07_KEGG.Rmd

# KEGG enrichment analysis {#clusterprofiler-kegg}


```{r include=FALSE}
library(knitr)
opts_chunk$set(message=FALSE, warning=FALSE, eval=TRUE, echo=TRUE, cache=TRUE)
library(clusterProfiler)
```

The KEGG FTP service is not freely available for academic use since 2012, and there are many software packages using out-dated KEGG annotation data. The `r Biocpkg("clusterProfiler")` package supports downloading the latest online version of KEGG data using the [KEGG website](https://www.kegg.jp), which is freely available for academic users. Both the KEGG pathway and module are supported in `r Biocpkg("clusterProfiler")`.


## Supported organisms {#clusterProfiler-kegg-supported-organisms}

The `r Biocpkg("clusterProfiler")` package supports all organisms that have KEGG annotation data available in the KEGG database. Users should pass an abbreviation of academic name to the `organism` parameter. The full list of KEGG supported organisms can be accessed via <http://www.genome.jp/kegg/catalog/org_list.html>. [KEGG Orthology](https://www.genome.jp/kegg/ko.html) (KO) Database is also supported by specifying `organism = "ko"`.


The `r Biocpkg("clusterProfiler")` package provides `search_kegg_organism()` function to help searching supported organisms.

```{r}
library(clusterProfiler)
search_kegg_organism('ece', by='kegg_code')
ecoli <- search_kegg_organism('Escherichia coli', by='scientific_name')
dim(ecoli)
head(ecoli)
```


## KEGG pathway over-representation analysis {#clusterprofiler-kegg-pathway-ora}


```{r}
data(geneList, package="DOSE")
gene <- names(geneList)[abs(geneList) > 2]

kk <- enrichKEGG(gene         = gene,
                 organism     = 'hsa',
                 pvalueCutoff = 0.05)
head(kk)
```

Input ID type can be `kegg`, `ncbi-geneid`, `ncbi-proteinid` or `uniprot` (see also [session 16.1.2](#bitr_kegg)). Unlike `enrichGO()`, there is no `readable` parameter for `enrichKEGG()`. However, users can use the [`setReadable()` function](#setReadable) if there is an `OrgDb` available for the species.


## KEGG pathway gene set enrichment analysis {#clusterprofiler-kegg-pathway-gsea}

```{r}
kk2 <- gseKEGG(geneList     = geneList,
               organism     = 'hsa',
               minGSSize    = 120,
               pvalueCutoff = 0.05,
               verbose      = FALSE)
head(kk2)
```



## KEGG module over-representation analysis {#clusterprofiler-kegg-module-ora}


[KEGG Module](http://www.genome.jp/kegg/module.html) is a collection of manually defined function units. In some situation, KEGG Modules have a more straightforward interpretation.

```{r}
mkk <- enrichMKEGG(gene = gene,
                   organism = 'hsa',
                   pvalueCutoff = 1,
                   qvalueCutoff = 1)
head(mkk)                   
```

## KEGG module gene set enrichment analysis {#clusterprofiler-kegg-module-gsea}


```{r}
mkk2 <- gseMKEGG(geneList = geneList,
                 organism = 'hsa',
                 pvalueCutoff = 1)
head(mkk2)
```


## Visualize enriched KEGG pathways

The `r Biocpkg("enrichplot")` package implements [several methods](#enrichplot) to visualize enriched terms. Most of them are general methods that can be used on GO, KEGG, MSigDb, and other gene set annotations. Here, we introduce the `clusterProfiler::browseKEGG()` and `pathview::pathview()` functions to help users explore enriched KEGG pathways with genes of interest.


To view the KEGG pathway, users can use the `browseKEGG` function, which will open a web browser and highlight enriched genes.


```{r eval=FALSE}
browseKEGG(kk, 'hsa04110')
```



(ref:browseKEGGscap) Explore selected KEGG pathway.

(ref:browseKEGGcap) **Explore selected KEGG pathway.** Differentially expressed genes that are enriched in the selected pathway will be highlighted.

```{r browseKEGG, out.width="100%", echo=FALSE, fig.cap="(ref:browseKEGGcap)", fig.scap="(ref:browseKEGGscap)"}
knitr::include_graphics("figures/browseKEGG.png")
```



Users can also use the `pathview()` function from the `r Biocpkg("pathview")` [@luo_pathview] to visualize enriched KEGG pathways identified by the `r Biocpkg("clusterProfiler")` package [@yu2012].

The following example illustrates how to visualize the "hsa04110" pathway, which was enriched in our previous analysis.

```{r eval=FALSE}
library("pathview")
hsa04110 <- pathview(gene.data  = geneList,
                     pathway.id = "hsa04110",
                     species    = "hsa",
                     limit      = list(gene=max(abs(geneList)), cpd=1))
```

(ref:pathviewscap) Visualze selected KEGG pathway by `pathview()`.

(ref:pathviewcap) **Visualize selected KEGG pathway by `pathview()`.** Gene expression values can be mapped to gradient color scale.

```{r pathview, out.width="100%", echo=FALSE, fig.cap="(ref:pathviewcap)", fig.scap="(ref:pathviewscap)"}
knitr::include_graphics("figures/hsa04110_pathview.png")
```