Data API

Note

File formats are being discussed in openpipelines-bio/openpipeline#102

Dataset

obs:
 index # cell id
 sample
 cell_type
 organism
 tissue

mod:
 rna:
 rnav:
 prot:
 atac:
 vdj:

RNA modality

mod:
 rna:
   layers:
     counts
     normalized
   obs:
     <qc metrics>
     doublet_score
     doublet_bool
     cluster
   var:
     index # feature_id, preferably an ensembl id
     feature_name
     <qc metrics>
     highly_variable
     highly_variable_score
   obsm:
     X_pca
     X_integrated
     X_umap
     annotation # scvi, bbknn, ...
   obsp:
     connectivities
     distances

ADT modality

mod:
 prot:
   layers:
     counts
   var:
      index # feature_id
      feature_name # Associated protein names

VDJ modality

mod:
 vdj:
   obsm:
     vdj_t
     vdj_b

ATAC modality

mod:
 atac:
   layers:
     counts
   var:
     interval

RNA velocity modality

mod:
 rnav:
   layers:
     spliced
     unspliced