CytoSnake Workflows#

Below are the available workflow modules in CytoSnake

cp_process#

Generates consensus profiles.

Consensus profiles contain unique signatures that are associated with a specific state of the cell (e.g., control, DMSO, chemical/genetic perturbations, etc.).

The development of this workflow was heavily influenced by this study: https://github.com/broadinstitute/cell-health

included modules:

  • common

  • aggregate

  • annotate

  • normalize

  • feature_select

  • generate_consensus

parameters

input:

  • profile: single-cell morphology datasets

  • metadata: metadata directory associated with plate data.

  • barcode (optional): file containing plate data to plate map pairings. Default is None.

outputs:

  • profiles: aggregated, annotated, normalized, selected features, and consensus profiles.

  • cell counts: Cell counts per well

cp_process_singlecells#

Converts SQLite plate data into parquet and returns selected features from single-cell morphology profiles.

The development of this workflow was heavily influenced by: https://github.com/WayScience/Benchmarking_NF1_data/tree/main/3_extracting_features

included modules

  • common

  • cytotable_convert

  • annotate

  • normalize

  • feature_select

parameters

inputs:

  • profiles: single-cell morphology datasets in SQLite format.

  • metadata: metadata file associated with single-cell morphology dataset.

outputs:

  • profiles: annotated, normalized, and feature selected profiles.