S3 integration. Orthopairs files are now going to be stored on S3 post-Orthopairs, and then drawn from S3 pre-orthoinference, rather than keeping the files in a static location. This will have a downstream benefit of removing the release number-specific filepath that Orthoinference requires to access the Orthopairs files currently.
japi dependency bumped from 1.0.30 to 1.0.31
File download/removal code was moved from the Main class to either 'runOrthopairs.sh' or the Jenkinsfile. This is so its easier to track when something goes wrong at this phase. Also keeping in line with a generalized theme of having the code do the 'work' while Jenkins/bash handles 'setup'
I found that the Duration.sleep of 2 seconds wasn't needed for querying gene names from UniProt. UniProt appears more than capable of taking the abuse, and since they restrict it to query sizes of 250, it seems like the natural rate is well within their limits. Orthopairs now takes roughly an hour now instead of 1.5-2 (addressing #5 )
@jweiser I looked into addressing #4 , which was a suggestion to use List.subclass method for creating lists within lists (a necessity for querying UniProt API since you can only have batches of 250 queries at a time), but I didn't really feel the reason of changing slightly bloated code for maximum brevity. I plead 'if it ain't broke don't fix it'. So I'd like to keep the code in the state it is in, and so would like to close #4
List of changes for v73:
Duration.sleep
of 2 seconds wasn't needed for querying gene names from UniProt. UniProt appears more than capable of taking the abuse, and since they restrict it to query sizes of 250, it seems like the natural rate is well within their limits. Orthopairs now takes roughly an hour now instead of 1.5-2 (addressing #5 )List.subclass
method for creating lists within lists (a necessity for querying UniProt API since you can only have batches of 250 queries at a time), but I didn't really feel the reason of changing slightly bloated code for maximum brevity. I plead 'if it ain't broke don't fix it'. So I'd like to keep the code in the state it is in, and so would like to close #4