Hello,

I am interested in making a KEGG pathway diagram to display the results of my KEGG pathway enrichment analysis (affymetrix zebrafish 1.1 ST microarray platform -> pd.zebgene.1.1.st).  Below is an example of the data returned after applying edgeR's kegga function.  Can anyone point me in the right direction as to how I can generate a colored pathway diagram (ex. Phototransduction - path:dre04744)?

Secondary to this, I would also like to generate a heatmap using all "up" and "down" genes for Glycolysis/Gluconeogenesis (path:dre00010).  If it were possible, I would also like to generate a heatmap using all "up" and "down" genes from multiple pathways (ex- path:dre00010 and path:dre00020). 

                                               Pathway  N Up Down         P.Up    P.Down
path:dre00010             Glycolysis / Gluconeogenesis 70 32    1 1.021135e-16 0.5755094
path:dre00020                Citrate cycle (TCA cycle) 29  4    0 2.262706e-01 1.0000000
path:dre00030                Pentose phosphate pathway 28 18    1 2.378557e-13 0.2896853
path:dre00040 Pentose and glucuronate interconversions 23 10    0 7.273981e-06 1.0000000
path:dre00051          Fructose and mannose metabolism 34 19    0 1.849783e-12 1.0000000

Here is the code I used to generate the pathway results:

setwd("C:\\Users\\mat149\\Desktop\\GG")
library(oligo)
CELlist <- list.celfiles("C:\\Users\\mat149\\Desktop\\GG\\CEL", full.names=TRUE, pattern=NULL, all.file=FALSE)
pdat<-read.AnnotatedDataFrame("C:\\Users\\mat149\\Desktop\\GG\\CEL\\phenoMOD.txt",header=TRUE,row.name="Filename",sep="\t")
CELdat <- read.celfiles(filenames = CELlist,experimentData=TRUE,phenoData=pdat,verbose=TRUE)

eset<-rma(CELdat, background=TRUE, normalize=TRUE, subset=NULL, target="core")     
library(affycoretools)
eset <- annotateEset(eset, annotation(eset))

library(org.Dr.eg.db)
fd <- fData(eset)
fd$ENTREZID <- mapIds(org.Dr.eg.db, as.character(fd$SYMBOL), "ENTREZID","SYMBOL",multiVals="first")
fData(eset) <- fd

ph = CELdat@phenoData 
ph@data[ ,1] = c("WT1","WT2","WT3","WT4","WT5","WT6","WT7","WT8","MO1","MO2","MO3","MO4","RS1","RS2","RS3","RS4")
ph@data[ ,2] = c("control","control","control","control","control","control","control","control","morphant","morphant","morphant","morphant","rescue","rescue","rescue","rescue")
colnames(ph@data)[2]="Treatment"
colnames(ph@data)[1]="Sample"
groups = ph@data$Treatment
f = factor(groups,levels=c("control","morphant","rescue"))

library(limma)
design = model.matrix(~ 0 + f)
colnames(design)=c("control","morphant","rescue")
data.fit = lmFit(eset,design)
contrast.matrix = makeContrasts(morphant-control,rescue-control,morphant-rescue,levels=design)
data.fit.con = contrasts.fit(data.fit,contrast.matrix)
data.fit.eb = eBayes(data.fit.con)

library(edgeR)

entzzvec<-as.vector(data.fit.eb$genes$ENTREZID)
MOkegg<-kegga(data.fit.eb,coef=1,geneid=entzzvec,FDR=0.05,species.KEGG="dre",convert=TRUE)
RSkegg<-kegga(data.fit.eb,coef=2,geneid=entzzvec,FDR=0.05,species.KEGG="dre",convert=TRUE)
MORSkegg<-kegga(data.fit.eb,coef=3,geneid=entzzvec,FDR=0.05,species.KEGG="dre",convert=TRUE)

Thanks,

Matt

When you say a "pathway diagram", I am guessing that you mean the well known wiring diagrams that have been produced over the years by the KEGG project, for example like this one:

http://www.genome.jp/dbget-bin/www_bget?dre04744

It is impossible for edgeR or limma to generate diagrams like this from raw data. Moreover, I think there would be copyright issues if we were to try to download these diagrams from KEGG, and we wouldn't be able to color them automatically anyway as they are just png pictures. I think if you want a coloured diagram like this, you would have to do it by hand. We did this ourselves for this paper:

https://www.ncbi.nlm.nih.gov/pubmed/26444724

see for example the supplementary figure:

http://iovs.arvojournals.org/data/Journals/IOVS/934564/i1552-5783-56-11-6438-s01.png