Can have list of sequence labels to exclude from combined fasta

refs #68

Arvados-DCO-1.1-Signed-off-by: Peter Amstutz <peter.amstutz@curii.com>

4 files changed, 35 insertions, 9 deletions

diff --git a/bh20seqanalyzer/main.py b/bh20seqanalyzer/main.py
index 28b5e31..31ad4c4 100644
--- a/bh20seqanalyzer/main.py
+++ b/bh20seqanalyzer/main.py
@@ -125,7 +125,8 @@ def start_pangenome_analysis(api,
                              analysis_project,
                              pangenome_workflow_uuid,
                              validated_project,
-                             schema_ref):
+                             schema_ref,
+                             exclude_list):
     validated = arvados.util.list_all(api.collections().list, filters=[["owner_uuid", "=", validated_project]])
     inputobj = {
         "inputReads": [],
@@ -134,6 +135,10 @@ def start_pangenome_analysis(api,
         "metadataSchema": {
             "class": "File",
             "location": schema_ref
+        },
+        "exclude": {
+            "class": "File",
+            "location": exclude_list
         }
     }
     validated.sort(key=lambda v: v["portable_data_hash"])
@@ -213,6 +218,8 @@ def main():
     parser.add_argument('--pangenome-workflow-uuid', type=str, default='lugli-7fd4e-mqfu9y3ofnpnho1', help='')
     parser.add_argument('--fastq-workflow-uuid', type=str, default='lugli-7fd4e-2zp9q4jo5xpif9y', help='')
 
+    parser.add_argument('--exclude-list', type=str, default='keep:lugli-4zz18-tzzhcm6hrf8ci8d/exclude.txt', help='')
+
     parser.add_argument('--latest-result-collection', type=str, default='lugli-4zz18-z513nlpqm03hpca', help='')
     parser.add_argument('--kickoff', action="store_true")
     args = parser.parse_args()
@@ -229,7 +236,8 @@ def main():
                                  args.pangenome_analysis_project,
                                  args.pangenome_workflow_uuid,
                                  args.validated_project,
-                                 schema_ref)
+                                 schema_ref,
+                                 args.exclude_list)
         return
 
     logging.info("Starting up, monitoring %s for uploads" % (args.uploader_project))
@@ -250,7 +258,8 @@ def main():
                                      args.pangenome_analysis_project,
                                      args.pangenome_workflow_uuid,
                                      args.validated_project,
-                                     schema_ref)
+                                     schema_ref,
+                                     args.exclude_list)
 
         copy_most_recent_result(api,
                                 args.pangenome_analysis_project,
diff --git a/workflows/pangenome-generate/pangenome-generate.cwl b/workflows/pangenome-generate/pangenome-generate.cwl
index ad8b27f..9118cf8 100644
--- a/workflows/pangenome-generate/pangenome-generate.cwl
+++ b/workflows/pangenome-generate/pangenome-generate.cwl
@@ -9,6 +9,7 @@ inputs:
   metadata: File[]
   metadataSchema: File
   subjects: string[]
+  exclude: File?
   bin_widths:
     type: int[]
     default: [ 1, 4, 16, 64, 256, 1000, 4000, 16000]
@@ -47,6 +48,7 @@ steps:
     in:
       readsFA: inputReads
       subjects: subjects
+      exclude: exclude
     out: [relabeledSeqs, originalLabels]
     run: relabel-seqs.cwl
   dedup:
diff --git a/workflows/pangenome-generate/relabel-seqs.cwl b/workflows/pangenome-generate/relabel-seqs.cwl
index c1f17a4..367b9bf 100644
--- a/workflows/pangenome-generate/relabel-seqs.cwl
+++ b/workflows/pangenome-generate/relabel-seqs.cwl
@@ -3,10 +3,13 @@ class: CommandLineTool
 inputs:
   readsFA: File[]
   subjects: string[]
+  exclude:
+    type: File?
+    inputBinding: {position: 2}
   script:
     type: File
     default: {class: File, location: relabel-seqs.py}
-    inputBinding: {}
+    inputBinding: {position: 1}
 outputs:
   relabeledSeqs:
     type: File
diff --git a/workflows/pangenome-generate/relabel-seqs.py b/workflows/pangenome-generate/relabel-seqs.py
index 6b022a0..25b4a08 100644
--- a/workflows/pangenome-generate/relabel-seqs.py
+++ b/workflows/pangenome-generate/relabel-seqs.py
@@ -1,5 +1,6 @@
 import os
 import json
+import sys
 
 def readitems(stem):
     items = []
@@ -16,15 +17,26 @@ subjects = readitems("subs")
 relabeled_fasta = open("relabeledSeqs.fasta", "wt")
 original_labels = open("originalLabels.ttl", "wt")
 
+blacklist = set()
+if len(sys.argv) > 1:
+    with open(sys.argv[1]) as bl:
+        for l in bl:
+            blacklist.add(l.strip())
+
 for i, r in enumerate(reads):
     with open(r["path"], "rt") as fa:
-        label = fa.readline()
-        original_labels.write("<%s> <http://biohackathon.org/bh20-seq-schema/original_fasta_label> \"%s\" .\n" % (subjects[i], label[1:].strip().replace('"', '\\"')))
-        relabeled_fasta.write(">"+subjects[i]+"\n")
+        label = fa.readline().strip()
+        original_labels.write("<%s> <http://biohackathon.org/bh20-seq-schema/original_fasta_label> \"%s\" .\n" % (subjects[i], label[1:].replace('"', '\\"')))
+        skip = (subjects[i] in blacklist or label[1:] in blacklist)
+        if skip:
+            original_labels.write("<%s> <http://biohackathon.org/bh20-seq-schema/excluded_from_graph> \"true\"^^<http://www.w3.org/2001/XMLSchema#boolean> .\n" % (subjects[i]))
+        if not skip:
+            relabeled_fasta.write(">"+subjects[i]+"\n")
         data = fa.read(8096)
         while data:
-            relabeled_fasta.write(data)
+            if not skip:
+                relabeled_fasta.write(data)
             endswithnewline = data.endswith("\n")
             data = fa.read(8096)
-        if not endswithnewline:
+        if not skip and not endswithnewline:
             relabeled_fasta.write("\n")