[BioC] hs.Mm.inp.db problem

[Iain Gallagher]

Hello List

I am trying to map ~5000 mouse genes to human genes using the inparanoid package and I am failing miserably!

Having followed the example in the documentation I can't get any of my 5000 mouse genes converted to human EG ids.

Example follows with 3 genes only:

rm(list=ls())

library(hom.Mm.inp.db)
library(org.Mm.eg.db)
library(org.Hs.eg.db)

#mouse genes in as symbols
dataIn <- c('Ints7', 'Upp1', 'Cdc2a')

#map these to mouse EG ids
egIds <- revmap(org.Mm.egSYMBOL)
mapped <- mappedkeys(egIds)
egIds <- as.list(egIds[mapped])
ind <- which(names(egIds)%in%dataIn)
egIdsIn <- egIds[ind]
#map these IDs to ENSEMBL protein Ids as used for the inparanoid mapping
mouseProtIds <- mget(unlist(egIdsIn),org.Mm.egENSEMBLPROT)
mouseProtIds <- mouseProtIds[!is.na(mouseProtIds)]

#this is the point of failure!
rawHumanProtIds <- mget(unlist(mouseProtIds),hom.Mm.inpHOMSA,ifnotfound=NA)


the returned list is full of NA

Using biomart on the Ensembl site I can get:

Ensembl Transcript ID    Human Ensembl Protein ID
ENSMUST00000020099       ENSP00000397973

For example, for Cdc2a, so I know there are homologs there, but for some reason the inparanoid package is not working for me.
Using the example in the documentation it does work though so I'm assuming the mistake is with me.

Can anyone help with this (more curiosity now - I can get the data through biomart)?

Cheers

Iain