Jim Hutchins's Avatar

Jim Hutchins

21 May, 2013 09:32 AM

I note that Treefam includes S.cerevisiae (Sc), S.pombe (Sp), Arabidopsis and rice as an "Outgroup" set of species. I find this a really good idea, to give a sense of how genes have been conserved (or not) during evolution, on a wider scale than 'just' animals. And also, these are key model organisms of interest to a great many researchers worldwide.

However in several cases where I know a (human) gene has S.pombe or S.cerevisiae orthologues, these are often not found in Treefam.

Just one example: Chk1 / CHEK1
Here are links to these genes in their "home" (species-specific) databases:

Searching Treefam for CHEK1, and opening the tree TF351441 does not include these Sp and Sc orthologs:

Oddly enough, Ensembl reports the Sc, but not the Sp orthologue:
As the chk1 gene was in fact first discovered in S.pombe (, to me it seems rather strange not to include Sp in the list of organisms that shares that gene!

In this case, the NCBI's HomoloGene correctly identifies the Sc and Sp orthologues:

Clearly these different sites (Treefam / Ensembl / HomoloGene) use different methods and stringencies to define what is included as being an orthologue.

Happily, Treefam does find Sc, Sp and At orthologues for the well-conserved gene DNA Polymerase alpha (POLA1):
- so at the "highly conserved" end of the conservation spectrum, this aspect of Treefam works just fine.

My questions related to this are:
- do you think it's possible that Treefam could be missing out on some genuine and important orthologues, by being a bit too stringent?
- what do you think of the idea of (automatically) 'cross-checking' between Treefam and Ensembl orthologues/paralogues, to ensure some consistency between these?

  1. 1 Posted by Fabian on 14 Jun, 2013 08:55 AM

    Fabian's Avatar

    Hi Jim,
    thanks for your feedback.

    You are right, there are differences between TreeFam, Ensembl Compara and HomoloGene, which are due to the different methodology or taxon sampling.
    Also, we keep our families fixed TreeFam and whereas e.g. Compara rebuild their families with every release. There are good arguments for either approach.

    At some time in the near future we will be looking into ways of testing our families. We will then also cross-check our predictions with e.g. Ensembl. That is a good idea.

    Thanks again for your valuable feedback!
    I will close the ticket for now, as we have added your suggestions to our todo list for the evaluation of our families


  2. Fabian closed this discussion on 14 Jun, 2013 08:55 AM.

Comments are currently closed for this discussion. You can start a new one.

Keyboard shortcuts


? Show this help
ESC Blurs the current field

Comment Form

r Focus the comment reply box
^ + ↩ Submit the comment

You can use Command ⌘ instead of Control ^ on Mac