How to get KEGG pathway names in KEGG enrichment analysis using clusterprofiler

Question

[tianshenbio]

I am doing KEGG enrichment analysis for a non-model organism. First I annotate the genome using KAAS and I got ko numbers associated with each gene (kegg2gene)

K11968  BANY.1.2.t00001
K04834  BANY.1.2.t00003
K13273  BANY.1.2.t00007
K04497  BANY.1.2.t00010
K09210  BANY.1.2.t00011
K10360  BANY.1.2.t00012
K10360  BANY.1.2.t00013

Then I have a list of DE genes (gene) and performed KEGG enrichment using clusterprofiler:

x <- enricher(gene,TERM2GENE=kegg2gene,pvalueCutoff = 0.05, pAdjustMethod = "BH",qvalueCutoff = 0.05)

Then I got a list of enriched ko numbers, then how do I translate the ko numbers to pathway names/descriptions? I know I can add TERM2NAME=kegg2name, but how do I get a list of all kegg numbers vs names?

[ATpoint]

https://www.genome.jp/kegg/ko.html

Paste them into the box on that page and translate to pathways.

[tianshenbio]

Hi, Since I hope to do pathway enrichment analysis, I need to get a list of genes vs pathways. Now I have genes vs K numbers so yes, I need to translate K numbers to pathways. But I can't download a list of K number vs pathways from the mapper page u shared....

[N15]

You don't need to translate them to use clusterProfiler, you can use KOs directly. set organism = "ko"

https://github.com/YuLab-SMU/clusterProfiler/issues/99

[tianshenbio]

Hi NRC

I read that post from github but only enricherKEGG has the organism option right? As I mentioned I used the enricher function not enrichKEGG since I am using a customized dataset. You can also have a look at the last comment from the link you shared, that's written by me.

[N15]

I am not sure about enricherKEGG, I am only familiar with enrichKEGG. You can use a list of KEGG IDs directly, e.g. K00011, K00678, etc. You set organisms="ko" and that's it, this definitely works for non-model organisms because all you need is a list of KO values

[tianshenbio]

Hi NRC Thank you for your reply. Sorry, I meant enrichKEGG. First of all, I have a list of all kegg2gene for my species: kegg2gene

K11968  BANY.1.2.t00001
K04834  BANY.1.2.t00003
K13273  BANY.1.2.t00007
K04497  BANY.1.2.t00010
K09210  BANY.1.2.t00011
K10360  BANY.1.2.t00012
K10360  BANY.1.2.t00013
K10360  BANY.1.2.t00015
K01530  BANY.1.2.t00016

Then I have a list of DE genes: gene

BANY.1.2.t20473
BANY.1.2.t12787
BANY.1.2.t10473
BANY.1.2.t10472
BANY.1.2.t08098

How do I perform the kegg enrichment analysis?

I tried this:

x <- enrichKEGG(gene,organism="ko",keyType="kegg",pvalueCutoff = 0.05, pAdjustMethod = "BH",kegg2gene,qvalueCutoff = 0.05)

and I got an error message:

--> No gene can be mapped....
--> Expected input gene ID: K01213,K01623,K00846,K00875,K00496,K18106
--> return NULL...

[N15]

Match your list of DE genes to their KO value, then try again.

x <- enrichKEGG(gene, organism='ko', keyType='kegg', universe = kegg2gene, pSjustMethod="BH")

[tianshenbio]

Hi, I am not sure if I understood correctly. I tried to map my DE genes with K numbers and used the K number only, now gene (a string of characters) is like:

[1] K00901 K01016 K06483 K06585 K00799

Also, only the K numbers in kegg2gene are used, now kegg2gene looks like:

[1] K04834 K13273 K04497 K09210 K10360

x <- enrichKEGG(gene,organism="ko",keyType="kegg",universe="kegg2gene", pAdjustMethod = "BH")

This gave me

No gene set have size > 10 ...
--> return NULL...

[N15]

hmm what if you take out the p adjusted cutoff?

[tianshenbio]

I tried to adjust p and minGSSize but did not work...I have 300 terms in gene and 9000 terms in kegg2go, should be enough for the analysis...Anyway, thank you for your help!NRC

[fanimata]

Hi! How you solved your problem using enrichKEGG( ) function? I have the same error message:

ca_kegg <- enrichKEGG(ca_list, organism = 'ko', keyType = 'kegg', universe = BBRB_KEGG, pAdjustMethod = "BH")

---> No gene can be mapped....
---> Expected input gene ID: K00895,K01810,K21622,K16370,K15779,K01218
---> return NULL...

ca_list is my list of DE gene ID's and BBRB_KEGG is a dataframe of two columns with gene ID's and KEGG annotations that I get with Trinotate.

If you can help me I'll really apreciate!

Answer 1

[Guangchuang Yu]

it works. see also https://doi.org/10.1016/j.xinn.2021.100141.