implement maskprimers align umi

nf-core · Sep 27, 2024 · 0432de0 · 0432de0
1 parent b0afe90
commit 0432de0
Show file tree

Hide file tree

Showing 10 changed files with 179 additions and 57 deletions.
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
@@ -64,7 +64,8 @@ jobs:
             "test_nebnext_umi",
             "test_rnaseq_bulk",
             "test_rnaseq_sc",
-            "test_maskprimers_extract"
+            "test_maskprimers_extract",
+            "test_maskprimers_align",
           ]
       fail-fast: false
     steps:

diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -17,6 +17,7 @@ and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.
 - [#344](https://github.com/nf-core/airrflow/pull/344) Simplified pRESTO Maskprimers score and Maskprimers extract processes.
 - [#344](https://github.com/nf-core/airrflow/pull/344) Fix clustersets approach, consider both reads clusters annotation.
 - [#344](https://github.com/nf-core/airrflow/pull/344) Allow for partial alignments in MakeDB for mouse TCR data, to bypass junction check for TRAV germlines with additional positions (e.g. 84A).
+- [#344](https://github.com/nf-core/airrflow/pull/344) Allow for maskprimers align with UMI on both V gene and C gene primers.
 
 ### `Dependencies`
 

diff --git a/conf/clontech_umi_bcr.config b/conf/clontech_umi_bcr.config
@@ -30,7 +30,7 @@ params {
 
 
     // Mask primer options
-    maskprimers_align = true
+    maskprimers_align_race = true
     primer_r2_extract_len = 7
     primer_r1_mask_mode = 'cut'
     primer_r2_mask_mode = 'cut'

diff --git a/conf/clontech_umi_tcr.config b/conf/clontech_umi_tcr.config
@@ -30,7 +30,7 @@ params {
 
 
     // Mask primer options
-    maskprimers_align = true
+    maskprimers_align_race = true
     primer_r2_extract_len = 7
     primer_r1_mask_mode = 'cut'
     primer_r2_mask_mode = 'cut'

diff --git a/conf/modules.config b/conf/modules.config
@@ -148,7 +148,7 @@ process {
         ext.args2 = '-f ID PRIMER ERROR'
     }
 
-    withName: PRESTO_MASKPRIMERS_ALIGN {
+    withName: PRESTO_ALIGN_PRIMERS {
         publishDir = [
             path: { "${params.outdir}/presto/02-maskprimers/${meta.id}" },
             mode: params.publish_dir_mode,
@@ -177,7 +177,7 @@ process {
         ext.args2 = '-f ID PRIMER ERROR PRSTART'
     }
 
-    withName: 'PRESTO_MASKPRIMERS_EXTRACT_R1' {
+    withName: 'PRESTO_MASKPRIMERS_ALIGN_UMI_R1' {
         publishDir = [
             path: { "${params.outdir}/presto/02-maskprimers/${meta.id}" },
             mode: params.publish_dir_mode,
@@ -186,13 +186,12 @@ process {
         ext.args2 = '-f ID PRIMER ERROR PRSTART'
     }
 
-    withName: 'PRESTO_MASKPRIMERS_EXTRACT_R2' {
+    withName: 'PRESTO_MASKPRIMERS_ALIGN_UMI_R2' {
         publishDir = [
             path: { "${params.outdir}/presto/02-maskprimers/${meta.id}" },
             mode: params.publish_dir_mode,
             pattern: "*{txt,log,tab}"
         ]
-        ext.args = '--pf EXTRACT_PRIMER'
         ext.args2 = '-f ID PRIMER ERROR PRSTART'
     }
 
@@ -326,7 +325,7 @@ process {
         ext.args3 = '-f ID BARCODE SEQCOUNT PRIMER PRCOUNT PRCONS PRFREQ CONSCOUNT ERROR'
     }
 
-    withName: PRESTO_BUILDCONSENSUS_ALIGN {
+    withName: PRESTO_BUILDCONSENSUS_ALIGN_RACE {
         publishDir = [
             path: { "${params.outdir}/presto/06-build-consensus/${meta.id}" },
             mode: params.publish_dir_mode,

diff --git a/conf/test_maskprimers_align.config b/conf/test_maskprimers_align.config
@@ -0,0 +1,39 @@
+/*
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+    Nextflow config file for running minimal tests
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+    Defines input files and everything required to run a fast and simple pipeline test.
+
+    Use as follows:
+        nextflow run nf-core/airrflow -profile test_maskprimers_align,<docker/singularity> --outdir <OUTDIR>
+
+----------------------------------------------------------------------------------------
+*/
+
+params {
+    config_profile_name        = 'Test profile'
+    config_profile_description = 'Minimal test dataset to check pipeline function'
+
+    // Limit resources so that this can run on GitHub Actions
+    max_cpus   = 2
+    max_memory = '6.GB'
+    max_time   = '6.h'
+
+    // Input data
+    input = pipelines_testdata_base_path + 'testdata-bcr/Metadata_test_airr.tsv'
+    cprimers = pipelines_testdata_base_path + 'testdata-bcr/C_primers.fasta'
+    vprimers = pipelines_testdata_base_path + 'testdata-bcr/V_primers.fasta'
+    reference_fasta = pipelines_testdata_base_path + 'database-cache/imgtdb_base.zip'
+    reference_igblast = pipelines_testdata_base_path + 'database-cache/igblast_base.zip'
+
+    mode = 'fastq'
+
+    library_generation_method = 'specific_pcr_umi'
+    cprimer_position = 'R1'
+    umi_length = 8
+    umi_start = 6
+    umi_position = 'R1'
+    index_file = true
+    isotype_column = 'c_primer'
+    maskprimers_align = true
+}
diff --git a/modules/local/presto/presto_maskprimers_align.nf b/modules/local/presto/presto_maskprimers_align.nf
@@ -12,8 +12,10 @@ process PRESTO_MASKPRIMERS_ALIGN {
     tuple val(meta), path(R1)
     path(primers)
     val(max_len)
+    val(barcode)
     val(max_error)
     val(mask_mode)
+    val(reverse_primers)
     val(suffix)
 
     output:
@@ -23,6 +25,8 @@ process PRESTO_MASKPRIMERS_ALIGN {
     path "versions.yml" , emit: versions
 
     script:
+    def barcode = barcode ? '--barcode' : ''
+    def revpr = reverse_primers ? '--revpr' : ''
     def args = task.ext.args?: ''
     def args2 = task.ext.args2?: ''
     """
@@ -32,6 +36,7 @@ process PRESTO_MASKPRIMERS_ALIGN {
     --maxlen ${max_len} \\
     --maxerror ${max_error} \\
     --mode ${mask_mode} \\
+    $barcode \\
     $args \\
     --outname ${meta.id}_${suffix} \\
     --log ${meta.id}_${suffix}.log > ${meta.id}_command_log_${suffix}.txt

diff --git a/nextflow.config b/nextflow.config
@@ -56,6 +56,7 @@ params {
     primer_r2_maxerror = 0.2
     primer_r1_mask_mode = 'cut'
     primer_r2_mask_mode = 'cut'
+    maskprimers_align_race = false
     maskprimers_align = false
     maskprimers_extract = false
     primer_r1_extract_len = 0
@@ -315,6 +316,7 @@ profiles {
     test_10x_sc { includeConfig 'conf/test_10x_sc.config' }
     test_clontech_umi { includeConfig 'conf/test_clontech_umi.config' }
     test_maskprimers_extract { includeConfig 'conf/test_maskprimers_extract.config' }
+    test_maskprimers_align {includeConfig 'conf/test_maskprimers_align.config' }
     test_nebnext_umi { includeConfig 'conf/test_nebnext_umi.config' }
     test_rnaseq_bulk { includeConfig 'conf/test_rnaseq_bulk.config' }
     test_rnaseq_sc { includeConfig 'conf/test_rnaseq_sc.config' }