aboutsummaryrefslogtreecommitdiff
path: root/workflows/pangenome-generate/pangenome-generate.cwl
diff options
context:
space:
mode:
authorPjotr Prins2020-05-12 13:55:55 -0500
committerPjotr Prins2020-05-12 13:55:55 -0500
commit8e9042879db7fe061f3d18737f4102b5fd0b8c65 (patch)
tree978645b86e805c218451beaf194bfd97f2fbd9d8 /workflows/pangenome-generate/pangenome-generate.cwl
parent40ca03045bbc4f6fb1258acb6f42a60ee5532e0d (diff)
parent55bab826fda6c6bc3190476f8e2c61e51ef092d6 (diff)
downloadbh20-seq-resource-8e9042879db7fe061f3d18737f4102b5fd0b8c65.tar.gz
bh20-seq-resource-8e9042879db7fe061f3d18737f4102b5fd0b8c65.tar.lz
bh20-seq-resource-8e9042879db7fe061f3d18737f4102b5fd0b8c65.zip
Merge branch 'master' of github.com:arvados/bh20-seq-resource
Diffstat (limited to 'workflows/pangenome-generate/pangenome-generate.cwl')
-rw-r--r--workflows/pangenome-generate/pangenome-generate.cwl83
1 files changed, 70 insertions, 13 deletions
diff --git a/workflows/pangenome-generate/pangenome-generate.cwl b/workflows/pangenome-generate/pangenome-generate.cwl
index 6794e2d..1d49904 100644
--- a/workflows/pangenome-generate/pangenome-generate.cwl
+++ b/workflows/pangenome-generate/pangenome-generate.cwl
@@ -1,17 +1,29 @@
+#!/usr/bin/env cwl-runner
cwlVersion: v1.1
class: Workflow
+requirements:
+ ScatterFeatureRequirement: {}
+ StepInputExpressionRequirement: {}
inputs:
inputReads: File[]
metadata: File[]
metadataSchema: File
subjects: string[]
+ bin_widths:
+ type: int[]
+ default: [ 1, 4, 16, 64, 256, 1000, 4000, 16000]
+ doc: width of each bin in basepairs along the graph vector
+ cells_per_file:
+ type: int
+ default: 100
+ doc: Cells per file on component_segmentation
outputs:
odgiGraph:
type: File
outputSource: buildGraph/odgiGraph
odgiPNG:
type: File
- outputSource: vizGraph/odgiPNG
+ outputSource: vizGraph/graph_image
seqwishGFA:
type: File
outputSource: induceGraph/seqwishGFA
@@ -20,10 +32,16 @@ outputs:
outputSource: odgi2rdf/rdf
readsMergeDedup:
type: File
- outputSource: dedup/readsMergeDedup
+ outputSource: dedup/reads_dedup
mergedMetadata:
type: File
outputSource: mergeMetadata/merged
+ indexed_paths:
+ type: File
+ outputSource: index_paths/indexed_paths
+ colinear_components:
+ type: Directory
+ outputSource: segment_components/colinear_components
steps:
relabel:
in:
@@ -32,17 +50,21 @@ steps:
out: [relabeledSeqs, originalLabels]
run: relabel-seqs.cwl
dedup:
- in: {readsFA: relabel/relabeledSeqs}
- out: [readsMergeDedup, dups]
- run: seqkit-rmdup.cwl
+ in: {reads: relabel/relabeledSeqs}
+ out: [reads_dedup, dups]
+ run: ../tools/seqkit/seqkit_rmdup.cwl
overlapReads:
- in: {readsFA: dedup/readsMergeDedup}
- out: [readsPAF]
- run: minimap2.cwl
+ in:
+ target: dedup/reads_dedup
+ query: dedup/reads_dedup
+ outputCIGAR:
+ default: true
+ out: [alignments]
+ run: ../tools/minimap2/minimap2_paf.cwl
induceGraph:
in:
- readsFA: dedup/readsMergeDedup
- readsPAF: overlapReads/readsPAF
+ readsFA: dedup/reads_dedup
+ readsPAF: overlapReads/alignments
out: [seqwishGFA]
run: seqwish.cwl
buildGraph:
@@ -50,9 +72,18 @@ steps:
out: [odgiGraph]
run: odgi-build.cwl
vizGraph:
- in: {inputODGI: buildGraph/odgiGraph}
- out: [odgiPNG]
- run: odgi-viz.cwl
+ in:
+ sparse_graph_index: buildGraph/odgiGraph
+ width:
+ default: 50000
+ height:
+ default: 500
+ path_per_row:
+ default: true
+ path_height:
+ default: 4
+ out: [graph_image]
+ run: ../tools/odgi/odgi_viz.cwl
odgi2rdf:
in: {odgi: buildGraph/odgiGraph}
out: [rdf]
@@ -66,3 +97,29 @@ steps:
originalLabels: relabel/originalLabels
out: [merged]
run: merge-metadata.cwl
+ bin_paths:
+ run: ../tools/odgi/odgi_bin.cwl
+ in:
+ sparse_graph_index: buildGraph/odgiGraph
+ bin_width: bin_widths
+ scatter: bin_width
+ out: [ bins, pangenome_sequence ]
+ index_paths:
+ label: Create path index
+ run : ../tools/odgi/odgi_pathindex.cwl
+ in:
+ sparse_graph_index: buildGraph/odgiGraph
+ out: [ indexed_paths ]
+ segment_components:
+ label: Run component segmentation
+ run: ../tools/graph-genome-segmentation/component_segmentation.cwl
+ in:
+ bins: bin_paths/bins
+ cells_per_file: cells_per_file
+ pangenome_sequence:
+ source: bin_paths/pangenome_sequence
+ valueFrom: $(self[0])
+ # the bin_paths step is scattered over the bin_width array, but always using the same sparse_graph_index
+ # the pangenome_sequence that is extracted is exactly the same for the same sparse_graph_index
+ # regardless of bin_width, so we take the first pangenome_sequence as input for this step
+ out: [ colinear_components ]