update workflows

docs/starsolo.rst

@@ -335,20 +335,12 @@ Below are inputs for *count* workflow. Notice that required inputs are in bold.
       - Disk space in GB needed for count per sample.
       - 500
       - 500
-    * - backend
-      - Cloud infrastructure backend to use. Available options:
-
-        - ``gcp`` for Google Cloud;
-        - ``aws`` for Amazon AWS;
-        - ``local`` for local machine.
-      - "gcp"
-      - "gcp"
     * - preemptible
-      - Number of maximum preemptible tries allowed. This works only when *backend* is ``gcp``.
+      - Number of maximum preemptible tries allowed. Only works for GCP
       - 2
       - 2
     * - awsQueueArn
-      - The AWS ARN string of the job queue to be used. This only works for ``aws`` backend.
+      - The AWS ARN string of the job queue to be used. Only works for AWS
       - "arn:aws:batch:us-east-1:xxx:job-queue/priority-gwf"
       - ""
 

workflows/cellranger/cellranger_atac_create_reference.wdl

@@ -16,8 +16,6 @@ workflow cellranger_atac_create_reference {
 
         # Number of preemptible tries
         Int preemptible = 2
-        # Backend
-        String backend = "gcp"
         # Arn string of AWS queue
         String awsQueueArn = ""
 
@@ -41,6 +39,11 @@ workflow cellranger_atac_create_reference {
     # Output directory, with trailing slashes stripped
     String output_directory_stripped = sub(output_directory, "[/\\s]+$", "")
 
+    # Backend: gcp, aws, local
+    Boolean use_gcp = sub(output_directory, "^gs://.+$", "gcp") == "gcp"
+    Boolean use_aws = sub(output_directory, "^s3://.+$", "aws") == "aws"
+    String backend = (if use_gcp then "gcp"  else (if use_aws then "aws" else "local"))
+
     call run_cellranger_atac_create_reference {
         input:
             docker_registry = docker_registry,

workflows/cellranger/cellranger_count.wdl

@@ -10,8 +10,8 @@ workflow cellranger_count {
         String input_fastqs_directories
         # A comma-separated list of input data types
         String? input_data_types
-        # A comma-separated list of input feature barcode files
-        String? input_fbf
+        # A comma-separated list of input auxiliary files
+        String? input_aux
         # CellRanger output directory, gs url
         String output_directory
 
@@ -20,9 +20,6 @@ workflow cellranger_count {
         # Index TSV file
         File acronym_file
 
-        # Target panel CSV for targeted gene expression analysis
-        File? target_panel
-
         # chemistry of the channel
         String chemistry = "auto"
         # Force pipeline to use this number of cells, bypassing the cell detection algorithm, mutually exclusive with expect_cells.
@@ -69,10 +66,9 @@ workflow cellranger_count {
             input_samples = input_samples,
             input_fastqs_directories = input_fastqs_directories,
             input_data_types = input_data_types,
-            input_fbf = input_fbf,
+            input_aux = input_aux,
             output_directory = output_directory,
             genome_file = genome_file,
-            target_panel = target_panel,
             chemistry = chemistry,
             force_cells = force_cells,
             expect_cells = expect_cells,
@@ -104,10 +100,9 @@ task run_cellranger_count {
         String? input_samples
         String input_fastqs_directories
         String? input_data_types
-        String? input_fbf
+        String? input_aux
         String output_directory
         File genome_file
-        File? target_panel
         String chemistry
         Int? force_cells
         Int? expect_cells
@@ -191,22 +186,17 @@ task run_cellranger_count {
                     for fastq_f in fastq_files:
                         check_fastq_file(fastq_f, sample_name)
 
-        samples = data_types = fbfs = None
+        samples = data_types = auxs = None
         fastqs_dirs = []
 
         if '~{input_samples}' != '':
-            assert version.parse('~{cellranger_version}') >= version.parse('3.0.2')
             samples = '~{input_samples}'.split(',')
             data_types = '~{input_data_types}'.split(',')
-            fbfs = '~{input_fbf}'.split(',')
+            auxs = '~{input_aux}'.split(',')
 
-            target_panel = set()
             feature_file = set()
-            for dtype, fbf in zip(data_types, fbfs):
-                if dtype == 'rna':
-                    target_panel.add(fbf)
-                else:
-                    feature_file.add(fbf)
+            for dtype, aux in zip(data_types, auxs):
+                feature_file.add(aux)
 
             def _locate_file(file_set, keyword):
                 if len(file_set) > 1:
@@ -220,7 +210,6 @@ task run_cellranger_count {
                 check_call(call_args)
                 return os.path.abspath(os.path.basename(file_loc))
 
-            target_panel = _locate_file(target_panel, 'target panel')
             feature_file = _locate_file(feature_file, 'feature reference')
             assert feature_file != ''
 
@@ -256,26 +245,14 @@ task run_cellranger_count {
 
         if samples is None: # not Feature Barcode
             call_args.extend(['--sample=~{sample_id}', '--fastqs=' + ','.join(fastqs_dirs)])
-            if ('~{target_panel}' != '') and (os.path.basename('~{target_panel}') != 'null'):
-                assert version.parse('~{cellranger_version}') >= version.parse('4.0.0')
-                call_args.append('--target-panel=~{target_panel}')
         else:
             call_args.extend(['--libraries=libraries.csv', '--feature-ref=' + feature_file])
-            if target_panel != '':
-                assert version.parse('~{cellranger_version}') >= version.parse('4.0.0')
-                call_args.append('--target-panel=' + target_panel)
 
         if '~{force_cells}' != '':
             call_args.append('--force-cells=~{force_cells}')
         if '~{expect_cells}' != '':
             call_args.append('--expect-cells=~{expect_cells}')
-        if '~{include_introns}' == 'true':
-            assert version.parse('~{cellranger_version}') >= version.parse('5.0.0')
-            if version.parse('~{cellranger_version}') >= version.parse('7.0.0'):
-                call_args.extend(['--include-introns', '~{include_introns}'])
-            else:
-                call_args.append('--include-introns')
-        elif version.parse('~{cellranger_version}') >= version.parse('7.0.0'):
+        if '~{include_introns}' == 'false':
             call_args.extend(['--include-introns', '~{include_introns}'])
 
         # For generating BAM output
@@ -286,7 +263,6 @@ task run_cellranger_count {
                 call_args.append('--create-bam=false')
         else:
             if '~{no_bam}' == 'true':
-                assert version.parse('~{cellranger_version}') >= version.parse('5.0.0')
                 call_args.append('--no-bam')
 
         if '~{secondary}' != 'true':

workflows/cellranger/cellranger_create_reference.wdl

@@ -36,15 +36,18 @@ workflow cellranger_create_reference {
 
         # Number of preemptible tries
         Int preemptible = 2
-        # Backend
-        String backend = "gcp"
         # Arn string of AWS queue
         String awsQueueArn = ""
     }
 
     # Output directory, with trailing slashes stripped
     String output_directory_stripped = sub(output_directory, "[/\\s]+$", "")
 
+    # Backend: gcp, aws, local
+    Boolean use_gcp = sub(output_directory, "^gs://.+$", "gcp") == "gcp"
+    Boolean use_aws = sub(output_directory, "^s3://.+$", "aws") == "aws"
+    String backend = (if use_gcp then "gcp"  else (if use_aws then "aws" else "local"))
+
     call generate_create_reference_config {
         input:
             input_sample_sheet = input_sample_sheet,