From 9ddcfeacb3191638f42b08af999889d867f0f81c Mon Sep 17 00:00:00 2001
From: Peter Amstutz
Date: Mon, 20 Apr 2020 14:57:25 -0400
Subject: Better handling of duplicate sequences

Also save original fasta label in metadata
---
 workflows/pangenome-generate/pangenome-generate.cwl | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

(limited to 'workflows/pangenome-generate/pangenome-generate.cwl')

diff --git a/workflows/pangenome-generate/pangenome-generate.cwl b/workflows/pangenome-generate/pangenome-generate.cwl
index 896f936..0cb1368 100644
--- a/workflows/pangenome-generate/pangenome-generate.cwl
+++ b/workflows/pangenome-generate/pangenome-generate.cwl
@@ -26,15 +26,11 @@ steps:
     in:
       readsFA: inputReads
       subjects: subjects
-    out: [relabeledSeqs]
+    out: [relabeledSeqs, originalLabels]
     run: relabel-seqs.cwl
-  common:
-    in: {readsFA: relabel/relabeledSeqs}
-    out: [duplicatedReads]
-    run: seqkit-common.cwl
   dedup:
     in: {readsFA: relabel/relabeledSeqs}
-    out: [readsMergeDedup]
+    out: [readsMergeDedup, dups]
     run: seqkit-rmdup.cwl
   overlapReads:
     in: {readsFA: dedup/readsMergeDedup}
@@ -63,5 +59,7 @@ steps:
       metadata: metadata
       metadataSchema: metadataSchema
       subjects: subjects
+      dups: dedup/dups
+      originalLabels: relabel/originalLabels
     out: [merged]
     run: merge-metadata.cwl
-- 
cgit v1.2.3