获取KEGG通路的基因列表 做单细胞GSEA、GSVA分析

   library(msigdbr)
    genesets = msigdbr(species = "Homo sapiens"  ) #msigdbr提供多个物种的基因集数据
    # View(msigdbr_collections()) #查看msigdbr包中所有的基因集
    unique(genesets$gs_subcat)  # 有多个数据库来源的基因集可选，这里选用KEGG
    genesets <- subset(genesets, gs_subcat=="CP:KEGG", select = c("gs_name", "gene_symbol"))
    unique(genesets$gs_name) #查看有多少条通路（186个）
   

方法二 使用 KEGGREST

#BiocManager::install("KEGGREST")
    #BiocManager::install("EnrichmentBrowser")
    library("KEGGREST")
    library("EnrichmentBrowser")
    KEGGREST:: listDatabases()
    KEGGREST::keggList(database ='kegg')
    keggList("organism") ## returns the list of KEGG organisms with
    #step2: check and obtain a list of entry identifiers (in this case: sar) and associated definition for a given database or a given set of database entries.
    res <- keggList("pathway", "hsa") ## returns the list of human pathways
    length(res)
    res=as.data.frame(res)
    head(res)
    #step 3: download the pathways of that organism:
    hsapathway <- downloadPathways("hsa")
    head(hsapathway)
    idTypes(org = 'hsa' )
    #step 4: retrieve gene sets for an organism from databases such as GO and KEGG:
    hsa_kegg_genesets <- getGenesets(org = "hsa", db = "kegg",
                                     gene.id.type = "SYMBOL",
                                     cache = TRUE, return.type="list")
    
    #step5: Parse and write the gene sets to a flat text file in GMT format for other pathway enrichment analysis programs (e.g., GSEA):
    writeGMT(hsa_kegg_genesets, gmt.file = "kegg_hsa_kegg_genesets_gmt")
    save(hsa_kegg_genesets,file = "~/heart_muscle/hsa_kegg_genesets.rds")

 library(GSVA);print(getwd())  load("~/heart_muscle/hsa_kegg_genesets.rds")  genesets_kegg=hsa_kegg_genesets  print(length(hsa_kegg_genesets))  #kegg---  gssea.res <- gsva(expr, genesets_kegg [1:50], method="ssgsea",kcdf="Poisson",min.sz > 3,max.sz=200 ,parallel.sz=10 )   saveRDS(gssea.res, paste0(new_dir,file_name,"_gssea.res_kegg.rds"  )  )    gssea.df <- data.frame(Genesets=rownames(gssea.res),gssea.res, check.names = F)  write.csv(gssea.df,  paste0(new_dir,file_name,"gssea_res_kegg.csv" ), row.names = F)        # ssgsea-----  #library(GSVA);print(getwd())  #gssea.res <- gsva(expr, genesets_GO [1:50], method="ssgsea",kcdf="Poisson",min.sz > 3,max.sz=200 ,parallel.sz=10 )  #,parallel.sz=10  #saveRDS(gssea.res, paste0(new_dir,file_name,"_gssea.res_go.rds"  )  )    #gssea.df <- data.frame(Genesets=rownames(gssea.res), gssea.res, check.names = F)  #write.csv(gssea.df,  paste0(new_dir,file_name,"gssea_res_go.csv"), row.names = F)   #print("done-------");print(getwd())​

参考：https://biobeat.wordpress.com/category/r/https://www.researchgate.net/post/How_i_can_get_a_list_of_KEGG_pathways_and_its_list_of_genes