1 target_gene_id <- unique(read.delim("miRNA-gene interactions.txt")$EntrezID)
 2 # BiocInstaller::biocLite("clusterProfiler")
 3 # BiocInstaller::biocLite("org.Hs.eg.db")
 4 
 5 display_number = c(15, 10, 15)
 6 ## GO enrichment with clusterProfiler
 7 library(clusterProfiler)
 8 ego_MF <- enrichGO(OrgDb="org.Hs.eg.db",
 9              gene = target_gene_id,
10              pvalueCutoff = 0.05,
11              ont = "MF",
12              readable=TRUE)
13 ego_result_MF <- as.data.frame(ego_MF)[1:display_number[1], ]
14 # ego_result_MF <- ego_result_MF[order(ego_result_MF$Count),]
15 
16 ego_CC <- enrichGO(OrgDb="org.Hs.eg.db",
17                    gene = target_gene_id,
18                    pvalueCutoff = 0.05,
19                    ont = "CC",
20                    readable=TRUE)
21 ego_result_CC <- as.data.frame(ego_CC)[1:display_number[2], ]
22 # ego_result_CC <- ego_result_CC[order(ego_result_CC$Count),]
23 
24 ego_BP <- enrichGO(OrgDb="org.Hs.eg.db",
25                    gene = target_gene_id,
26                    pvalueCutoff = 0.05,
27                    ont = "BP",
28                    readable=TRUE)
29 ego_result_BP <- na.omit(as.data.frame(ego_BP)[1:display_number[3], ])
30 # ego_result_BP <- ego_result_BP[order(ego_result_BP$Count),]
31 
32 go_enrich_df <- data.frame(ID=c(ego_result_BP$ID, ego_result_CC$ID, ego_result_MF$ID),
33                                    Description=c(ego_result_BP$Description, ego_result_CC$Description, ego_result_MF$Description),
34                                    GeneNumber=c(ego_result_BP$Count, ego_result_CC$Count, ego_result_MF$Count),
35                                    type=factor(c(rep("biological process", display_number[1]), rep("cellular component", display_number[2]),
36                                           rep("molecular function", display_number[3])), levels=c("molecular function", "cellular component", "biological process")))
37 
38 ## numbers as data on x axis
39 go_enrich_df$number <- factor(rev(1:nrow(go_enrich_df)))
40 ## shorten the names of GO terms
41 shorten_names <- function(x, n_word=4, n_char=40){
42   if (length(strsplit(x, " ")[[1]]) > n_word || (nchar(x) > 40))
43   {
44     if (nchar(x) > 40) x <- substr(x, 1, 40)
45     x <- paste(paste(strsplit(x, " ")[[1]][1:min(length(strsplit(x," ")[[1]]), n_word)],
46                        collapse=" "), "...", sep="")
47     return(x)
48   } 
49   else
50   {
51     return(x)
52   }
53 }
54 
55 labels=(sapply(
56   levels(go_enrich_df$Description)[as.numeric(go_enrich_df$Description)],
57   shorten_names))
58 names(labels) = rev(1:nrow(go_enrich_df))
59 
60 ## colors for bar // green, blue, orange
61 CPCOLS <- c("#8DA1CB", "#FD8D62", "#66C3A5")
62 library(ggplot2)
63 p <- ggplot(data=go_enrich_df, aes(x=number, y=GeneNumber, fill=type)) +
64   geom_bar(stat="identity", width=0.8) + coord_flip() + 
65   scale_fill_manual(values = CPCOLS) + theme_bw() + 
66   scale_x_discrete(labels=labels) +
67   xlab("GO term") + 
68   theme(axis.text=element_text(face = "bold", color="gray50")) +
69   labs(title = "The Most Enriched GO Terms")
70 
71 p
72 
73 pdf("go_enrichment_of_miRNA_targets.pdf")
74 p
75 dev.off()
76 
77 svg("go_enrichment_of_miRNA_targets.svg")
78 p
79 dev.off()