X-Git-Url: https://git.arvados.org/arvados.git/blobdiff_plain/42acc196ca8f5908f3cc214d5f715eb1db82c7a8..ca06cfbda0e84d469f7810a280cfa4dfa8997260:/crunch_scripts/arvados-bcbio-nextgen.py

diff --git a/crunch_scripts/arvados-bcbio-nextgen.py b/crunch_scripts/arvados-bcbio-nextgen.py
index 8781087667..b7e19ecddb 100755
--- a/crunch_scripts/arvados-bcbio-nextgen.py
+++ b/crunch_scripts/arvados-bcbio-nextgen.py
@@ -1,10 +1,15 @@
 #!/usr/bin/python
+# Copyright (C) The Arvados Authors. All rights reserved.
+#
+# SPDX-License-Identifier: Apache-2.0
 
 import arvados
 import subprocess
-import subst
+import crunchutil.subst as subst
 import shutil
 import os
+import sys
+import time
 
 if len(arvados.current_task()['parameters']) > 0:
     p = arvados.current_task()['parameters']
@@ -55,23 +60,25 @@ with open("/usr/local/share/bcbio-nextgen/galaxy/tool_data_table_conf.xml", "w")
 os.mkdir("/usr/local/share/bcbio-nextgen/galaxy/tool-data")
 
 with open("/usr/local/share/bcbio-nextgen/galaxy/tool-data/bowtie2_indices.loc", "w") as f:
-    f.write(subst.do_substitution(p, "GRCh37\tGRCh37\tHuman (GRCh37)\t$(dir $(bowtie2_indices))"))
+    f.write(subst.do_substitution(p, "GRCh37\tGRCh37\tHuman (GRCh37)\t$(dir $(bowtie2_indices))\n"))
 
-with open("/usr/local/share/bcbio-nextgen/galaxy/tool-data/bwa_indices.loc", "w") as f:
-    f.write(subst.do_substitution(p, "GRCh37\tGRCh37\tHuman (GRCh37)\t$(file $(bwa_indices))"))
+with open("/usr/local/share/bcbio-nextgen/galaxy/tool-data/bwa_index.loc", "w") as f:
+    f.write(subst.do_substitution(p, "GRCh37\tGRCh37\tHuman (GRCh37)\t$(file $(bwa_index))\n"))
 
 with open("/usr/local/share/bcbio-nextgen/galaxy/tool-data/gatk_sorted_picard_index.loc", "w") as f:
-    f.write(subst.do_substitution(p, "GRCh37\tGRCh37\tHuman (GRCh37)\t$(file $(gatk_sorted_picard_index))"))
+    f.write(subst.do_substitution(p, "GRCh37\tGRCh37\tHuman (GRCh37)\t$(file $(gatk_sorted_picard_index))\n"))
 
 with open("/usr/local/share/bcbio-nextgen/galaxy/tool-data/picard_index.loc", "w") as f:
-    f.write(subst.do_substitution(p, "GRCh37\tGRCh37\tHuman (GRCh37)\t$(file $(picard_index))"))
+    f.write(subst.do_substitution(p, "GRCh37\tGRCh37\tHuman (GRCh37)\t$(file $(picard_index))\n"))
 
 with open("/usr/local/share/bcbio-nextgen/galaxy/tool-data/sam_fa_indices.loc", "w") as f:
-    f.write(subst.do_substitution(p, "index\tGRCh37\t$(file $(sam_fa_indices))"))
+    f.write(subst.do_substitution(p, "index\tGRCh37\t$(file $(sam_fa_indices))\n"))
 
-with open("/tmp/crunch-job/gatk-variant.yaml", "w") as f:
+with open("/tmp/crunch-job/freebayes-variant.yaml", "w") as f:
     f.write('''
-# Template for whole genome Illumina variant calling with GATK pipeline
+# Template for whole genome Illumina variant calling with FreeBayes
+# This is a GATK-free pipeline without post-alignment BAM pre-processing
+# (recalibration and realignment)
 ---
 details:
   - analysis: variant2
@@ -81,20 +88,22 @@ details:
     #   batch: your-arbitrary-batch-name
     algorithm:
       aligner: bwa
-      mark_duplicates: picard
-      recalibrate: gatk
-      realign: gatk
-      variantcaller: gatk-haplotype
+      mark_duplicates: true
+      recalibrate: false
+      realign: false
+      variantcaller: freebayes
       platform: illumina
       quality_format: Standard
-      coverage_interval: genome
       # for targetted projects, set the region
       # variant_regions: /path/to/your.bed
 ''')
 
+os.unlink("/usr/local/share/bcbio-nextgen/gemini_data")
+os.symlink(arvados.get_job_param_mount("gemini_data"), "/usr/local/share/bcbio-nextgen/gemini_data")
+
 os.chdir(arvados.current_task().tmpdir)
 
-rcode = subprocess.call(["bcbio_nextgen.py", "--workflow", "template", "/tmp/crunch-job/gatk-variant.yaml", "project1",
+rcode = subprocess.call(["bcbio_nextgen.py", "--workflow", "template", "/tmp/crunch-job/freebayes-variant.yaml", "project1",
                          subst.do_substitution(p, "$(file $(R1))"),
                          subst.do_substitution(p, "$(file $(R2))")])
 
@@ -102,4 +111,35 @@ os.chdir("project1/work")
 
 os.symlink("/usr/local/share/bcbio-nextgen/galaxy/tool-data", "tool-data")
 
-rcode = subprocess.call(["bcbio_nextgen.py", "../config/project1.yaml"])
+rcode = subprocess.call(["bcbio_nextgen.py", "../config/project1.yaml", "-n", os.environ['CRUNCH_NODE_SLOTS']])
+
+print("run-command: completed with exit code %i (%s)" % (rcode, "success" if rcode == 0 else "failed"))
+
+if rcode == 0:
+    os.chdir("../final")
+
+    print("arvados-bcbio-nextgen: the follow output files will be saved to keep:")
+
+    subprocess.call(["find", ".", "-type", "f", "-printf", "arvados-bcbio-nextgen: %12.12s %h/%f\\n"])
+
+    print("arvados-bcbio-nextgen: start writing output to keep")
+
+    done = False
+    api = arvados.api('v1')
+    while not done:
+        try:
+            out = arvados.CollectionWriter()
+            out.write_directory_tree(".", max_manifest_depth=0)
+            outuuid = out.finish()
+            api.job_tasks().update(uuid=arvados.current_task()['uuid'],
+                                                 body={
+                                                     'output':outuuid,
+                                                     'success': (rcode == 0),
+                                                     'progress':1.0
+                                                 }).execute()
+            done = True
+        except Exception as e:
+            print("arvados-bcbio-nextgen: caught exception: {}".format(e))
+            time.sleep(5)
+
+sys.exit(rcode)