Completed GO-MWU for new analyses

Speeding Things Up As of my last post (yesterday), I was trying to figure out a way to rapidly obtain large numbers of GO IDs. Last time, I used Sam’s shell script by calling it inside this script. However, Sam’s script took an extremely long time to run. A rough ballpark: for my 2 analyses (each with ~120,000 accession IDs), it would take a total of about 8 days on my local machine. That’s a huge pipeline bottleneck, and so we found an alternative. New Script for Uniprot to GO I created an R script for obtaining GO terms from accession IDs. Downside: it requires you to manually download the SwissProt database – with all GO terms included – from Upside: it’s much, much, much faster. It ran in a few minutes, which by my calculation, is a bit faster than 8 days. After getting GO terms, you need…

from Aidan F. Coyle