You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
While comparing full output files from scpca-nf might be ideal, the fact that the output files are large suggests that we might prefer to pre-compute a set of metrics as part of the output. Aa first step of this, we will need a list of specific metrics that we will calculate on each workflow run to identify changes in the workflow outputs.
Some of these, such as number of cells, may already exist in metadata files that we produce, but others will require new calculations.
The expected output here will be a document in this repository that describes the key metrics, including which input files may be required for each metric. We will use that document to guide the next steps of creating an output file with these metrics.
While comparing full output files from
scpca-nf
might be ideal, the fact that the output files are large suggests that we might prefer to pre-compute a set of metrics as part of the output. Aa first step of this, we will need a list of specific metrics that we will calculate on each workflow run to identify changes in the workflow outputs.Some of these, such as number of cells, may already exist in metadata files that we produce, but others will require new calculations.
The expected output here will be a document in this repository that describes the key metrics, including which input files may be required for each metric. We will use that document to guide the next steps of creating an output file with these metrics.
A partial list of metrics and more information can be found at https://data-lab-knowledge.slab.com/posts/discussion-testing-scpca-nf-outputs-3d4pra87
The text was updated successfully, but these errors were encountered: