Search
Question: Questions about Pathview, maybe also gage
0
23 months ago by
Germany
ChIP-Tease0 wrote:

Hello everybody,

Today i found the R Pathview package and could already produce some output pictures.
Nevertheless i have some problems and hope that someone might help me out.

1. There is a difference between the original KEGG picture and the modified Pathview picture.
I guess this is not supposed to be this way? I don’t really know what to do about this!?
This happens for all the exported pictures.

http://pic-hoster.net/view/64307/hsa04514.png.htm

http://pic-hoster.net/view/64308/hsa04514.Down.png.htm

2. I would love to change the color, since there are people how cannot distinguish red and green, but i don’t find anything, which might help.

3. When i spilt up my genes to up and downregulated genes and only check the significantly upregulated ones, i don’t get any result. Only NAs. Is there a way to change the p.value threshold somehow?
In contrast to gage the Webgestalt (http://bioinfo.vanderbilt.edu/webgestalt/option.php) tool finds an enrichment for my upregulated genes (92 genes)

4. I’m not really sure, what the shortcuts should tell me. I’d be happy for an explanation:
p.geomean   stat.mean        p.val       q.val set.size         exp1

My R code

library(dplyr)
library(pathview)

library("AnnotationDbi")
library("org.Hs.eg.db")

my_df <- data.frame(my_data)

UPDOWN <- „Up"
UpReg <- my_df[my_df $log2FoldChange < log2(0.4) & my_df$baseMean > 20 & my_df $padj < 0.05, ] UpReg$symbol <- mapIds(org.Hs.eg.db,
keys=row.names(UpReg),
column="SYMBOL",
keytype="SYMBOL",
multiVals="first")

UpReg$entrez <- mapIds(org.Hs.eg.db, keys=row.names(UpReg), column="ENTREZID", keytype="SYMBOL", multiVals="first") UpReg$name =   mapIds(org.Hs.eg.db,
keys=row.names(UpReg),
column="GENENAME",
keytype="SYMBOL",
multiVals="first")

library(gage)
library(gageData)
data(kegg.sets.hs)
data(sigmet.idx.hs)
kegg.sets.hs = kegg.sets.hs[sigmet.idx.hs]

foldChange <- UpReg$log2FoldChange names(foldChange) <- UpReg$entrez

keggres = gage(foldChange, gsets=kegg.sets.hs, same.dir=TRUE)

library(dplyr)
keggrespathways <- data.frame(id=rownames(keggres$less), keggres$less) %>% tbl_df() %>%filter(row_number()<=5) %>% .\$id %>%  as.character()
keggrespathways

keggresids <- substr(keggrespathways, start = 1, stop = 8)
keggresids

plot_pathway <- function(pid) pathview(gene.data = foldChange, pathway.id = pid, species = „hsa“, new.signature = FALSE, out.suffix = UPDOWN)
detach(„package:dplyr“, unload=TRUE)# otherwise a get an error message discussed ion another thread
tmp <- sapply(keggresids, plot_pathway) # exports the original KEGG pig, pathview pic and some xml file

Thanks a lot, Alex

modified 17 months ago by bioinf0 • written 23 months ago by ChIP-Tease0
0
17 months ago by
bioinf0
bioinf0 wrote:

Hello there,

Did you find the explanation? If so, it would be nice if you can share it.

Thanks!