Filter and intersect

CHANGELOG.md

@@ -7,10 +7,14 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Added
 
+- Added a second BEDTools intersect step to allow filtering and overlapping in the same workflow.
+
 ### Changed
 
 ### Fixed
 
+- Updated PINTS version from 1.1.6 to 1.1.8 ([Fixes an issue where PINTS fails if one of the predictions was empty](https://github.com/hyulab/PINTS/issues/12))
+
 ### Deprecated
 
 ### Removed

conf/modules.config

@@ -175,7 +175,12 @@ process {
 
     withName: BEDTOOLS_INTERSECT_FILTER {
         ext.prefix = { "${meta.id}_filtered" }
-        ext.args = { params.no_overlap ? "-v": "" }
+        ext.args = "-v"
+    }
+
+    withName: BEDTOOLS_INTERSECT {
+        ext.prefix = { "${meta.id}_intersect" }
+        ext.args = "-wa"
     }
 
 }

conf/test.config

@@ -29,12 +29,6 @@ params {
     assay_type = "GROseq"
     skip_grohmm = true // FIXME Fails due to higher memory requirements
     tuning_file = "${projectDir}/tests/subworkflows/local/grohmm/tuningparams_small.csv"
-    filter_bed = "${projectDir}/tests/samplesheets/region.bed"
-}
-
-process {
-    // FIXME https://github.com/hyulab/PINTS/issues/12
-    withName: PINTS_CALLER {
-        ext.when = false
-    }
+    filter_bed = "${projectDir}/tests/samplesheets/unwanted_region.bed"
+    intersect_bed = "${projectDir}/tests/samplesheets/wanted_region.bed"
 }

modules.json

@@ -107,7 +107,7 @@
                     },
                     "pints/caller": {
                         "branch": "master",
-                        "git_sha": "5e34754d42cd2d5d248ca8673c0a53cdf5624905"
+                        "git_sha": "a8c6d338a76df3736215c93c27d9ccc7b88e9cf4"
                     },
                     "preseq/ccurve": {
                         "branch": "master",

nextflow.config

@@ -36,7 +36,7 @@ params {
     tuning_file                = null
 
     filter_bed                 = null
-    no_overlap                 = true
+    intersect_bed              = null
 
     // MultiQC options
     multiqc_config             = null

nextflow_schema.json

@@ -137,13 +137,16 @@
                 "filter_bed": {
                     "type": "string",
                     "fa_icon": "fas fa-filter",
-                    "pattern": "^\\S+\\.bed(\\.gz)?$"
+                    "pattern": "^\\S+\\.bed(\\.gz)?$",
+                    "description": "Undesired regions, that transcripts should not overlap with",
+                    "help_text": "Promoter regions, or gene regions are a good example"
                 },
-                "no_overlap": {
-                    "type": "boolean",
-                    "default": true,
-                    "fa_icon": "far fa-chart-bar",
-                    "description": "Only report those entries in A that have no overlap in B"
+                "intersect_bed": {
+                    "type": "string",
+                    "fa_icon": "fas fa-sort-amount-desc",
+                    "pattern": "^\\S+\\.bed(\\.gz)?$",
+                    "description": "Desired regions, that transcripts should overlap with",
+                    "help_text": "Histone Modifications(H3K4me1 and H3K27ac), or known TREs from the PINTS element matrix are examples"
                 }
             },
             "required": ["assay_type"],

subworkflows/local/transcript_identification.nf

@@ -9,6 +9,7 @@ include { CAT_CAT } from '../../modules/nf-core/cat/cat/main'
 include { BEDTOOLS_MERGE } from '../../modules/nf-core/bedtools/merge/main'
 include { BEDTOOLS_SORT } from '../../modules/nf-core/bedtools/sort/main'
 include { BEDTOOLS_INTERSECT as BEDTOOLS_INTERSECT_FILTER } from '../../modules/nf-core/bedtools/intersect/main'
+include { BEDTOOLS_INTERSECT } from '../../modules/nf-core/bedtools/intersect/main'
 
 include { HOMER_GROSEQ } from '../nf-core/homer/groseq/main'
 
@@ -58,12 +59,14 @@ workflow TRANSCRIPT_INDENTIFICATION {
     if(params.filter_bed) {
         ch_filter_bed = Channel.from(params.filter_bed)
         BEDTOOLS_INTERSECT_FILTER ( ch_identification_bed.combine(ch_filter_bed), "bed" )
+        ch_identification_bed = BEDTOOLS_INTERSECT_FILTER.out.intersect
         ch_versions = ch_versions.mix(BEDTOOLS_INTERSECT_FILTER.out.versions.first())
     }
-
-    // Use non-filtered bed files if we skip filtering
-    if(!params.filter_bed) {
-        ch_identification_bed = BEDTOOLS_INTERSECT_FILTER.out.intersect
+    if(params.intersect_bed) {
+        ch_intersect_bed = Channel.from(params.intersect_bed)
+        BEDTOOLS_INTERSECT ( ch_identification_bed.combine(ch_intersect_bed), "bed" )
+        ch_identification_bed = BEDTOOLS_INTERSECT.out.intersect
+        ch_versions = ch_versions.mix(BEDTOOLS_INTERSECT_FILTER.out.versions.first())
     }
 
     ch_identification_bed

tests/samplesheets/wanted_region.bed

@@ -0,0 +1,2 @@
+chr21	37267780	37267800
+chr21	41362070	41362130