Several insight reports require annotations of cell phenotypes. Typically, a phenotype annotation is a label of cell type that annotates each cell in the dataset, with the exception of cells that were excluded from analysis due to data quality reasons by Artifact Detection or Cell Quality Control. Good quality cells that cannot be confidently assigned a phenotype can be labeled “Other”.

We offer several tools through the Enable Cloud Platform to facilitate self-serve generation of these annotations, as well as our annotation services.

The appropriate approach for generating phenotype annotations varies depending on the data at hand. For mid-plex datasets like those generated by the Akoya PhenoImager HT, we recommend the analysis approaches described here. For higher-plex datasets like those generated using Enable’s lab services on the Akoya PhenoCycler Fusion platform, we recommend the analysis approaches described here.

If you already have cell phenotype annotations generated on another analysis platform, we have some ability to ingest the annotations provided they are available in the following format:

1. A cell segmentation mask conforming to these specifications.
2. A .csv file containing:
   1. cell_id values that correspond to the integer valued segment ids from the cell segmentation mask in one column
   2. Cell annotation labels for each corresponding cell_id in the second column

We can also import cell annotations associated with cell segmentation results in GeoJSON format (like the type exported by QuPath).