What genesets can I use for pathway analysis of honeybee seq data?

[wmseq]

bigmw,
Thanks a lot for your quick answer!
I have solved the problem, because GeneID is in column2.

>names(edger.fc) = test$GeneID
> head(exp.fc)
409677 100576979 100577819 552035 413550 413908
5.557823 4.667221 4.516693 4.127615 3.986429 3.937341

The first line is GeneID, and the second line is logFC.

Now I met another problem that the geneset file of kegg.gs could not be used for honey bee, because I could not run the following command:
> pv.out.list <- sapply(path.ids2[1:3], function(pid) pathview(gene.data = exp.fc, pathway.id = pid, species ="ame", out.suffix = out.suffix))

Then I checked the fc.kegg.p$greater, I got the following results.

> head(fc.kegg.p$greater)


hsa00010 Glycolysis / Gluconeogenesis NA NaN NA
hsa00020 Citrate cycle (TCA cycle) NA NaN NA
hsa00030 Pentose phosphate pathway NA NaN NA
hsa00040 Pentose and glucuronate interconversions NA NaN NA
hsa00051 Fructose and mannose metabolism NA NaN NA
hsa00052 Galactose metabolism NA NaN NA
q.val set.size exp1
hsa00010 Glycolysis / Gluconeogenesis NA 0 NA
hsa00020 Citrate cycle (TCA cycle) NA 0 NA
hsa00030 Pentose phosphate pathway NA 0 NA
hsa00040 Pentose and glucuronate interconversions NA 0 NA
hsa00051 Fructose and mannose metabolism NA 0 NA
hsa00052 Galactose metabolism NA 0 NA

The kegg.gs file seems for anaysis of human genes.

What genesets files can I use for honeybee genes, and If there is not any available, How can I create one by myself from the kegg pathway files?

Have a good weekend!

Richard

[bigmw]

gage package has a function, kegg.gsets, that generates up-to-date pathway gene sets for essentially all species (~2300 at this time) supported by KEGG. For details:
library(gage)
?kegg.gsets

Because honey bee is a KEGG species, you can use the function directly.
kg.ame=kegg.gsets("ame")
names(kg.ame)
lapply(kg.ame, head, 3)
#save the gene set data for future use
save(kg.ame, file="kg.ame.RData")

Usually, you want to use the signaling pathways and metabolic pathways in your analysis. So extract those pathways for your analysis:
kegg.gs=kg.ame$kg.sets[kg.ame$sigmet.idx]

now you can call gage with your data like:
fc.kegg.p <- gage(exp.fc, gsets = kegg.gs, ref = NULL, samp = NULL)

[wmseq]

Thank you vary much, bigmw!
Another question:
If I'd like to use all the pathways or a special one, how to extract them or it for my analysis?

[bigmw]

Read the documentation of kegg.gsets function, particular the Value section:
?kegg.gsets

Check the structure and elements of the actual gene set list, i.e. kg.ame:
str(kg.ame)
lapply(kg.ame, head, 3)

[wmseq]

Hi bigmw,
Why I could not find the function of "kegg.gsets" in gage package?

> library(gage)
> ?kegg.gsets
No documentation for ‘kegg.gsets’ in specified packages and libraries:
you could try ‘??kegg.gsets’

Thanks a lot!!

Richard

[rghan]

gage package has a function, kegg.gsets, that generates up-to-date pathway gene sets for essentially all species (~2300 at this time) supported by KEGG. For details:
library(gage)
?kegg.gsets

Because honey bee is a KEGG species, you can use the function directly.
kg.ame=kegg.gsets("ame")
names(kg.ame)
lapply(kg.ame, head, 3)
#save the gene set data for future use
save(kg.ame, file="kg.ame.RData")

Usually, you want to use the signaling pathways and metabolic pathways in your analysis. So extract those pathways for your analysis:
kegg.gs=kg.ame$kg.sets[kg.ame$sigmet.idx]

now you can call gage with your data like:
fc.kegg.p <- gage(exp.fc, gsets = kegg.gs, ref = NULL, samp = NULL)

Hi bigmw-

I have begun to play around with gage for pathway analysis in wine grapes (kegg code = vvi). I tried to apply your above instructions to create a gene set that I can employ in my pipeline. I noticed in the recent documentation that kegg.gsets is now kegg.gs, so I updated my code accordingly. However, I also am getting an error (Error: could not find function "kegg.gs") whenever I attempt to call kegg.gs, kegg.gs("vvi"), or when assigning to a variable like: kg.vvi=kegg.gs("vvi"). Any help you might provide will be most helpful.

-Cheers,
Ryan

[bigmw]

Not sure where you read about the "change". The function has never been changed, it is still kegg.gsets (not kegg.gs). so please use the following line to generate your kegg gene set for wine grapes instead:
kg.vvi=kegg.gsets("vvi")

[beeman]

Noob question. Saving pathways to flatfile

This thread is a big help for me as I'm also trying to analyse some honeybee GE data. I'm really a complete novice when it comes to R and I would like to just push the signaling pathways and metabolic pathways to a flatfile I can parse to find matches for specific gid's.

I can't seem to figure out how to write out to a text file as each pathway is a separate matrix/array (not sure of correct structural definition).

This is what the data looks like..

head(kegg.gs)

signaling pathways and metabolic pathways$`ame03010 Ribosome`
[1] "100191002" "100576457" "100577575" "100577702" "100578551" "100578560" "406099" "406120" "406126" "408404" "408515"
[12] "408526" "408675" "408799" "409129" "409202" "409290" "409294" "409326" "409479" "409528" "409552"
[23] "409589" "409599" "409637" "409723" "409728" "409832" "409926" "409951" "410024" "410188" "410486"
[34] "411103" "411380" "411515" "411618" "411862" "412266" "412367" "412473" "412549" "413121" "413137"
[45] "413296" "413398" "413470" "413868" "413875" "413928" "550651" "550711" "550715" "551107" "551125"
[56] "551158" "551330" "551397" "551418" "551584" "551644" "551867" "551870" "552097" "552106" "552241"
[67] "552266" "552272" "552318" "552445" "552494" "552517" "552564" "552593" "552628" "552632" "552641"
[78] "552676" "552726" "552774" "724125" "724142" "724162" "724164" "724186" "724191" "724233" "724369"
[89] "724493" "724531" "724631" "724708" "724744" "724868" "724966" "725062" "725147" "725201" "725291"
[100] "725383" "725432" "725647" "725659" "725866" "725884" "725943" "726013" "726056" "726171" "726270"
[111] "726295" "726439" "726779" "726789" "727128"

$`ame03013 RNA transport`
[1] "100576125" "100576967" "100577501" "100578660" "100578852" "100578879" "408335" "408385" "408386" "408556" "408625"
[12] "408979" "409118" "409169" "409229" "409258" "409308" "409476" "409550" "409560" "409647" "409672"
[23] "409734" "409796" "409839" "409866" "410030" "410099" "410100" "410103" "410239" "410280" "410343"
[34] "410355" "410447" "410580" "410664" "410864" "410972" "411154" "411217" "411474" "411735" "411778"
[45] "411909" "411918" "411974" "411989" "412169" "412508" "412584" "412602" "412721" "412835" "413018"
[56] "413053" "413145" "413146" "413173" "413343" "413344" "413365" "413378" "413650" "413675" "413815"
[67] "413877" "414004" "544670" "550834" "551173" "551184" "551242" "551281" "551283" "551690" "551707"
[78] "551751" "551824" "551901" "551953" "552010" "552225" "552351" "552515" "552780" "724123" "724288"
[89] "724532" "724628" "724659" "725039" "725134" "725227" "725259" "725332" "725396" "725830" "725896"
[100] "726086" "726272" "726338" "726500" "726546" "726874" "727419" "727447"

Can someone please give me a little R schooling??

[WhatsOEver]

The "write.table" function is probably what your are looking for. See here: https://stat.ethz.ch/R-manual/R-deve...ite.table.html

[beeman]

Great thanks for pointing me in the right direction ^_^

[_jim_bo_]

Has anyone worked with MsigDB for genesets? It seems to be pretty comprehensive.

Are there any comparisons between this and KEGG based ones?

[bigmw]

KEGG is the most comprehensive pathway database, with over 400 pathways and over 3000 species.
MsigDB include multiple type s of gene sets besides pathway (hence collections c1-c7), such as GO, transcription factor or miRNA targets, co-expressed gene sets or gene signatures. However, MsigDB only focuses on human genes with an emphasis on cancer related gene sets. In the meantime, it also has substantial redundancy and heterogeneity among different collections, i.e. 1 pathway can have multiple different gene sets.

GAGE can be applied to all these different types of gene sets. In fact, gage package provides functions kegg.gsets and go.gsets, which generate up-to-date gene sets for all 3000 KEGG species and 19 GO species. It also can read in MSigDB gene set data as discribed in the data preparation tutorial:

http://www.bioconductor.org/packages/release/bioc/vignettes/gage/inst/doc/dataPrep.pdf