thanks to reply my question. I also have tried ensemble IDs. But the same problem occurs is about ensemble IDs. I saw in many related articles researchers used GENE-SYMBOLs for transferring genes from GEO to STRING and I was wondering about that because they lose many DEGs.
I performed a differential expression analysis with R. Now I have a list of differential expressed gene symbols and I want to visualize them with Cytoscape. When I enter DE gene symbols to sring db plugin of Cytoscape, it knows some of the gene symbols for more than one protein and asks to choose between them and I don't know how to sure which one is my desired one. So I have to choose a safer way to transfer DEGs to string. Does exist a way to transfer exactly the DEGs themselves to sring by R?
1 answer
There are several approaches here, but fundamentally, gene symbols are not unique (as you discovered) and are not a great way to identify differentially expressed genes. I would suggest translating your gene symbols to uniprot or ensemble identifiers. These should work much better when you hand them off to string-db.
-- scooter
Log in to answer this question.