From 4618a271193fa5cd5cb1e10d6f8acee9c6e8132a Mon Sep 17 00:00:00 2001
From: Michael R. Crusoe
Date: Mon, 4 May 2020 18:37:52 +0200
Subject: add pangenome browser prep

---
 .gitmodules                                        |  3 ++
 .../pangenome-generate/pangenome-generate.cwl      | 44 ++++++++++++++++++++++
 workflows/tools                                    |  1 +
 3 files changed, 48 insertions(+)
 create mode 100644 .gitmodules
 create mode 160000 workflows/tools

diff --git a/.gitmodules b/.gitmodules
new file mode 100644
index 0000000..c7d7f99
--- /dev/null
+++ b/.gitmodules
@@ -0,0 +1,3 @@
+[submodule "workflows/tools"]
+	path = workflows/tools
+	url = https://github.com/common-workflow-library/bio-cwl-tools.git
diff --git a/workflows/pangenome-generate/pangenome-generate.cwl b/workflows/pangenome-generate/pangenome-generate.cwl
index 6794e2d..05e3511 100644
--- a/workflows/pangenome-generate/pangenome-generate.cwl
+++ b/workflows/pangenome-generate/pangenome-generate.cwl
@@ -1,10 +1,22 @@
+#!/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
@@ -24,6 +36,12 @@ outputs:
   mergedMetadata:
     type: File
     outputSource: mergeMetadata/merged
+  indexed_paths:
+    type: File
+    outputSource: index_paths/indexed_paths
+  colinear_components:
+    type: File[]
+    outputSource: segment_components/colinear_components
 steps:
   relabel:
     in:
@@ -66,3 +84,29 @@ 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 ]
diff --git a/workflows/tools b/workflows/tools
new file mode 160000
index 0000000..45c2d6d
--- /dev/null
+++ b/workflows/tools
@@ -0,0 +1 @@
+Subproject commit 45c2d6dab199cb931b28c54863924ff3cf9079c1
-- 
cgit v1.2.3