Merge pull request #305 from ggabernet/cellranger_vdj

Updates cellranger_vdj
nf-core · Feb 27, 2024 · a0ad699 · a0ad699
2 parents ebbd620 + afa4c1d
commit a0ad699
Show file tree

Hide file tree

Showing 13 changed files with 229 additions and 84 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -9,6 +9,7 @@ and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.
 
 - [#294](https://github.com/nf-core/airrflow/pull/294) Merge template updates nf-core/tools v2.11.1
 - [#299](https://github.com/nf-core/airrflow/pull/299) Add profile for common NEB and TAKARA protocols
+- Add possibility to merge multi-lane samples when starting from fastq files
 
 ### `Fixed`
 

diff --git a/bin/check_samplesheet.py b/bin/check_samplesheet.py
@@ -160,9 +160,12 @@ def check_samplesheet(file_in, assembled):
 
         ## Check that sample ids are unique
         if len(tab["sample_id"]) != len(set(tab["sample_id"])):
-            print_error(
-                "Sample IDs are not unique! The sample IDs in the input samplesheet should be unique for each sample."
-            )
+            if assembled:
+                print_error(
+                    "Sample IDs are not unique! The sample IDs in the input samplesheet should be unique for each sample."
+                )
+            else:
+                print("WARNING: Sample IDs are not unique! FastQs with the same sample ID will be merged.")
 
         ## Check that pcr_target_locus is IG or TR
         for val in tab["pcr_target_locus"]:

diff --git a/bin/reveal_add_metadata.R b/bin/reveal_add_metadata.R
@@ -61,8 +61,12 @@ if (!("INPUTID" %in% names(opt))) {
 # Read metadata file
 metadata <- read.csv(opt$METADATA, sep = "\t", header = TRUE, stringsAsFactors = F)
 
+# Merging samples over multiple lanes introduces multi-rows per sample
+# We expect only one row per sample
 metadata <- metadata %>%
-    filter(sample_id == opt$INPUTID)
+    dplyr::filter(sample_id == opt$INPUTID) %>%
+    dplyr::select(!starts_with("filename_")) %>%
+    dplyr::distinct()
 
 if (nrow(metadata) != 1) {
     stop("Expecting nrow(metadata) == 1; nrow(metadata) == ", nrow(metadata), " found")
@@ -81,10 +85,7 @@ internal_fields <-
         "id",
         "filetype",
         "valid_single_cell",
-        "valid_pcr_target_locus",
-        "filename_R1",
-        "filename_R2",
-        "filename_I1"
+        "valid_pcr_target_locus"
     )
 metadata <- metadata[, !colnames(metadata) %in% internal_fields]
 

diff --git a/conf/test_10x_sc.config b/conf/test_10x_sc.config
@@ -18,7 +18,6 @@ params {
 
     // params
     mode = 'fastq'
-    sc_raw = true
     library_generation_method = 'sc_10x_genomics'
     clonal_threshold = 0
 

diff --git a/modules.json b/modules.json
@@ -5,6 +5,11 @@
         "https://github.com/nf-core/modules.git": {
             "modules": {
                 "nf-core": {
+                    "cat/fastq": {
+                        "branch": "master",
+                        "git_sha": "02fd5bd7275abad27aad32d5c852e0a9b1b98882",
+                        "installed_by": ["modules"]
+                    },
                     "cellranger/mkvdjref": {
                         "branch": "master",
                         "git_sha": "3f5420aa22e00bd030a2556dfdffc9e164ec0ec5",

diff --git a/modules/nf-core/cat/fastq/environment.yml b/modules/nf-core/cat/fastq/environment.yml
diff --git a/modules/nf-core/cat/fastq/main.nf b/modules/nf-core/cat/fastq/main.nf
diff --git a/modules/nf-core/cat/fastq/meta.yml b/modules/nf-core/cat/fastq/meta.yml
diff --git a/nextflow.config b/nextflow.config
@@ -120,7 +120,6 @@ params {
     // Single cell raw input options
     // -----------------------
     reference_10x = null
-    sc_raw = false
 
 
     // -----------------------
@@ -296,7 +295,6 @@ profiles {
     test_assembled_immcantation_devel_mm { includeConfig 'conf/test_assembled_immcantation_devel_mm.config' }
     test_nocluster { includeConfig 'conf/test_nocluster.config' }
     test_fetchimgt { includeConfig 'conf/test_fetchimgt.config' }
-    test_igblast { includeConfig 'conf/test_igblast.config' }
     test_10x_sc { includeConfig 'conf/test_10x_sc.config' }
     test_clontech_umi { includeConfig 'conf/test_clontech_umi.config' }
     test_nebnext_umi { includeConfig 'conf/test_nebnext_umi.config' }

diff --git a/nextflow_schema.json b/nextflow_schema.json
@@ -459,12 +459,6 @@
                     "description": "Path to the reference directory required by cellranger. Can either be directory or tar.gz.",
                     "help_text": "See for [IMGT](https://support.10xgenomics.com/single-cell-vdj/software/pipelines/latest/advanced/references#imgt) or [default](https://www.10xgenomics.com/support/software/cell-ranger/downloads).",
                     "fa_icon": "fas fa-database"
-                },
-                "sc_raw": {
-                    "type": "boolean",
-                    "description": "Must be given when raw single cell data should be run.",
-                    "help_text": "Must be given when raw single cell data should be run.",
-                    "fa_icon": "fas fa-database"
                 }
             },
             "help_text": "Options for running raw single cell data.",

diff --git a/subworkflows/local/fastq_input_check.nf b/subworkflows/local/fastq_input_check.nf
@@ -3,8 +3,7 @@
  */
 
 include { SAMPLESHEET_CHECK } from '../../modules/local/samplesheet_check'
-//TODO: when enchantr supports input samplesheet from raw sequencing, update code here to commented one.
-//include { VALIDATE_INPUT } from '../../modules/local/enchantr/validate_input'
+include { CAT_FASTQ } from '../../modules/nf-core/cat/fastq/main'
 
 workflow FASTQ_INPUT_CHECK {
     take:
@@ -15,22 +14,41 @@ workflow FASTQ_INPUT_CHECK {
         .tsv
         .splitCsv ( header:true, sep:'\t' )
         .map { create_fastq_channels(it) }
+        .dump (tag: 'fastq_channel_before_merge_samples')
+        .groupTuple(by: [0])
+        .dump(tag: 'fastq_channel_after_merge_samples_grouped')
+        .branch {
+            meta, fastqs ->
+                single: fastqs.size() == 1
+                    return [ meta, fastqs.flatten() ]
+                multiple: fastqs.size() > 1
+                    return [ meta, fastqs.flatten() ]
+        }
         .set { ch_reads }
-    // VALIDATE_INPUT(
-    //     samplesheet,
-    //     params.miairr,
-    //     params.collapseby,
-    //     params.cloneby
-    // )
+    ch_versions = SAMPLESHEET_CHECK.out.versions
+
+    // Merge multi-lane sample fastq for protocols except for 10x genomics (cellranger handles multi-fastq per sample)
+    if (params.library_generation_method == 'sc_10x_genomics') {
+
+        ch_merged_reads = ch_reads.single.mix( ch_reads.multiple )
+
+    } else {
+
+        CAT_FASTQ (
+            ch_reads.multiple
+        )
+        .reads
+        .mix( ch_reads.single )
+        .dump (tag: 'fastq_channel_after_merge_samples')
+        .set { ch_merged_reads }
 
-    // VALIDATE_INPUT.out.validated_input
-    //                     .splitCsv(header: true, sep:'\t')
-    //                     .map { get_meta(it) }
-    //                     .set{ ch_reads }
+        ch_versions = ch_versions.mix( CAT_FASTQ.out.versions )
+
+    }
 
     emit:
-    reads = ch_reads // channel: [ val(meta), [ reads ] ]
-    versions = SAMPLESHEET_CHECK.out.versions // channel: [ versions.yml ]
+    reads = ch_merged_reads // channel: [ val(meta), [ reads ] ]
+    versions = ch_versions // channel: [ versions.yml ]
     samplesheet = SAMPLESHEET_CHECK.out.tsv // tsv metadata file
 }
 
@@ -47,6 +65,7 @@ def create_fastq_channels(LinkedHashMap col) {
     meta.filetype           = "fastq"
     meta.single_cell        = col.single_cell.toLowerCase()
     meta.locus              = col.pcr_target_locus
+    meta.single_end         = false
 
     def array = []
     if (!file(col.filename_R1).exists()) {