Co-localization of two proteins and interaction

Question

[mahdijalili]

Hi,

I am seeking any evidence (paper or other reference) for co-localization of two proteins lead to interaction in PPI network. Any help will be greatly appreciated.

Thanks.

Answer 1

[Sam]

You can try using STRING. All you have to do is to give the protein / gene name to the database and if there are any evidence of them being co-localized or interacted together, then you should be able to find it. Another possible database will be GeneMania, however, I never managed to extract the exact paper for a particular evidence. You might want to give both a choice

[Jean-Karim Heriche]

I think you may misunderstand what GeneMania does: "however, I never managed to extract the exact paper for a particular evidence.". GeneMania computes a measure of functional similarity between genes based on a number of data sources. For each link between a pair of genes, it gives some weight to the data sources and you can retrieve these but it doesn't mean that the paper or database contains a direct evidence for the link, only that it provides the stated amount of support to it.

Also I don't think GeneMania includes explicit co-localization data unless you consider its use of GO cellular component annotations as indication of co-localization.

[Sam]

I am not particular sure about the evidence part of GeneMania. That is always confusing to me. But as for the co-localization part, they actually provide an annotation of co-localization. You just have to select it under advance options. There will be the options for co-localization (at least for human and mouse). As for getting the database and paper, it seems like you just have to click on the edge on the network and you will be given the information.

[Jean-Karim Heriche]

Co-localization usually means (at least for me) subcellular co-localization, i.e. sharing some space inside a cell. GeneMania seems to have a different notion of co-localization because the sources for their co-localization data are microarray and chip-seq papers. Actually, from their doc:

Co-localization: Genes expressed in the same tissue, or proteins found in the same location. Two genes are linked if they are both expressed in the same tissue or if their gene products are both identified in the same cellular location.

It's confusing to biologists like me because, for me, the first part of their definition would be co-expression.

Anyway, GeneMania is a great tool, one just has to be careful in the choice of data sources.

[Sam]

True. The definitions are usually confusing. In most database, I guess even for STRING, the definitions of co-localization are usually genes or protein that are found in the same location. Whereas for co-expression, not only does these genes or proteins appears in the same location, but they also have a correlation in their expression level .

I usually ignore the co-localization pairs as it seems so easier to happen when compare to co-expression. So, yes, we must be-careful of the definition used by different tools, not only in GeneMania. This definition thing are often the most annoying problem we face everyday...