Generate reports for CosMx data

[dorien-er]

Goal: update the existing workflows/generate_qc_report workflow to accept and run a report for CosMx H5MU

Implementation

• Update the resources_test_script/spatial_qc_sample_data.sh to generate cosmx data with QC metrics

• Update the src/ingestion_qc/h5mu_to_qc_json component:

  - `--ingestion_method` should include `cosmx`
  - Final JSON should include standard QC metrics (total counts, nonzero vars, fraction mitochondrial, fraction ribosomal)
  - Final JSON should include spatial coordinates (found in .obsm["spatial"])
  - Final JSON should include spatial QC metrics (found in .obs): `Area`, `AspectRatio`, `Mean.MembraneStain`, `Max.MembraneStain`, `Mean.PanCK`, `Max.PanCK`, `Mean.CD45`, `Max.CD45`, `Mean.CD3`, `Max.CD3`, `Mean.DAPI`, `Max.DAPI`
  

• Update the siqc reporting engine (https://github.com/openpipelines-bio/siqc) to accept a CosMx-based JSON and generate a report including CosMx QC metrics.

  Default visualizations in the report:
  
  - Total counts histogram (per sample per cell)
  - Non-zero vars histogram (per sample per cell)
  - Mitochondrial fraction histogram (per sample per cell)
  - Ribosomal fraction histogram (per sample per cell)
  - Cell area histogram (per sample per cell)
  - Aspect ratio histogram (per sample per cell)
  - DAPI intensity histogram (per sample per cell)
  - MembraneStain intensity histogram (per sample per cell)
  
  Optional visualizations in the report:
  
  - CD45 intensity histogram (per sample per cell)
  - CD3 intensity histogram (per sample per cell)
  - PanCK intensity histogram (per sample per cell)
  

Data

For development: s3://openpipelines-bio/openpipeline_spatial/resources_test/cosmx/Lung5_Rep2_tiny.h5mu
For running example report: https://nanostring-public-share.s3.us-west-2.amazonaws.com/SMI-Compressed/Lung5_Rep2/Lung5_Rep2+SMI+Flat+data.tar.gz

Test run/example report

Once finalized, run the reporting workflow on Seqera Cloud on the full-sized dataset (see above).