View Single Post
Old 10-07-2012, 07:11 PM   #3
A_Morozov
Member
 
Location: Russia, Irkutsk

Join Date: Feb 2011
Posts: 40
Default

For start, you can just call pairs of genes with alignment score above certain treshold common, and call those that don't have such pairing unique. This is grossly oversimplified approach, because unless you have some really small genomes (like viral or organellar) some genes are gonna have more than one copy. Are these the same or different or you? And what if some genes are indeed of the same function but derived from distant groups eg via HGT? And are you gonna compare anything other than gene content?
I agree with westerman - you need to tell us more about what exactly question your project must answer if you want us to give valuable advice.
A_Morozov is offline   Reply With Quote