I am working on protein-protein interaction. I am trying to find interaction between two different organisms like host and pathogen can I use cytoscape for protein-protein interaction. also suggest me another best.
Yes and no.
Cytoscape may make it easier to notice patterns across interaction networks but it's mostly just a way to organize the data. The easiest way to use it to predict protein-protein interactions, given interaction data from different species, is a guilt-by-association approach. That is, if two or more proteins interact in one species (or between a host-pathogen pair, perhaps) and the two proteins are also present in another species or pair, then we have evidence that they also interact in that second species AND we have a suggestion that related proteins have similar functions. This kind of approach works especially well when it's combined from a few different data types, like a mix of binary protein interactions and spoke-expansion interactions from protein complexes. This is one of the approaches used by the STRING database (http://string-db.org/), among others.
The critical part is to have a well-defined set of orthologs to group the proteins. That way, you can provide Cytoscape with networks of interactions among orthologous groups rather than individual proteins. I like to use eggNOG (http://eggnogdb.embl.de/) for this.
There are some Cytoscape plugins like PEPPER (http://apps.cytoscape.org/apps/pepper) designed to predict protein-protein interactions based on topological properties but depending upon the data you are working with the approach may not work well. It also depends upon things like GO term annotations for defining known interactions, so if you're working with hosts and pathogens, those annotations may not exist yet.
Good luck!
- Badges
- Science topic
- Similar topics
- Psychology