg.db,
                ont = "ALL",
                pvalueCutoff = 0.05,
                pAdjustMethod = "BH",
                minGSSize = 10,
                maxGSSize = 500,
                readable = FALSE,
                pool = FALSE
)
go_data <- ego@result
p = ggplot(go_data,aes(-log(pvalue),Description))
p=p + geom_point() 
p=p + geom_point(aes(size=Count))
pbubble = p+ geom_point(aes(size=Count,color=-log10(pvalue)))
pr = pbubble+scale_color_gradient(low="#FDE725FF",high = "#2C738EFF")
pr = pr+labs(color=expression(-log(pvalue)),size="Count",  
             x="-log10(pvalue)",y="Description",title="BP Term")+
  scale_y_discrete(labels = function(x) str_wrap(x, width = 30) )
pr<-pr + theme_bw()
pr<-pr+theme(axis.title.x =element_text(family="sans",size=26,color="black",face="plain"), axis.title.y=element_text(family="sans",size=26,color="black",face="plain"))
pr<-pr+theme(axis.text.x =element_text(family="sans",size=26,color="black",face="plain"), axis.text.y=element_text(family="sans",size=26,color="black",face="plain"))
pr<-pr+theme(legend.title = element_text(family="sans",size = 26,color="black",face="plain",hjust=0.5))
pr<-pr+theme(legend.text = element_text(family="sans",size = 20,color="black",face="plain",hjust=0.5))
pr<-pr+theme(plot.title = element_text(family="sans",face="bold",size =26,hjust=0.5))
print(pr)
ggsave("GO.png",width=10.5,height=6.5)



 










四、multiVolcano

setwd("C:/Users/runru/Desktop/H")
library(magrittr)
library(ggplot2)
library(ggrepel)
library(janitor)
library(cowplot)
Sys.setenv(LANGUAGE = "en") #Display English error message
options(stringsAsFactors = FALSE) #Prohibit chr to factor
data <- read.csv("C:/Users/runru/Desktop/H/VOL/DMSinput.csv", check.names=FALSE)
row.names(data) <- data[, 1]
colnames(data)[1] <- "Geneid"
head(data)
p1 <- ggplot(data, aes(deltaBeta, -log10(FDR))) +
  geom_text_repel(aes(deltaBeta, -log10(FDR),
                      label = ifelse(-log10(FDR) > 60, rownames(data), ""))) + 
  geom_point(aes(color = feature)) +
  guides(colour = guide_legend(override.aes = list(size=5))) + # Modify legend size
  theme(legend.position = "none") + # remove legend
  labs(x = "Methylation difference (beta)",
       y = bquote(~-log[10]~(italic("FDR")))) + #参考了FigureYa59Volcano
  theme(axis.title.x = element_text(color="black", size = 14, face = "bold"),
        axis.title.y = element_text(color="black", size = 14, face = "bold"))
p1
# Remove background and gridlines
p1 <- p1 + theme_classic() +
  theme(#panel.background = element_rect(fill = NA),
    panel.border = element_rect(color = NA, fill = NA, size = 2),
    panel.grid.major.x = element_blank(),
    panel.grid.minor.x = element_blank(),
    panel.grid.minor.y = element_blank(),
    panel.grid.major.y = element_blank()) +
  scale_y_continuous(breaks = c(0, 2.5, 5, 7.5, 10),
                     labels = c(0, 2.5, 5, 7.5, ""),
                     limits = c(0, 10))
p1
```
## Drawing of the column chart `p2` on the right side of the scatter plot
```{r}
### get percentage information
Percent <- data[, -1] %>%
  tabyl(feature) %>%
  adorn_totals("row") %>%
  adorn_pct_formatting() %>%
  .[-9,]
Percent
# Get the maximum value of n of the count column in Percent
n_max = max(Percent$n)
n_max

### Construct data for drawing colored dots
# Note that the filling here should be the same as the previous order, that is to say, the color should correspond to the color in the scatter plot
Percent$A <- 1:8
test = data.frame(x = c(1, 2, 3, 4, 5, 6, 7,8),
                  y = rep(-n_max/15, 8),
                  type = LETTERS[1:8])
rect_data <- data.frame(
  xstart = c(0.55, 1.55, 2.55, 3.55, 4.55, 5.55, 6.55),
  xend   = c(1.45, 2.45, 3.45, 4.45, 5.45, 6.45, 7.45), 
  ystart = c(rep(-n_max/2, 7)),
  yend   = c(rep(0, 7)))
### draw
p2 <- ggplot(Percent, aes(A, n)) +
  geom_bar(stat = "identity", fill = "#BEBEBE", color = "black") + 
  geom_point(data = test, 
             aes(x, y, color = type), size = 4) +
  geom_text(aes(label = Percent$percent,
                hjust = ifelse(n > 1000, 2, -0.2)),
            #nudge_y = -120, 
            vjust = 0,
            fontface = "bold", size = 3) +
  geom_text(data = Percent,
            aes(A, -n_max/3, label = c("1stExon", "3'UTR", "5'UTR", "Body", "IGR", "TSS1500", "TSS200")), 
            color = "black", fontface = "bold",
            size = 3, hjust = 0.5) +
  geom_rect(data = rect_data, inherit.aes = FALSE,
            aes(xmin = xstart, xmax = xend,
                ymin = ystart, ymax = yend), 
            fill = "NA", color = "black") +
  theme_bw() +
  coord_flip() +
  scale_x_continuous(expand = c(0, 0)) +
  scale_y_continuous(expand = c(0, 0)) +
  theme(panel.background = element_rect(fill = NA),
        panel.border = element_blank(),
        panel.grid.major.x = element_blank(),
        panel.grid.minor.x = element_blank(),
        panel.grid.minor.y = element_blank(),
        panel.grid.major.y = element_blank(),
        axis.line.x = element_blank(),
        axis.ticks.x = element_blank(),
        axis.ticks.y = element_blank(),
        axis.text.y = element_blank(),
        axis.text.x = element_blank(),
        legend.position = "none"
  ) +
  guides(fill = FALSE) +
  xlab("") + ylab("") 
p2
```
## Drawing of top percentage graph `p3`
### Calculate the percentage
percent_Hypo <- sum(data$deltaBeta > 0) / nrow(data)
paste("Hypo was", round(percent_Hypo * 100, 2), "%" )
percent_Hyper <- sum(data$deltaBeta < 0) / nrow(data)
paste("Hyper was", round(percent_Hyper * 100, 2), "%" )
df1 <- data.frame(xmin = c(-350, 500, 1928, 4591),
                  xmax = c(500, 1928, 4591, 6500),
                  ymin = 00,
                  ymax = 100,
                  class  = c("A","B","C","A"),
                  text = c(paste(round(percent_Hyper * 100, 2), "%" ),  "", "", paste(round(percent_Hypo * 100, 2), "%" )))
df1
### draw
p3 <- ggplot(df1, aes(xmin= xmin, xmax = xmax, 
                      ymin= ymin , ymax= ymax, 
                      fill = class, label = text )) + 
  geom_rect(color = "black") +
  geom_text(aes(label = df1$text, x = xmin + 400, y = ymax - 50),
            position = position_dodge(0.9),
            vjust = 0, fontface = "bold", size = 3) +
  scale_fill_manual(values = c("#FFFFFF", "#4D4D4D", "#BEBEBE", "#FFFFFF")) +
  theme_bw() +
  scale_x_continuous(expand = c(0, 0)) +
  scale_y_continuous(expand = c(0, 0)) +
  
  theme(panel.background = element_rect(fill = NA),
        panel.border = element_rect(color = NA, fill = NA, size = 2),
        panel.grid.major.x = element_blank(),
        panel.grid.minor.x = element_blank(),
        panel.grid.minor.y = element_blank(),
        panel.grid.major.y = element_blank(),
        axis.line.x = element_blank(),
        axis.ticks.x = element_blank(),
        axis.ticks.y = element_blank(),
        axis.text.y = element_blank(),
        axis.text.x = element_blank()) +
  guides(fill = FALSE) +
  xlab("") + ylab("") 
p3
``
## Group picture

```{r,fig.width = 10,fig.height = 6}
combined_plot <- insert_xaxis_grob(p1, p3, 
                                   position = "top",
                                   height = grid::unit(0.4, "in") # Adjust the height ratio
) 

combined_plot <- insert_yaxis_grob(combined_plot, p2, 
                                   position = "right",
                                   width = grid::unit(2.5, "in")
)
# ggdraw(combined_plot) 

subtitle_theme_1 <- ggdraw() +
  draw_label("Hyermethylated DMPs          Hypomethylated DMPs",
             x = 0.1, y = 0.25,
             hjust = 0, vjust = 1.01, size = 10,  fontface = "bold")

p <- plot_grid(subtitle_theme_1, combined_plot, ncol = 1, 
               rel_heights = c(0.1, 1)) # Top percentage and scatter scale
p
# save to file
ggsave("multiVolcano.pdf")