Sfitz by readgroup and add FastQC

.gitignore

@@ -80,3 +80,8 @@ work/
 *.gz
 *.tar
 *.zip
+
+# Other
+test/*
+test/*/*
+slurm-*.out

CHANGELOG.md

@@ -10,10 +10,14 @@ This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.htm
 ## [Unreleased]
 
 ### Added
+- Add FastQC workflow
+- Add per readgroup and per library functionality
+- Add `process_afterscript`
 - Add Nextflow version requirement to `README`
 
 ### Changed
 - Update SAMtools 1.18 to 1.20
+- Update NFTest for FastQC and new test sample
 - Update repository/pipeline description
 - Update Nextflow configuration test workflows
 

README.md

@@ -71,7 +71,7 @@ input:
 
 | Field | Type | Required | Description |
 | ----- | ---- | ------------ | ------------------------ |
-| `algorithms` | list | no | List of tools to be run: ['stats', 'collectwgsmetrics', 'bamqc'], default = ['stats', 'collectwgsmetrics'] |
+| `algorithm` | list | no | List of tools to be run: ['stats', 'collectwgsmetrics', 'bamqc'], default = ['stats', 'collectwgsmetrics'] |
 | `reference` | path | yes/no | Reference fasta is required only for `CollectWgsMetrics` |
 | `output_dir` | path | yes | Not required if `blcds_registered_dataset` = `true` |
 | `blcds_registered_dataset` | boolean | no | Default is `false`. Only `uclahs_cds` users should change this. When `true`, BLCDS folder structure is used |
@@ -80,8 +80,16 @@ input:
 #### SAMtools specific configuration
 | Field | Type | Required | Description |
 | ----- | ---- | ------------ | ------------------------ |
-| remove_duplicates | boolean | no | Ignore reads marked as duplicate. default = `false` |
-| samtools_stats_additional_options | string | no | Any additional options recognized by `samtools stats` |
+| stats_max_rgs_per_sample | integer | no | If a sample has more than this number of readgroups, `SAMtools stats` will not run per readgroup analysis. Default = 20 |
+| stats_max_libs_per_sample | integer | no | If a sample has more than this number of libraries, `SAMtools stats` will not run per library analysis. Default = 20 |
+| stats_remove_duplicates | boolean | no | Ignore reads marked as duplicate. default = `false` |
+| stats_additional_options | string | no | Any additional options recognized by `samtools stats` |
+
+#### FastQC specific configuration
+| Field | Type | Required | Description |
+| ----- | ---- | ------------ | ------------------------ |
+| fastqc_level | string | yes | 'readgroup', 'library' or 'sample' |
+| fastqc_additional_options | string | no | Any additional options recognized by `FastQC` |
 
 #### Picard specific configuration
 | Field | Type | Required | Description |
@@ -95,7 +103,7 @@ input:
 #### Qualimap specific configuration
 | Field | Type | Required | Description |
 | ----- | ---- | ------------ | ------------------------ |
-| bamqc_outformat | string | no | Choice of 'pdf' or 'html', default = 'pdf' |
+| bamqc_output_format | string | no | Choice of 'pdf' or 'html', default = 'pdf' |
 | bamqc_additional_options | string | no | Any additional options recognized by `bamqc` |
 
 #### Base resource allocation updaters
@@ -124,23 +132,23 @@ base_resource_update {
     ]
 }
 ```
-- To double memory for `run_CollectWgsMetrics_Picard` and triple memory for `run_stats_SAMtools` and `run_bamqc_Qualimap`:
+- To double memory for `run_CollectWgsMetrics_Picard` and triple memory for `run_statsSamples_SAMtools` and `run_bamqc_Qualimap`:
 ```Nextflow
 base_resource_update {
     memory = [
         ['run_CollectWgsMetrics_Picard', 2],
-        [['run_stats_SAMtools', 'run_bamqc_Qualimap'], 3]
+        [['run_statsSamples_SAMtools', 'run_bamqc_Qualimap'], 3]
     ]
 }
 ```
-- To double CPUs and memory for `run_CollectWgsMetrics_Picard` and double memory for `run_stats_SAMtools`:
+- To double CPUs and memory for `run_CollectWgsMetrics_Picard` and double memory for `run_statsSamples_SAMtools`:
 ```Nextflow
 base_resource_update {
     cpus = [
         ['run_CollectWgsMetrics_Picard', 2]
     ]
     memory = [
-        [['run_CollectWgsMetrics_Picard', 'run_stats_SAMtools'], 2]
+        [['run_CollectWgsMetrics_Picard', 'run_statsSamples_SAMtools'], 2]
     ]
 }
 ```

config/F16.config

@@ -3,7 +3,57 @@ process {
         cpus = 1
         memory = 250.MB
     }
-    withName: run_stats_SAMtools {
+    withName: assess_ReadQualityReadgroups_FastQC {
+        cpus = 2
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: assess_ReadQualityLibraries_FastQC {
+        cpus = 2
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: assess_ReadQualitySamples_FastQC {
+        cpus = 2
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: run_statsReadgroups_SAMtools {
+        cpus = 1
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: run_statsLibraries_SAMtools {
+        cpus = 1
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: run_statsSamples_SAMtools {
         cpus = 1
         memory = 1.GB
         retry_strategy {

config/F2.config

@@ -3,7 +3,57 @@ process {
         cpus = 1
         memory = 250.MB
     }
-    withName: run_stats_SAMtools {
+    withName: assess_ReadQualityReadgroups_FastQC {
+        cpus = 1
+        memory = 1500.MB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 2000.MB
+            }
+        }
+    }
+    withName: assess_ReadQualityLibraries_FastQC {
+        cpus = 1
+        memory = 1500.MB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 2000.MB
+            }
+        }
+    }
+    withName: assess_ReadQualitySamples_FastQC {
+        cpus = 1
+        memory = 1500.MB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 2000.MB
+            }
+        }
+    }
+    withName: run_statsReadgroups_SAMtools {
+        cpus = 1
+        memory = 1500.MB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 2000.MB
+            }
+        }
+    }
+    withName: run_statsLibraries_SAMtools {
+        cpus = 1
+        memory = 1500.MB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 2000.MB
+            }
+        }
+    }
+    withName: run_statsSamples_SAMtools {
         cpus = 1
         memory = 1500.MB
         retry_strategy {

config/F32.config

@@ -3,7 +3,57 @@ process {
         cpus = 1
         memory = 250.MB
     }
-    withName: run_stats_SAMtools {
+    withName: assess_ReadQualityReadgroups_FastQC {
+        cpus = 2
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: assess_ReadQualityLibraries_FastQC {
+        cpus = 2
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: assess_ReadQualitySamples_FastQC {
+        cpus = 2
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: run_statsReadgroups_SAMtools {
+        cpus = 1
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: run_statsLibraries_SAMtools {
+        cpus = 1
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: run_statsSamples_SAMtools {
         cpus = 1
         memory = 1.GB
         retry_strategy {

config/F4.config

@@ -3,7 +3,57 @@ process {
         cpus = 1
         memory = 250.MB
     }
-    withName: run_stats_SAMtools {
+    withName: assess_ReadQualityReadgroups_FastQC {
+        cpus = 2
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: assess_ReadQualityLibraries_FastQC {
+        cpus = 2
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: assess_ReadQualitySamples_FastQC {
+        cpus = 2
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: run_statsReadgroups_SAMtools {
+        cpus = 1
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 3.GB
+            }
+        }
+    }
+    withName: run_statsLibraries_SAMtools {
+        cpus = 1
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 3.GB
+            }
+        }
+    }
+    withName: run_statsSamples_SAMtools {
         cpus = 1
         memory = 1.GB
         retry_strategy {

config/F72.config

@@ -3,7 +3,57 @@ process {
         cpus = 1
         memory = 250.MB
     }
-    withName: run_stats_SAMtools {
+    withName: assess_ReadQualityReadgroups_FastQC {
+        cpus = 2
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: assess_ReadQualityLibraries_FastQC {
+        cpus = 2
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: assess_ReadQualitySamples_FastQC {
+        cpus = 2
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: run_statsReadgroups_SAMtools {
+        cpus = 1
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: run_statsLibraries_SAMtools {
+        cpus = 1
+        memory = 1.GB
+        retry_strategy {
+            memory {
+                strategy = 'add'
+                operand = 4.GB
+            }
+        }
+    }
+    withName: run_statsSamples_SAMtools {
         cpus = 1
         memory = 1.GB
         retry_strategy {