How Do I Match Orthologues In One Species To Another, Genome Scale?

Question

[Jason]

I've been doing genome wide expression studies in S. cerevisiae, but a recent publication performed similar experiments using my gene of interest's orthologue in C. elegans. I want to see if genes I found to be highly/lowly expressed in my microarray study (yeast) were highly/lowly in their study (worms). Does anyone have a file which has genes and their orthologues (yeast/worms)? Or is there a way I could do this through NCBI without going gene by gene? I know how to write in python so if someone could suggest a method using script that would be helpful too. Thanks

[Jarretinha]

Did you try the common solutions like NCBI KOGs?

Answer 1

[Michael]

Can be quite simple: This BioMart query will do the job. And this should be also the most comfortable solution

This will provide a mapping of yeast genes to orthologous C.elegans genes with some additional attributes. In the results page click Go to export this mapping as a file. Chose other attributes on the left if you like.

[Ian Simpson]

Funny, I tried to do this in the Ensembl MART and had to end up adding the second dataset as I wasn't being offered the other species through the multi species comparison option of the first. Wonder if that's a bug ?

[Michael]

I can do exactly the same query in ensmbl mart. Under attributes chose Homologs, then open C.ELEGANS ORTHOLOGS:

[Ian Simpson]

Hmmm. Did that, must have done something odd. Thought it was strange that it wasn't working as I had done similar before with no problem. Thanks.

[Jason]

I just started to use the biomart site, but it looks like it is exactly what I needed. Thanks a ton!

Answer 2

[Ian Simpson]

If you want to do it with scripting (Perl) the very powerful Ensembl Compara database is really very good. It has what is probably the best orthology walking pipeline out there at the moment.

http://www.ensembl.org/info/docs/api/compara/compara_tutorial.html

If you want to make your life ball-bouncingly easy just use the Ensembl BioMart service http://www.ensembl.org/biomart/martview just chose one of your species then under the "filter" chose "multi species comparison" then from the pull-down menu pick the other species. At the bottom you can add an additional dataset, chose the other species and then from the filter fro this dataset again chose the "multi-species comparison" and pick the other species from the pull-down menu. Hit results and it will provide you with a table mapping genes from one species on to the other. If you want to use smaller subsets of genes just provide those gene ids in the gene section of the filter selection and provide the ids. Hope that helps.

Answer 3

[Adrian]

InParanoid is another database of pre-computed orthologs with convenient downloadable files.

Answer 4

[Perry]

Try ROUNDUP or homologene

Answer 5

[Nkululeko]

While BioMart is great it is limited in the number of genomes it has e.g. I recently was trying to find all the orthologues shared between a number of mycobacterium species - BioMart does not at present have these species catalogued.

An excellent alternative when looking for orthologues in microbial genomes is the 'Microbial Genome Database for Comparative Analysis'.

To create a table of homologous genes across various species check out: mbgd.genome.ad.jp/htbin/SelectOrganism.pl

Answer 6

[Larry_Parnell]

I like Ian's approach. The only thing I would add is the importance of knowing the date that the comparisons in any pre-computed dataset such as this done. For example, the S. cerevisiae genome has undergone revision of gene models with the completion and subsequent analysis of several of yeast species geneomes. Was the ortholog comparison done before or after the revised gene models came on-line? Thus, it is always wise to either check a few special examples by hand or to run a script.

Answer 7

[David Managadze]

NCBI has HomoloGene service for that. See file ftp://ftp.ncbi.nih.gov/pub/HomoloGene/current/homologene.data for protein accessions, gene IDs, taxonomy IDs, etc. S. cerevisiae and C. elegans are there.