Test data needs to be updated after subsetting spatial genes #60
Description
With #59 we subset the spatial data to the genes that occur in the scRNAseq data. This solution is fine, at some point we might want to switch to the approach in point 2. though.
- With this new filter the test data changes. Currently the data hasn't been updated. With the new approach we should leave out the gene sampling. For now it's not problematic to continue with the current data, but at some point:
- Leave out the gene sampling
- Recreate the test data
- Upload to aws
- filtering to the shared genes might not be the optimal solution to begin with as we want the processing methods to work on all available counts of the spatial data. To do it properly ideally we'd do a technology specific filter to filter out just the control probes. In this case we need to fix some bugs in the metric calculation. In some metrics we run into an error when sc and sp don't have the same gene sets.
future todos:
- Add gene filter in the processor that covers all spatial technologies
- Fix metrics to work on gene sets that differ
- revert the filtering step to shared genes in the processor to just filter the scrnaseq genes to spatial genes