clusterProfiler优雅的处理KEGG数据

最近在系统学习clusterProfiler软件包，越用越觉得Y叔的强大，本节来介绍几个实用的小功能

k转ko

bitr_kegg可以对多个ko号进行转换此处传入字符向量即可

k <- c("K00799","K00844")
x <- bitr_kegg(k,"kegg","Path","ko") 

> x
     kegg    Path
1  K00844 ko00010
2  K00844 ko00051
3  K00844 ko00052
4  K00799 ko00480
5  K00844 ko00500
6  K00844 ko00520
7  K00844 ko00521

翻译ko

y <- ko2name(x$Path)

        ko                                              name
1  ko00010                      Glycolysis / Gluconeogenesis
2  ko00051                   Fructose and mannose metabolism
3  ko00052                              Galactose metabolism
4  ko00480                            Glutathione metabolism
5  ko00500                     Starch and sucrose metabolism
6  ko00520       Amino sugar and nucleotide sugar metabolism

left_join(x,y,by=c("Path"="ko"))

     kegg    Path                                              name
1  K00844 ko00010                      Glycolysis / Gluconeogenesis
2  K00844 ko00051                   Fructose and mannose metabolism
3  K00844 ko00052                              Galactose metabolism
4  K00799 ko00480                            Glutathione metabolism
5  K00844 ko00500                     Starch and sucrose metabolism
6  K00844 ko00520       Amino sugar and nucleotide sugar metabolism
7  K00844 ko00521                         Streptomycin biosynthesis
8  K00844 ko00524   Neomycin, kanamycin and gentamicin biosynthesis

下载全部的k号对应的pathway信息

通过阅读clusterProfiler源代码发现有kegg_link此函数，源代码如下

## http://www.genome.jp/kegg/rest/keggapi.html
## kegg_link('hsa', 'pathway')
kegg_link <- function(target_db, source_db) {
    url“ ”） <- paste0("http://rest.kegg.jp/link/", target_db, "/", source_db, collapse="")
    kegg_rest(url)
}

clusterProfiler:::kegg_link("ko","pathway")

          from        to
1 path:map00010 ko:K00001
2  path:ko00010 ko:K00001
3 path:map00010 ko:K00002
4  path:ko00010 ko:K00002
5 path:map00010 ko:K00016
6  path:ko00010 ko:K00016

kegg_list <- function(db) {
    url <- paste0("http://rest.kegg.jp/list/", db, collapse="")
    kegg_rest(url)
}

clusterProfiler:::kegg_list("pathway")

clusterProfiler:::kegg_list("ko")

列出与人类相关的通路

clusterProfiler:::kegg_link("hsa","pathway")

           from        to
1 path:hsa00010 hsa:10327
2 path:hsa00010   hsa:124
3 path:hsa00010   hsa:125
4 path:hsa00010   hsa:126
5 path:hsa00010   hsa:127
6 path:hsa00010   hsa:128

ID转换

id <- c("4597","7111")
eg2np <- bitr_kegg(id,fromType='kegg',toType='ncbi-proteinid',organism='hsa')

  kegg ncbi-proteinid
1 4597      NP_002452
2 7111   NP_001159588

np2up <- bitr_kegg(eg2np[,2], fromType='ncbi-proteinid',toType='uniprot',organism='hsa')

  ncbi-proteinid uniprot
1      NP_002452  P53602
2   NP_001159588  P28289

参考：https://guangchuangyu.github.io/2016/05/convert-biological-id-with-kegg-api-using-clusterprofiler/
https://guangchuangyu.github.io/cn/2017/09/ko-id-conversion/
https://github.com/YuLab-SMU/clusterProfiler