pipeline.cwl

#!/usr/bin/env cwl-runner

class: Workflow
cwlVersion: v1.0
label: Pipeline for concatenating sc-ATAC-seq datasets into MuData object

requirements:
  ScatterFeatureRequirement: {}

inputs: 
    data_directory:
        label: "Path to directory containing cell by gene and cell by bin files"
        type: Directory
    
    uuids_file:
        label: "Path to a file containing a list of uuids and other metadata for the dataset to be indexed"
        type: File
    
    tissue:
        label: "Two letter tissue type code"
        type: string?
      
    access_key_id:
        label: "AWS access key id"
        type: string
    
    secret_access_key:
        label: "AWS secret access key"
        type: string

outputs:
    mudata_file:
        type: File
        outputSource: concatenate/mudata_file
    
    metadata_json:
        type: File
        outputSource: concatenate/metadata_json

steps:

    - id: concatenate
      in: 
        - id: data_directory
          source: data_directory
        - id: uuids_file
          source: uuids_file
        - id: tissue
          source: tissue
    
      out:
        - mudata_file
        - metadata_json
      run: steps/concatenate.cwl
      label: "Concatenates h5ad files in directory"

    - id: upload
      in: 
        - id: mudata_file
          source: concatenate/mudata_file
        - id: metadata_json
          source: concatenate/metadata_json
        - id: access_key_id
          source: access_key_id
        - id: secret_access_key
          source: secret_access_key
    
      out:
        - finished_text
      run: steps/upload.cwl
      label: "Uploads the pipeline outputs to s3"