[BioC] Visualizing KEGG pathways

[Marc Carlson]

Hi Tim,

For humans, KEGG uses entrez gene IDs for gene identifiers (they just
preceed them with "hsa:").  So we can extract those by just doing this:

egList = gsub("hsa:", "", glist)

##And then do this to see the gene symbols:
library(org.Hs.eg.db)
res = mget(as.character(egList), org.Hs.egSYMBOL, ifnotfound=NA) 

The org.Hs.eg.db library can also help you find out a lot of other
things about your genes.  And, when you look at those gene symbols I
think you will see the answer to your second question.  As an example,
the picture lists a single box with "stat" in it, but you can see from
your list of gene symbols that you have about 7 different kinds of stat
genes.  Also, from the kegg image you will see the many genes listed if
you just click on those green boxes (but you have to scroll down a bit
to see them).


  Marc




Tim Smith wrote:
> Hi,
>
> I was trying to explore KEGG pathways. I have two questions:
>
> Question 1:
> -------------
>
>  I used some code given by Saroj:
>
> #---------------------------
>    
> library("KEGGSOAP")
>
> # Suppose you are looking at the pathway for glycolysis / gluconeogenesis
> pathstr = "path:hsa00010"
>
> # get all genes
> glist = get.genes.by.pathway(pathstr)
>
> # First fix what color you want for which node
> # Suppose, you want to color first enzyme hsa:10327 (EC 1.1.1.2)
> # white on blue background and rest all nodes in black on white
> # fgcols = color of text and border
> # bgcols = color of the rectangular area
>
> fgcols = rep("black", length(glist))
> fgcols[1] = "black"
> bgcols = rep("#e6e6fa", length(glist))
> bgcols[1] = "blue"
>
> # color the pathway accordingly
> url = color.pathway.by.objects(pathstr, glist, fg=fgcols, bg=bgcols)
> browseURL(url)
>
> #---------------------------
>
> I get the following graph (with one of the nodes shown as blue) at "http://soap.genome.jp/tmp/mark_pathway_www3_api14519/hsa00010.gif"
>
> The original Glycolysis pathway in KEGG is given at "http://www.genome.jp/kegg/pathway/map/map00010.html"
>
> My question is whether I can get the actual gene name (in this case 'AKR1A1') instead of '1.1.1.2' on the node that is given in the graph - i.e., how can I get the gene names (for 'hsa') on the graph ?
>
>
> Question 2:
> ------------
>
> This is not exactly an R question, so please excuse me - I'm just trying to understand things in KEGG. I used the following code to get the genes for Jak-stat pathway:
>
> #------------
> pathstr = "path:hsa04630"
>
> # get all genes
> glist = get.genes.by.pathway(pathstr)
>
> #------------------
>   
>> length(glist)
>>     
> [1] 155
>
> This lists 155 genes as being associated with the pathway. But if I go to the KEGG graph ("http://www.genome.jp/kegg/pathway/hsa/hsa04630.html") I just see perhaps 20 nodes/genes. Where is my disconnect?
>
> Thanks for the replies!
>
>
>       
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
>
>