Hi,
I am doing the SPIA analysis via **graphite package** for I could do the SPIA not just on KEGG pathway but on others such as reactome, nci etc.
So, one could use in graphite
prepareSPIA(db,"user_defined_string") (where, db can be KEGG, reactome, nci etc)
res <-runSPIA(de=my_DE, all=my_ALL, "user_defined_string")
head(res) ...
Name pSize NDE pNDE tA pPERT pG pGFdr pGFWER Status
1 Circadian rhythm 28 17 5.979811e-05 10.42313596 0.000005 6.856035e-09 1.028405e-06 1.028405e-06 Activated
2 ECM-receptor interaction 73 32 2.996008e-04 24.56547212 0.000005 3.193615e-08 2.395211e-06 4.790422e-06 Activated
-
Is there a way to extract the DE gene list using graphite/runSPIA ? For instance, how to list all the 17 (NDE) gene ids for circadian rhythm in the above example ?
Any suggestions/idea ?
Thanks in advance.
Ramya Gamini
Hi Ramya,
prepareSPIA creates a R data file, "keggexSPIA.RData", in the current working directory and the file contains a list with all the pathways to be tested.
prepareSPIA(kegg, "keggex")
You can load that file:
load("keggexSPIA.RData")
and see the vignette of SPIA how to extract information from it. To get the Entrez IDs for instance you can use:
entids= sapplypath.info,"[","nodes")
names(entids)=gsub(".nodes","",names(entids))
# Entrez ids for the "Wnt signaling pathway"
entids[['Wnt signaling pathway']]
These ids may need to be intersected with the list of all genes on the array to get the hits per pathway used in the analysis.
Adi Tarca