Closed benmwebb closed 5 years ago
I misunderstood what you suggested, and I deleted my previous comment. Sorry about it
Since I'm not that familiar with PMI2, I added a wiki page cheat sheet for transitioning from PMI1 to PMI2. Please take a look at that wiki page and make corrections where necessary and add more notes. This will eventually go into the manual to help others transition to PMI2.
Obviously there won't always be a 1:1 mapping between PMI1 and PMI2 methods. Where this isn't the case, a note pointing to the new PMI2 way of doing things would be helpful.
We should follow the imp_tutorial/develop example as much as possible
Note that realistically it will take a long time to transition published applications to PMI2, and there's little benefit to doing so. salilab/imp#997 may be a better solution. We should definitely transition examples and tutorials though.
We should follow the imp_tutorial/develop example as much as possible
This would require the development version of the tutorial to work, which isn't yet the case. See the latest Travis failures.
All tutorials and examples should now use PMI2. Legacy systems will continue to use PMI1, although new applications use PMI2.
We should port each of the IMP systems to PMI2. This should ensure that these systems continue to work in the future, expose any areas where PMI2 is lacking relative to PMI1, and allow us to promote PMI2 as the default PMI.
For each system this requires the following steps:
develop
branch in the GitHub repository (themaster
branch is required to work with the last IMP stable release) and make any necessary changes there.master
branch.@saltzberg has already volunteered to handle the IMP tutorial. I'll handle the glycophorin and integrin example systems to start with, before moving on to "real" systems.