main.nf 7.48 KB
Newer Older
Gervaise Henry's avatar
Gervaise Henry committed
1
#!/usr/bin/env nextflow
Gervaise Henry's avatar
Gervaise Henry committed
2
3
4
5
6
7
8
9
/*
main.nf
*
* --------------------------------------------------------------------------
* Licensed under MIT (https://git.biohpc.swmed.edu/BICF/Astrocyte/cellranger_mkfastq/blob/develop/LICENSE)
* --------------------------------------------------------------------------
*
*/
Gervaise Henry's avatar
Gervaise Henry committed
10

11
12
13
14
15
16
17
18
//  ########  ####  ######  ######## 
//  ##     ##  ##  ##    ## ##       
//  ##     ##  ##  ##       ##       
//  ########   ##  ##       ######   
//  ##     ##  ##  ##       ##       
//  ##     ##  ##  ##    ## ##       
//  ########  ####  ######  ##       

Gervaise Henry's avatar
Gervaise Henry committed
19
// Define input variables
20
params.name = "run"
21
22
params.bcl = "${baseDir}/../test_data/simple1/*.tar.gz"
params.designFile = "${baseDir}/../test_data/single1/cellranger-tiny-bcl-simple-1_2_0.csv"
23
params.mask = ""
Gervaise Henry's avatar
Gervaise Henry committed
24
params.astrocyte = false
Holly Ruess's avatar
Holly Ruess committed
25
26
params.outDir = "${baseDir}/output"

Gervaise Henry's avatar
Gervaise Henry committed
27
// Define list of files
Gervaise Henry's avatar
Gervaise Henry committed
28
29
tarList = Channel
  .fromPath( params.bcl )
30
bclCount = Channel
Gervaise Henry's avatar
Gervaise Henry committed
31
32
  .fromPath( params.bcl )
  .count()
Gervaise Henry's avatar
Gervaise Henry committed
33
34

// Define regular variables
35
pipelineVersion = "2.1.2"
Gervaise Henry's avatar
Gervaise Henry committed
36
name = params.name
37
38
39
designLocation = Channel
  .fromPath(params.designFile)
  .ifEmpty { exit 1, "design file not found: ${params.designFile}" }
40
mask = params.mask
41
outDir = params.outDir
Gervaise Henry's avatar
Gervaise Henry committed
42

Gervaise Henry's avatar
Gervaise Henry committed
43
44
45
46
47
// Define script files
check_designScript = Channel.fromPath("$baseDir/scripts/check_design.py")
untarBCLScript = Channel.fromPath("$baseDir/scripts/untarBCL.sh")
countDesignScript = Channel.fromPath("$baseDir/scripts/countDesign.sh")
fastqcScript = Channel.fromPath("$baseDir/scripts/fastqc.sh")
48
49
50
versionsScript = Channel.fromPath("$baseDir/scripts/generate_versions.py")
referencesScript = Channel.fromPath("$baseDir/scripts/generate_references.py")
versions_pythonScript = Channel.fromPath("$baseDir/scripts/versions_python.sh")
Gervaise Henry's avatar
Gervaise Henry committed
51
//versions_pigzScript = Channel.fromPath("$baseDir/scripts/versions_pigz.sh")
52
53
54
55
56
versions_cellrangerScript = Channel.fromPath("$baseDir/scripts/versions_cellranger.sh")
versions_bcl2fastqScript = Channel.fromPath("$baseDir/scripts/versions_bcl2fastq.sh")
versions_fastqcScript = Channel.fromPath("$baseDir/scripts/versions_fastqc.sh")

// Define report files
Gervaise Henry's avatar
Gervaise Henry committed
57
multiqcConf = Channel.fromPath("${baseDir}/configs/multiqc_config.yaml")
58
59
references = Channel.fromPath("${baseDir}/../docs/references.md")

Gervaise Henry's avatar
Gervaise Henry committed
60

Gervaise Henry's avatar
Gervaise Henry committed
61
62
63
64
/*
 * trackStart: track start of pipeline
 */
params.ci = false
65
params.dev = false
Gervaise Henry's avatar
Gervaise Henry committed
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
process trackStart {
  script:
  """
  hostname
  ulimit -a
  export https_proxy=\${http_proxy}
 
  curl -H 'Content-Type: application/json' -X PUT -d '{ \
      "sessionId": "${workflow.sessionId}", \
      "pipeline": "cellranger_mkfastq", \
      "pipelineVersion": "${pipelineVersion}", \
      "start": "${workflow.start}", \
      "astrocyte": ${params.astrocyte}, \
      "status": "started", \
      "nextflowVersion": "${workflow.nextflow.version}",
81
      "ci": ${params.ci},
Gervaise Henry's avatar
Gervaise Henry committed
82
      "dev": ${params.dev}}' \
Gervaise Henry's avatar
Gervaise Henry committed
83
84
85
86
87
  "https://xku43pcwnf.execute-api.us-east-1.amazonaws.com/ProdDeploy/pipeline-tracking"
  """
}


Gervaise Henry's avatar
Gervaise Henry committed
88
process checkDesignFile {
Holly Ruess's avatar
Holly Ruess committed
89
90

  tag "${name}"
91
  module 'python/3.6.1-2-anaconda'
Gervaise Henry's avatar
Gervaise Henry committed
92
93

  input:
Gervaise Henry's avatar
Gervaise Henry committed
94
    file check_designScript
95
    file versions_pythonScript
Holly Ruess's avatar
Holly Ruess committed
96
    file designLocation
Gervaise Henry's avatar
Gervaise Henry committed
97
98

  output:
Holly Ruess's avatar
Holly Ruess committed
99
100
    file("design.checked.csv") into designPaths
    file("design.checked.csv") into designCount
101
102
    file("version_pipeline.txt") into version_pipeline
    file("version_nextflow.txt") into version_nextflow
103
    file("version_python.txt") into version_python
Gervaise Henry's avatar
Gervaise Henry committed
104
105

  script:
Holly Ruess's avatar
Holly Ruess committed
106
107
    """
    hostname
108
    ulimit -u 16384
Holly Ruess's avatar
Holly Ruess committed
109
    ulimit -a
110
111
112
113
    noSpaceDesign=\$(echo "${designLocation}" | tr -d ' ')
    if [[ "\${noSpaceDesign}" != "${designLocation}" ]]; then
      mv "${designLocation}" "\${noSpaceDesign}"
    fi
Gervaise Henry's avatar
Gervaise Henry committed
114
    python3 check_design.py -d \${noSpaceDesign}
115
    bash versions_python.sh > version_python.txt
116
    echo "${workflow.nextflow.version}" > version_nextflow.txt
117
    echo "${pipelineVersion}" > version_pipeline.txt
Holly Ruess's avatar
Holly Ruess committed
118
119
    """

Gervaise Henry's avatar
Gervaise Henry committed
120
}
121
122
123
/* nextflow workflow manifest version calls that aren't compatible with Asrcocyte
    echo "${workflow.manifest.version}" > version_pipeline.txt
*/
Gervaise Henry's avatar
Gervaise Henry committed
124

125
process untarBCL {  
Holly Ruess's avatar
Holly Ruess committed
126

127
  tag "${tar.simpleName}"
128
  module 'pigz/2.4'
Gervaise Henry's avatar
Gervaise Henry committed
129
130

  input:
Gervaise Henry's avatar
Gervaise Henry committed
131
    file untarBCLScript
Gervaise Henry's avatar
Gervaise Henry committed
132
    //file versions_pigzScript
133
    each file(tar) from tarList
Gervaise Henry's avatar
Gervaise Henry committed
134
135

  output:
136
    file("*[!version_pigz.txt]") into bclPaths mode flatten
Gervaise Henry's avatar
Gervaise Henry committed
137
    //file("version_pigz.txt") into version_pigz
Gervaise Henry's avatar
Gervaise Henry committed
138
139

  script:
Holly Ruess's avatar
Holly Ruess committed
140
141
    """
    hostname
142
    ulimit -u 16384
Holly Ruess's avatar
Holly Ruess committed
143
    ulimit -a
144
    bash untarBCL.sh -t ${tar}
Gervaise Henry's avatar
Gervaise Henry committed
145
    #bash versions_pigz.sh > version_pigz.txt
Holly Ruess's avatar
Holly Ruess committed
146
147
    """

Gervaise Henry's avatar
Gervaise Henry committed
148
149
}

Gervaise Henry's avatar
Gervaise Henry committed
150

Gervaise Henry's avatar
Gervaise Henry committed
151
process mkfastq {
Holly Ruess's avatar
Holly Ruess committed
152

153
  tag "${bcl.simpleName}"
Holly Ruess's avatar
Holly Ruess committed
154
  publishDir "${outDir}/${task.process}", mode: 'copy', pattern: "{*/outs/**/*.fastq.gz}"
155
  queue '128GB,256GB,256GBv1,384GB'
156
  module 'cellranger/3.1.0:bcl2fastq/2.19.1'
Gervaise Henry's avatar
Gervaise Henry committed
157
158

  input:
159
160
    file versions_cellrangerScript
    file versions_bcl2fastqScript
161
    each file(bcl) from bclPaths.collect()
Holly Ruess's avatar
Holly Ruess committed
162
    file design from designPaths
Gervaise Henry's avatar
Gervaise Henry committed
163
164

  output:
165
166
167
    file("fq/${bcl.simpleName}/*.fastq.gz") into fastqPaths
    val "${bcl.simpleName}" into bclName
    file("**/outs/**/*.fastq.gz") into cellrangerCount mode flatten
Gervaise Henry's avatar
Gervaise Henry committed
168
    file("**/outs/fastq_path/Stats/*") into bqcPaths
169
170
    file("version_cellranger.txt") into version_cellranger
    file("version_bcl2fastq.txt") into version_bcl2fastq
Gervaise Henry's avatar
Gervaise Henry committed
171
172

  script:
Holly Ruess's avatar
Holly Ruess committed
173
174
    """
    hostname
175
176
    ulimit -u 16384
    ulimit -a
177
    cellranger mkfastq --id=mkfastq_${bcl.simpleName} --run=${bcl} --csv=${design} --ignore-dual-index ${mask}
178
179
180
181
182
    mkdir fq
    mkdir "fq/${bcl.simpleName}"
    find . -name "*.fastq.gz" -exec cp {} fq/${bcl.simpleName}/ \\;
    bash versions_cellranger.sh > version_cellranger.txt
    bash versions_bcl2fastq.sh > version_bcl2fastq.txt
Holly Ruess's avatar
Holly Ruess committed
183
184
    """

Gervaise Henry's avatar
Gervaise Henry committed
185
}
Gervaise Henry's avatar
Gervaise Henry committed
186

Gervaise Henry's avatar
Gervaise Henry committed
187

188
if (bclCount.value == 1) {
Holly Ruess's avatar
Holly Ruess committed
189

190
  process countDesign {
Holly Ruess's avatar
Holly Ruess committed
191
192

    tag "${name}"
193
    publishDir "${outDir}/${task.process}/${name}", mode: 'copy'
194
195

    input:
Gervaise Henry's avatar
Gervaise Henry committed
196
      file countDesignScript
Holly Ruess's avatar
Holly Ruess committed
197
198
      file fastqs from cellrangerCount.collect()
      file design from designCount
199
200

    output:
Holly Ruess's avatar
Holly Ruess committed
201
      file("Cellranger_Count_Design.csv") into CountDesign
202
203

    script:
Holly Ruess's avatar
Holly Ruess committed
204
      """
205
206
207
      hostname
      ulimit -u 16384
      ulimit -a
Gervaise Henry's avatar
Gervaise Henry committed
208
      bash countDesign.sh
Holly Ruess's avatar
Holly Ruess committed
209
210
      """

211
  }
Holly Ruess's avatar
Holly Ruess committed
212

213
214
}

Gervaise Henry's avatar
Gervaise Henry committed
215

216
process fastqc {
Holly Ruess's avatar
Holly Ruess committed
217

218
  tag "${bcl}"
219
  module 'fastqc/0.11.5:parallel'
220
221

  input:
222
223
224
225
    file fastqcScript    
    file versions_fastqcScript
    file(fastq) from fastqPaths
    each bcl from bclName
226
227

  output:
228
229
    file("*fastqc.zip") into fqcPaths mode flatten
    file("version_fastqc.txt") into version_fastqc
230
231

  script:
Holly Ruess's avatar
Holly Ruess committed
232
233
    """
    hostname
234
    ulimit -u 16384
Holly Ruess's avatar
Holly Ruess committed
235
    ulimit -a
236
    find *.fastq.gz -exec mv {} ${bcl}.{} \\;
Gervaise Henry's avatar
Gervaise Henry committed
237
    bash fastqc.sh
238
    bash versions_fastqc.sh > version_fastqc.txt
Holly Ruess's avatar
Holly Ruess committed
239
240
    """

241
242
}

243

244
process versions {
Holly Ruess's avatar
Holly Ruess committed
245
246

  tag "${name}"
247
  module 'python/3.6.1-2-anaconda:cellranger/3.1.0:bcl2fastq/2.19.1:fastqc/0.11.5:pandoc/2.7'
248
249

  input:
250
251
  file versionsScript
  file referencesScript
252
253
  file version_pipeline
  file version_nextflow
254
  file version_python
Gervaise Henry's avatar
Gervaise Henry committed
255
  //file version_pigz
256
257
258
259
  file version_cellranger
  file version_bcl2fastq
  file version_fastqc
  file references
260
261

  output:
Holly Ruess's avatar
Holly Ruess committed
262
    file("*.yaml") into yamlPaths
263
264

  script:
Holly Ruess's avatar
Holly Ruess committed
265
266
    """
    hostname
267
    ulimit -u 16384
Holly Ruess's avatar
Holly Ruess committed
268
    ulimit -a
269
270
    python3 generate_versions.py -f version_*.txt -o versions
    python3 generate_references.py -r ${references} -o references
Holly Ruess's avatar
Holly Ruess committed
271
272
    """

273
274
}

Gervaise Henry's avatar
Gervaise Henry committed
275

Gervaise Henry's avatar
Gervaise Henry committed
276
process multiqc {
Holly Ruess's avatar
Holly Ruess committed
277
278
279

  tag "${name}"
  publishDir "${outDir}/${task.process}/${name}", mode: 'copy', pattern: "{multiqc*}"
280
  module 'multiqc/1.7'
Gervaise Henry's avatar
Gervaise Henry committed
281
282

  input:
283
    file multiqcConf
Holly Ruess's avatar
Holly Ruess committed
284
    file bqc name "bqc/?/*" from bqcPaths.collect()
285
    file fqc name "fqc/?/*" from fqcPaths.collect()
Holly Ruess's avatar
Holly Ruess committed
286
    file yamlPaths
Gervaise Henry's avatar
Gervaise Henry committed
287
288

  output:
Holly Ruess's avatar
Holly Ruess committed
289
    file("multiqc_report.html") into mqcPaths
Gervaise Henry's avatar
Gervaise Henry committed
290
291

  script:
Holly Ruess's avatar
Holly Ruess committed
292
293
    """
    hostname
294
    ulimit -u 16384
Holly Ruess's avatar
Holly Ruess committed
295
    ulimit -a
296
297
    #export LC_ALL=C.UTF-8
    #export LANG=C.UTF-8
Holly Ruess's avatar
Holly Ruess committed
298
299
300
    multiqc -c ${multiqcConf} .
    """

Gervaise Henry's avatar
Gervaise Henry committed
301
}