I have a list of ENSEMBL gene IDs from sheep and I would like to perform a GSEA analysis on them. From what I understand of GSEA, the easiest (best?) way to perform such analysis in my case is to transform my sheep IDs into orthologous human IDs.
I obtained the right translation using biomaRt but I am not completely satisfied by the result since only ~70% of my genes have an associated human gene ID. Moreover, I saw quite often duplications in the translation and don't know which human ID to choose in those cases:
ENSOARG000000011111 ==> ENSG000001234 ENSOARG000000011111 ==> ENSG000001235 ENSOARG000000011111 ==> ENSG000001236
Question 1: Is converting my sheep IDs like I'm doing the best way to conduct a GSEA analysis? Or can you propose me another method to analysis pathway/GO enrichment in my situation?
Question 2: What is the best way to obtain the most accurate translation of my sheep IDs in order to run a GSEA?
Can a blast analysis of my 400 Ovis Aries genes against a human database solves partially my problem or I should expect that it won't give me more information than the biomaRt query?
Thanks in advance,