Question: odd results "No enrichment found in any of gene cluster"
gravatar for joseph
6 months ago by
joseph40 wrote:

I have two gene list as below, one of them is able to obtain an enriched KEGG pathway, the other one cannot.

I consolidate them to one list, and carry on the KEGG enrichment, none of pathway has been found.

The link of .Rdata of these lists, please use list, list_edge, list_limma to access.


list_limma = list(doublePos_vs_singlePos_limma = DEG_doublePos_vs_singlePos_limma$ENTREZID,
                  doublePos_vs_singlePos_limma_Up = DEG_doublePos_vs_singlePos_limma_Up$ENTREZID,
                  doublePos_vs_singlePos_limma_Down  = DEG_doublePos_vs_singlePos_limma_Down$ENTREZID,
                  doublePos_vs_doubleNeg_limma = DEG_doublePos_vs_doubleNeg_limma$ENTREZID,
                  doublePos_vs_doubleNeg_limma_Up = DEG_doublePos_vs_doubleNeg_limma_Up$ENTREZID,
                  doublePos_vs_doubleNeg_limma_Down = DEG_doublePos_vs_doubleNeg_limma_Down$ENTREZID)

list_edge = list(doublePos_vs_singlePos_edge = DEG_doublePos_vs_singlePos_edge$table$ENTREZID,
                 doublePos_vs_singlePos_edge_Up = DEG_doublePos_vs_singlePos_edge_Up$table$ENTREZID,
                 doublePos_vs_singlePos_edge_Down = DEG_doublePos_vs_singlePos_edge_Down$table$ENTREZID,
                 doublePos_vs_doubleNeg_edge = DEG_doublePos_vs_doubleNeg_edge$table$ENTREZID,
                 doublePos_vs_doubleNeg_edge_Up = DEG_doublePos_vs_doubleNeg_edge_Up$table$ENTREZID,
                 doublePos_vs_doubleNeg_edge_Down = DEG_doublePos_vs_doubleNeg_edge_Down$table$ENTREZID)

levels(list_limma) = c("doublePos_vs_singlePos","doublePos_vs_singlePos_Up","doublePos_vs_singlePos_Down",

levels(list_edge) = c("doublePos_vs_singlePos","doublePos_vs_singlePos_Up","doublePos_vs_singlePos_Down",

list = list(doublePos_vs_singlePos = c(DEG_doublePos_vs_singlePos_limma$ENTREZID,DEG_doublePos_vs_singlePos_edge$table$ENTREZID),
            doublePos_vs_singlePos_Up = c(DEG_doublePos_vs_singlePos_limma_Up$ENTREZID,DEG_doublePos_vs_singlePos_edge_Up$table$ENTREZID),
            doublePos_vs_singlePos_Down  = c(DEG_doublePos_vs_singlePos_limma_Down$ENTREZID,DEG_doublePos_vs_singlePos_edge_Down$table$ENTREZID),
            doublePos_vs_doubleNeg = c(DEG_doublePos_vs_doubleNeg_limma$ENTREZID,DEG_doublePos_vs_doubleNeg_edge$table$ENTREZID),
            doublePos_vs_doubleNeg_Up = c(DEG_doublePos_vs_doubleNeg_limma_Up$ENTREZID,DEG_doublePos_vs_doubleNeg_edge_Up$table$ENTREZID),
            doublePos_vs_doubleNeg_Down = c(DEG_doublePos_vs_doubleNeg_limma_Down$ENTREZID,DEG_doublePos_vs_doubleNeg_edge_Down$table$ENTREZID))

levels(list) = c("doublePos_vs_singlePos","doublePos_vs_singlePos_Up","doublePos_vs_singlePos_Down",


CompKEGG_limma <- compareCluster(geneCluster = list_limma, fun = "enrichKEGG",organism="hsa", pvalueCutoff=1)

CompKEGG_edge <- compareCluster(geneCluster = list_edge, fun = "enrichKEGG",organism="hsa", pvalueCutoff=1)

CompKEGG <- compareCluster(geneCluster = list, fun = "enrichKEGG",organism="hsa", pvalueCutoff=1)


ADD COMMENTlink modified 6 months ago by Guangchuang Yu870 • written 6 months ago by joseph40
gravatar for Guangchuang Yu
6 months ago by
Hong Kong
Guangchuang Yu870 wrote:

your code is not reproducible.

pls see also C: Subset of genes can find a term but whole set cannot.

ADD COMMENTlink written 6 months ago by Guangchuang Yu870

I can understand the principle of gene enrichment. I just so surprised about nonpathway can be matched because the list is approx 1000+, even without significance (I set pvalueCutoff = 1), should have some gene match to KEGG pathway. I doubt if the list feature causes the problem? I check several times, not conclude, so have to raise the question.

What's the error do you get when you reproduce my code?

ADD REPLYlink modified 6 months ago • written 6 months ago by joseph40

> CompKEGG <- compareCluster(geneCluster = list, fun = "enrichKEGG",organism="hsa", pvalueCutoff=1)
Error in compareCluster(geneCluster = list, fun = "enrichKEGG", organism = "hsa",  :
  No enrichment found in any of gene cluster, please check your input...
> CompKEGG <- compareCluster(geneCluster = list, fun = "enrichKEGG",organism="hsa", pvalueCutoff=1, qvalueCutoff=1)
> head(CompKEGG, 2)
                 Cluster       ID
1 doublePos_vs_singlePos hsa04141
2 doublePos_vs_singlePos hsa05120
                                                 Description GeneRatio  BgRatio
1                Protein processing in endoplasmic reticulum    25/626 166/7271
2 Epithelial cell signaling in Helicobacter pylori infection    13/626  68/7271
       pvalue  p.adjust    qvalue
1 0.003948754 0.6872773 0.6813311
2 0.004707379 0.6872773 0.6813311
1 10802/3320/29979/22926/23645/10483/6746/27102/10484/824/9373/6747/9532/7353/23471/51726/7466/9978/54788/64215/5611/10273/80343/56886/80700
2                                                                           102/5803/3576/245973/8992/1445/5336/83700/3551/526/523/5058/1956
1    25
2    13

ADD REPLYlink written 6 months ago by Guangchuang Yu870

In the compareCluster documents,  I can't find the default qvalueCutoff setting. In the above situation, no enrichment can be found when pvalueCutoff = 1, but 2 enrichment show up when set q = 1. I suppose the default q had been set less than 1. Is that right?

ADD REPLYlink written 6 months ago by joseph40

yes, and the parameter list of compareCluster depends on the fun parameter. In your case, all enrichKEGG parameters can be passed to compareCluster.

ADD REPLYlink modified 6 months ago • written 6 months ago by Guangchuang Yu870
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.2.0
Traffic: 297 users visited in the last hour