[BioC] drosphila2PATH question

[Nianhua Li]

Dear Ariel,

The short answer is: this is a bug/missing feature.

The package drosophila2 was built by using ABPkgBuilder, which is
designed for
human, mouse and rat, not for fly. To get probe to KEGG mapping, the
function
first gets probe to Entrez Gene mapping, and then use
ftp://ftp.genome.jp/pub/kegg/pathways/dme/dme_gene_map.tab
to get KEGG pathway IDs. dem_gene_map.tab contains two columns: KEGG
gene IDs
and KEGG pathway IDs. KEGG has similar files for each organism. For human,
mouse and rat, KEGG Gene IDs happen to be the same as Entrez Gene ID.
Therefore,
function ABPkgBuilder can use this file to get probe to KEGG pathway mapping
right away. But for other organisms KEGG gene IDs are in a different format.
Therefore ABPkgBuilder will fail to get KEGG pathway IDs.

One solution is to modify ABPkgBuilder to get Entrez Gene to KEGG gene ID
mapping from
ftp://ftp.genome.jp/pub/kegg/genomes/dme/dme_ncbi-geneid.list

The other solution is to write specific builder functions for fly, just like
what happened to Arabidopsis and yeast. It is because ABPkgBuilder uses
Entrez
Gene ID as the base identifier for gene. But most people may perfer flybase
(rather than Entrez Gene). Therefore, it is more appealling to write a fly
specific builder function. But this is also time consuming...

sorry there is no direct answer, but hope it provides some background
information.

nianhua