main.nf 7.46 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"
Gervaise Henry's avatar
Gervaise Henry committed
23
params.astrocyte = false
Holly Ruess's avatar
Holly Ruess committed
24
25
params.outDir = "${baseDir}/output"

Gervaise Henry's avatar
Gervaise Henry committed
26
// Define list of files
Gervaise Henry's avatar
Gervaise Henry committed
27
28
tarList = Channel
  .fromPath( params.bcl )
29
bclCount = Channel
Gervaise Henry's avatar
Gervaise Henry committed
30
31
  .fromPath( params.bcl )
  .count()
Gervaise Henry's avatar
Gervaise Henry committed
32
33

// Define regular variables
34
pipelineVersion = "2.1.3"
Gervaise Henry's avatar
Gervaise Henry committed
35
name = params.name
36
37
38
39
designLocation = Channel
  .fromPath(params.designFile)
  .ifEmpty { exit 1, "design file not found: ${params.designFile}" }
outDir = params.outDir
Gervaise Henry's avatar
Gervaise Henry committed
40

Gervaise Henry's avatar
Gervaise Henry committed
41
42
43
44
45
// 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")
46
47
48
versionsScript = Channel.fromPath("$baseDir/scripts/generate_versions.py")
referencesScript = Channel.fromPath("$baseDir/scripts/generate_references.py")
versions_pythonScript = Channel.fromPath("$baseDir/scripts/versions_python.sh")
Gervaise Henry's avatar
Gervaise Henry committed
49
//versions_pigzScript = Channel.fromPath("$baseDir/scripts/versions_pigz.sh")
50
51
52
53
54
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
55
multiqcConf = Channel.fromPath("${baseDir}/configs/multiqc_config.yaml")
56
57
references = Channel.fromPath("${baseDir}/../docs/references.md")

Gervaise Henry's avatar
Gervaise Henry committed
58

Gervaise Henry's avatar
Gervaise Henry committed
59
60
61
62
/*
 * trackStart: track start of pipeline
 */
params.ci = false
63
params.dev = false
Gervaise Henry's avatar
Gervaise Henry committed
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
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}",
79
      "ci": ${params.ci},
Gervaise Henry's avatar
Gervaise Henry committed
80
      "dev": ${params.dev}}' \
Gervaise Henry's avatar
Gervaise Henry committed
81
82
83
84
85
  "https://xku43pcwnf.execute-api.us-east-1.amazonaws.com/ProdDeploy/pipeline-tracking"
  """
}


Gervaise Henry's avatar
Gervaise Henry committed
86
process checkDesignFile {
Holly Ruess's avatar
Holly Ruess committed
87
88

  tag "${name}"
89
  module 'python/3.6.1-2-anaconda'
Gervaise Henry's avatar
Gervaise Henry committed
90
91

  input:
Gervaise Henry's avatar
Gervaise Henry committed
92
    file check_designScript
93
    file versions_pythonScript
Holly Ruess's avatar
Holly Ruess committed
94
    file designLocation
Gervaise Henry's avatar
Gervaise Henry committed
95
96

  output:
Holly Ruess's avatar
Holly Ruess committed
97
98
    file("design.checked.csv") into designPaths
    file("design.checked.csv") into designCount
99
100
    file("version_pipeline.txt") into version_pipeline
    file("version_nextflow.txt") into version_nextflow
101
    file("version_python.txt") into version_python
Gervaise Henry's avatar
Gervaise Henry committed
102
103

  script:
Holly Ruess's avatar
Holly Ruess committed
104
105
    """
    hostname
106
    ulimit -u 16384
Holly Ruess's avatar
Holly Ruess committed
107
    ulimit -a
108
109
110
111
    noSpaceDesign=\$(echo "${designLocation}" | tr -d ' ')
    if [[ "\${noSpaceDesign}" != "${designLocation}" ]]; then
      mv "${designLocation}" "\${noSpaceDesign}"
    fi
Gervaise Henry's avatar
Gervaise Henry committed
112
    python3 check_design.py -d \${noSpaceDesign}
113
    bash versions_python.sh > version_python.txt
114
    echo "${workflow.nextflow.version}" > version_nextflow.txt
115
    echo "${pipelineVersion}" > version_pipeline.txt
Holly Ruess's avatar
Holly Ruess committed
116
117
    """

Gervaise Henry's avatar
Gervaise Henry committed
118
}
119
120
121
/* 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
122

123
process untarBCL {  
Holly Ruess's avatar
Holly Ruess committed
124

125
  tag "${tar.simpleName}"
126
  module 'pigz/2.4'
Gervaise Henry's avatar
Gervaise Henry committed
127
128

  input:
Gervaise Henry's avatar
Gervaise Henry committed
129
    file untarBCLScript
Gervaise Henry's avatar
Gervaise Henry committed
130
    //file versions_pigzScript
131
    each file(tar) from tarList
Gervaise Henry's avatar
Gervaise Henry committed
132
133

  output:
134
    file("*[!version_pigz.txt]") into bclPaths mode flatten
Gervaise Henry's avatar
Gervaise Henry committed
135
    //file("version_pigz.txt") into version_pigz
Gervaise Henry's avatar
Gervaise Henry committed
136
137

  script:
Holly Ruess's avatar
Holly Ruess committed
138
139
    """
    hostname
140
    ulimit -u 16384
Holly Ruess's avatar
Holly Ruess committed
141
    ulimit -a
142
    bash untarBCL.sh -t ${tar}
Gervaise Henry's avatar
Gervaise Henry committed
143
    #bash versions_pigz.sh > version_pigz.txt
Holly Ruess's avatar
Holly Ruess committed
144
145
    """

Gervaise Henry's avatar
Gervaise Henry committed
146
147
}

Gervaise Henry's avatar
Gervaise Henry committed
148

Gervaise Henry's avatar
Gervaise Henry committed
149
process mkfastq {
Holly Ruess's avatar
Holly Ruess committed
150

151
  tag "${bcl.simpleName}"
Holly Ruess's avatar
Holly Ruess committed
152
  publishDir "${outDir}/${task.process}", mode: 'copy', pattern: "{*/outs/**/*.fastq.gz}"
153
  queue '128GB,256GB,256GBv1,384GB'
154
  module 'cellranger/3.1.0:bcl2fastq/2.19.1'
Gervaise Henry's avatar
Gervaise Henry committed
155
156

  input:
157
158
    file versions_cellrangerScript
    file versions_bcl2fastqScript
159
    each file(bcl) from bclPaths.collect()
Holly Ruess's avatar
Holly Ruess committed
160
    file design from designPaths
Gervaise Henry's avatar
Gervaise Henry committed
161
162

  output:
163
164
165
    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
166
    file("**/outs/fastq_path/Stats/*") into bqcPaths
167
168
    file("version_cellranger.txt") into version_cellranger
    file("version_bcl2fastq.txt") into version_bcl2fastq
Gervaise Henry's avatar
Gervaise Henry committed
169
170

  script:
Holly Ruess's avatar
Holly Ruess committed
171
172
    """
    hostname
173
174
    ulimit -u 16384
    ulimit -a
175
    cellranger mkfastq --id=mkfastq_${bcl.simpleName} --run=${bcl} --csv=${design} --ignore-dual-index --use-bases-mask=Y*,I8n*,n*,Y*
176
177
178
179
180
    mkdir fq
    mkdir "fq/${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
181
182
    """

Gervaise Henry's avatar
Gervaise Henry committed
183
}
Gervaise Henry's avatar
Gervaise Henry committed
184

Gervaise Henry's avatar
Gervaise Henry committed
185

186
if (bclCount.value == 1) {
Holly Ruess's avatar
Holly Ruess committed
187

188
  process countDesign {
Holly Ruess's avatar
Holly Ruess committed
189
190

    tag "${name}"
191
    publishDir "${outDir}/${task.process}/${name}", mode: 'copy'
192
193

    input:
Gervaise Henry's avatar
Gervaise Henry committed
194
      file countDesignScript
Holly Ruess's avatar
Holly Ruess committed
195
196
      file fastqs from cellrangerCount.collect()
      file design from designCount
197
198

    output:
Holly Ruess's avatar
Holly Ruess committed
199
      file("Cellranger_Count_Design.csv") into CountDesign
200
201

    script:
Holly Ruess's avatar
Holly Ruess committed
202
      """
203
204
205
      hostname
      ulimit -u 16384
      ulimit -a
Gervaise Henry's avatar
Gervaise Henry committed
206
      bash countDesign.sh
Holly Ruess's avatar
Holly Ruess committed
207
208
      """

209
  }
Holly Ruess's avatar
Holly Ruess committed
210

211
212
}

Gervaise Henry's avatar
Gervaise Henry committed
213

214
process fastqc {
Holly Ruess's avatar
Holly Ruess committed
215

216
  tag "${bcl}"
217
  module 'fastqc/0.11.5:parallel'
218
219

  input:
220
221
222
223
    file fastqcScript    
    file versions_fastqcScript
    file(fastq) from fastqPaths
    each bcl from bclName
224
225

  output:
226
227
    file("*fastqc.zip") into fqcPaths mode flatten
    file("version_fastqc.txt") into version_fastqc
228
229

  script:
Holly Ruess's avatar
Holly Ruess committed
230
231
    """
    hostname
232
    ulimit -u 16384
Holly Ruess's avatar
Holly Ruess committed
233
    ulimit -a
234
    find *.fastq.gz -exec mv {} ${bcl}.{} \\;
Gervaise Henry's avatar
Gervaise Henry committed
235
    bash fastqc.sh
236
    bash versions_fastqc.sh > version_fastqc.txt
Holly Ruess's avatar
Holly Ruess committed
237
238
    """

239
240
}

241

242
process versions {
Holly Ruess's avatar
Holly Ruess committed
243
244

  tag "${name}"
245
  module 'python/3.6.1-2-anaconda:cellranger/3.1.0:bcl2fastq/2.19.1:fastqc/0.11.5:pandoc/2.7'
246
247

  input:
248
249
  file versionsScript
  file referencesScript
250
251
  file version_pipeline
  file version_nextflow
252
  file version_python
Gervaise Henry's avatar
Gervaise Henry committed
253
  //file version_pigz
254
255
256
257
  file version_cellranger
  file version_bcl2fastq
  file version_fastqc
  file references
258
259

  output:
Holly Ruess's avatar
Holly Ruess committed
260
    file("*.yaml") into yamlPaths
261
262

  script:
Holly Ruess's avatar
Holly Ruess committed
263
264
    """
    hostname
265
    ulimit -u 16384
Holly Ruess's avatar
Holly Ruess committed
266
    ulimit -a
267
268
    python3 generate_versions.py -f version_*.txt -o versions
    python3 generate_references.py -r ${references} -o references
Holly Ruess's avatar
Holly Ruess committed
269
270
    """

271
272
}

Gervaise Henry's avatar
Gervaise Henry committed
273

Gervaise Henry's avatar
Gervaise Henry committed
274
process multiqc {
Holly Ruess's avatar
Holly Ruess committed
275
276
277

  tag "${name}"
  publishDir "${outDir}/${task.process}/${name}", mode: 'copy', pattern: "{multiqc*}"
278
  module 'multiqc/1.7'
Gervaise Henry's avatar
Gervaise Henry committed
279
280

  input:
281
    file multiqcConf
Holly Ruess's avatar
Holly Ruess committed
282
    file bqc name "bqc/?/*" from bqcPaths.collect()
283
    file fqc name "fqc/?/*" from fqcPaths.collect()
Holly Ruess's avatar
Holly Ruess committed
284
    file yamlPaths
Gervaise Henry's avatar
Gervaise Henry committed
285
286

  output:
Holly Ruess's avatar
Holly Ruess committed
287
    file("multiqc_report.html") into mqcPaths
Gervaise Henry's avatar
Gervaise Henry committed
288
289

  script:
Holly Ruess's avatar
Holly Ruess committed
290
291
    """
    hostname
292
    ulimit -u 16384
Holly Ruess's avatar
Holly Ruess committed
293
    ulimit -a
294
295
    #export LC_ALL=C.UTF-8
    #export LANG=C.UTF-8
Holly Ruess's avatar
Holly Ruess committed
296
297
298
    multiqc -c ${multiqcConf} .
    """

Gervaise Henry's avatar
Gervaise Henry committed
299
}