Ancestral homolog assignment

Question

[nbeveryday]

I'm going to make co-evolution analysis of genes involved in a disease phenotype.And current methods I choose need homologs of these genes .Though pfam and inparanoid can help but I am not satisfied with the result they provide . As my study focus mainly on primitive eukaryotes,the ancestral homologs (I don't expect ortholog) can be so different from the human disease genes.So I 'm eager to know if there is a "keen"(but not too many false positive results) method to help me find out disease gene homologs in primitive eukaryotes such as entamoeba histolytica & Dictyostelium discoideum, and how is the possibility to develop a method based on current bioinformatics.

[Juke34]

If you are not satisfied by tools that create (putative) orthologous groups by reciprocal best-hit ... it will be difficult.
The research for highly diverged homolog genes is an hard problem. Many research works work in this area. You can use gene/protein profiles. See bibliography ...
Apparently you want to work at large scale. I never heard about this kind of tool to work at large scale (I am not in the field, you have to verify). I suppose you should develop an approach integrating this kind of tool.