update aaas with parallel workflows

[dorien-er]

• update AaaS with parallel integration/annotation workflow
• extend existing parameters of OP to AaaS

[jakubmajercik]

This might be relevant elsewhere too but I think that with the number of params with similar names, it gets a bit confusing to see what's what. What do you think about this refactor?

Purpose	Current	Proposed
scVI integration — query	--obs_categorical_covariates	--scvi_obs_categorical_covariates
scVI integration — query	--obs_numerical_covariates	--scvi_obs_numerical_covariates
scANVI annotation — query	--input_obs_categorical_covariates	--scanvi_obs_categorical_covariates
scANVI annotation — query	--input_obs_numerical_covariates	--scanvi_obs_numerical_covariates
scANVI annotation — reference	--reference_obs_categorical_covariates	--scanvi_reference_obs_categorical_covariates
scANVI annotation — reference	--reference_obs_numerical_covariates	--scanvi_reference_obs_numerical_covariates

[jakubmajercik]

Was this change intentional? Shouldn't it point to the composed bucket?

Suggested change

	params.resources_test = "s3://openpipelines-bio/openpipeline_incubator/resources_test/"
	params.resources_test = "s3://openpipelines-bio/openpipeline_composed/resources_test/"

[dorien-er]

no, leftover from before merge of main :-) updated now

[dorien-er]

This might be relevant elsewhere too but I think that with the number of params with similar names, it gets a bit confusing to see what's what. What do you think about this refactor?
Purpose Current Proposed
scVI integration — query --obs_categorical_covariates --scvi_obs_categorical_covariates
scVI integration — query --obs_numerical_covariates --scvi_obs_numerical_covariates
scANVI annotation — query --input_obs_categorical_covariates --scanvi_obs_categorical_covariates
scANVI annotation — query --input_obs_numerical_covariates --scanvi_obs_numerical_covariates
scANVI annotation — reference --reference_obs_categorical_covariates --scanvi_reference_obs_categorical_covariates
scANVI annotation — reference --reference_obs_numerical_covariates --scanvi_reference_obs_numerical_covariates

Good catch - pushed a proposed simplification, where only --input_obs_<x>_covariates (and equivalent --reference_obs_<x>_covariates) are surfaced and applied to both scvi and scanvi, rather than having multiple params for the same purpose - in practice it doesn't make sense to pass different covariates to scvi or scanvi anyway