---
title: "09.GSEA"
output: html_document
date: '2022-07-18'
---


#### GSEA analysis ####
```{r}
options(stringsAsFactors = F)
library(org.Hs.eg.db)
library(clusterProfiler)
library(enrichplot)
library(tidyverse)
library(ggstatsplot)
library(dplyr)
library(plyr)
```

```{r,create genelist for GSEA}
organism = 'hsa'  
OrgDb = 'org.Hs.eg.db'


need_DEG <- res_tbl[,c(1,3,6)] 

colnames(need_DEG) <- c('gene','log2FoldChange','pvalue')
need_DEG$SYMBOL <- need_DEG$gene


#convert id  
df <- bitr(need_DEG$SYMBOL, 
           fromType = "SYMBOL",
           toType =  "ENTREZID",
           OrgDb = OrgDb)
need_DEG <- merge(need_DEG, df, by='SYMBOL')  
#write.csv(need_DEG,'need_DEG.csv')
geneList <- need_DEG$log2FoldChange
names(geneList) <- need_DEG$ENTREZID
geneList <- sort(geneList, decreasing = T)   #descending order
```

```{r, enrichement by clusterprofiler}
KEGG_kk_entrez <- gseKEGG(geneList = geneList,
                   organism     = organism,
                   pvalueCutoff = 0.05)  
KEGG_kk <- DOSE::setReadable(KEGG_kk_entrez, 
                             OrgDb=OrgDb,
                             keyType='ENTREZID')#convert id             
  

#save(KEGG_kk_entrez, file = "./data/epithelium/by_cancer_by_pheno/BC/GSEA_KEGG_results.RData")
#GSEA Hall markers
##define the TERM2GENE list from msigdb.  #Homo sapiens or Mus musculus
library(msigdbr)
library(ExperimentHub)
library(GSEABase)

C2_t2g <- msigdbr(species = "Homo sapiens",#Homo sapiens or Mus musculus
                  category = "C5"#, subcategory = "BP"
                  ) %>% 
  dplyr::select(gs_name, entrez_gene)
head(C2_t2g)

C2_kk_entrez <- GSEA(geneList, TERM2GENE = C2_t2g, pAdjustMethod = "BH", pvalueCutoff = 0.05
                     )
C2_kk <- DOSE::setReadable(C2_kk_entrez, 
                             OrgDb=OrgDb,
                             keyType='ENTREZID')
head(C2_kk)
head(C2_kk@result)
#write.csv(C2_kk@result, '~/Desktop/C5_0.05_GSEA.csv')
```