search

bokulich-lab / RESCRIPt

REference Sequence annotation and CuRatIon Pipeline
BSD 3-Clause "New" or "Revised" License
85 stars 26 forks source link

Migrate transformers, etc... to qiime2/q2-types #105

Open mikerobeson opened 3 years ago

mikerobeson commented 3 years ago

Given the recent updates to q2-types (qiime2/q2-types#256) to handle RNA data, and the current required functionality for RESCRIPt (e.g. DNA-to-RNA conversions), we should migrate some of these transformers and related content to q2-types. Slate this for the next QIIME 2 (2021.6??) and RESCRIPt releases.

thermokarst commented 3 years ago

EDIT: this was resolved in #102

This actually makes it so that it is impossible to have both q2-types and rescript installed in the same env (which is a problem, since rescript depends on q2-types):

ValueError: Duplicate semantic type ('RNASequence') defined in plugins: 'types' and 'rescript'

Unfortunately, this has caused some problems with the QIIME 2 2021.4 release - we need to train new feature classifiers in order to publish the release, but we can't train right now because rescript is uninstallable in this state. I'll take a crack at fixing this, PR coming in soon.

nbokulich commented 2 years ago

@thermokarst @mikerobeson can this issue be closed? It looks like this was resolved in #102 ... is there still a lingering transformer that needs to migrate?

mikerobeson commented 2 years ago

There are a few transformers that need to be moved over still... and utility methods here and here.

nbokulich commented 2 years ago

the first utility you linked (reverse_transcribe) is exposing the transformer as a stand-alone method and should not be moved (at least not to q2-types).