aboutsummaryrefslogtreecommitdiff
path: root/workflows/pangenome-generate/pangenome-generate_spoa.cwl
diff options
context:
space:
mode:
authorPeter Amstutz2020-08-05 16:06:11 -0400
committerGitHub2020-08-05 16:06:11 -0400
commitfdb1b012fc04ee07f401541e181e28fe442c9454 (patch)
tree8486db1087692dffcea9d93814e436d9cf150b47 /workflows/pangenome-generate/pangenome-generate_spoa.cwl
parent86f31ef60f65a820bf9ac25c3fc01c88f2a9ebfe (diff)
parent2d20bf90497588a297ca98a78ee0fbbcadf95569 (diff)
downloadbh20-seq-resource-fdb1b012fc04ee07f401541e181e28fe442c9454.tar.gz
bh20-seq-resource-fdb1b012fc04ee07f401541e181e28fe442c9454.tar.lz
bh20-seq-resource-fdb1b012fc04ee07f401541e181e28fe442c9454.zip
Merge pull request #99 from AndreaGuarracino/patch-2
several fixes in the website, added links to video talk and poster, new pangenome generation workflow
Diffstat (limited to 'workflows/pangenome-generate/pangenome-generate_spoa.cwl')
-rw-r--r--workflows/pangenome-generate/pangenome-generate_spoa.cwl122
1 files changed, 122 insertions, 0 deletions
diff --git a/workflows/pangenome-generate/pangenome-generate_spoa.cwl b/workflows/pangenome-generate/pangenome-generate_spoa.cwl
new file mode 100644
index 0000000..958ffb6
--- /dev/null
+++ b/workflows/pangenome-generate/pangenome-generate_spoa.cwl
@@ -0,0 +1,122 @@
+#!/usr/bin/env cwl-runner
+cwlVersion: v1.1
+class: Workflow
+requirements:
+ ScatterFeatureRequirement: {}
+ StepInputExpressionRequirement: {}
+inputs:
+ inputReads: File[]
+ metadata: File[]
+ metadataSchema: File
+ subjects: string[]
+ exclude: File?
+ 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/graph_image
+ spoaGFA:
+ type: File
+ outputSource: induceGraph/spoaGFA
+ odgiRDF:
+ type: File
+ outputSource: odgi2rdf/rdf
+ readsMergeDedup:
+ type: File
+ 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:
+ readsFA: inputReads
+ subjects: subjects
+ exclude: exclude
+ out: [relabeledSeqs, originalLabels]
+ run: relabel-seqs.cwl
+ dedup:
+ in: {reads: relabel/relabeledSeqs}
+ out: [reads_dedup, dups]
+ run: ../tools/seqkit/seqkit_rmdup.cwl
+ sort_by_quality_and_len:
+ in: {reads: dedup/reads_dedup}
+ out: [reads_sorted_by_quality_and_len]
+ run: sort_fasta_by_quality_and_len.cwl
+ induceGraph:
+ in:
+ readsFA: sort_by_quality_and_len/reads_sorted_by_quality_and_len
+ out: [spoaGFA]
+ run: spoa.cwl
+ buildGraph:
+ in: {inputGFA: induceGraph/spoaGFA}
+ out: [odgiGraph]
+ run: odgi-build-from-spoa-gfa.cwl
+ vizGraph:
+ 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]
+ run: odgi_to_rdf.cwl
+ mergeMetadata:
+ in:
+ metadata: metadata
+ metadataSchema: metadataSchema
+ subjects: subjects
+ dups: dedup/dups
+ 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 ]