From 0b0fb1c8a68df989bb2e1f593d717ac62e31d952 Mon Sep 17 00:00:00 2001
From: Peter Amstutz
Date: Mon, 19 Oct 2020 21:04:19 -0400
Subject: Extract subset of the all-sequences fasta by running a sparql query.

Arvados-DCO-1.1-Signed-off-by: Peter Amstutz <peter.amstutz@curii.com>
---
 workflows/pangenome-generate/from_sparql.cwl | 23 +++++++++++++++++++++++
 workflows/pangenome-generate/from_sparql.py  |  8 ++++++++
 2 files changed, 31 insertions(+)
 create mode 100644 workflows/pangenome-generate/from_sparql.cwl
 create mode 100644 workflows/pangenome-generate/from_sparql.py

(limited to 'workflows/pangenome-generate')

diff --git a/workflows/pangenome-generate/from_sparql.cwl b/workflows/pangenome-generate/from_sparql.cwl
new file mode 100644
index 0000000..5bc0792
--- /dev/null
+++ b/workflows/pangenome-generate/from_sparql.cwl
@@ -0,0 +1,23 @@
+cwlVersion: v1.1
+class: CommandLineTool
+requirements:
+  DockerRequirement:
+    dockerFile: |
+      FROM debian:10
+      RUN apt-get update && apt-get -yq --no-install-recommends install samtools python3-rdflib
+    dockerImageId: rdflib-and-samtools
+inputs:
+  script:
+    type: File
+    default:
+      class: File
+      location: from_sparql.py
+  metadata: File
+  fasta:
+    type: File
+    secondaryFiles: [.fai]
+  query: string
+stdout: selected.fasta
+outputs:
+  selected: stdout
+arguments: [python3, $(inputs.script), $(inputs.metadata), $(inputs.fasta), $(inputs.query)]
diff --git a/workflows/pangenome-generate/from_sparql.py b/workflows/pangenome-generate/from_sparql.py
new file mode 100644
index 0000000..4610cad
--- /dev/null
+++ b/workflows/pangenome-generate/from_sparql.py
@@ -0,0 +1,8 @@
+from rdflib import Graph
+import sys
+import subprocess
+g = Graph()
+g.parse(sys.argv[1], format="nt")
+res = g.query(sys.argv[3])
+for r in res:
+    subprocess.run(["samtools", "faidx", sys.argv[2], r[0]])
-- 
cgit v1.2.3


From 0d1831449d7541e0ec36fd75793915d144a02b5d Mon Sep 17 00:00:00 2001
From: Peter Amstutz
Date: Mon, 9 Nov 2020 16:06:01 -0500
Subject: Rename schema param to metadataSchema

Arvados-DCO-1.1-Signed-off-by: Peter Amstutz <peter.amstutz@curii.com>
---
 workflows/pangenome-generate/arv-main.cwl     | 2 +-
 workflows/pangenome-generate/collect-seqs.cwl | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'workflows/pangenome-generate')

diff --git a/workflows/pangenome-generate/arv-main.cwl b/workflows/pangenome-generate/arv-main.cwl
index dae47e6..1d71ee3 100644
--- a/workflows/pangenome-generate/arv-main.cwl
+++ b/workflows/pangenome-generate/arv-main.cwl
@@ -36,7 +36,7 @@ steps:
     run: collect-seqs.cwl
     in:
       src_project: src_project
-      schema: metadataSchema
+      metadataSchema: metadataSchema
       exclude: exclude
     out: [relabeledSeqs, mergedMetadata]
   pangenome-generate:
diff --git a/workflows/pangenome-generate/collect-seqs.cwl b/workflows/pangenome-generate/collect-seqs.cwl
index 3511df1..635108f 100644
--- a/workflows/pangenome-generate/collect-seqs.cwl
+++ b/workflows/pangenome-generate/collect-seqs.cwl
@@ -25,7 +25,7 @@ inputs:
   src_project:
     type: string
     inputBinding: {position: 2}
-  schema:
+  metadataSchema:
     type: File
     inputBinding: {position: 3}
   exclude:
-- 
cgit v1.2.3


From 1082b907d816f5da52aba6233073737632d0242f Mon Sep 17 00:00:00 2001
From: Peter Amstutz
Date: Mon, 9 Nov 2020 17:20:27 -0500
Subject: Make resource link work for both portable data hashes and sample id

Arvados-DCO-1.1-Signed-off-by: Peter Amstutz <peter.amstutz@curii.com>
---
 bh20simplewebuploader/main.py                | 14 ++++++++++++++
 workflows/pangenome-generate/collect-seqs.py |  2 +-
 2 files changed, 15 insertions(+), 1 deletion(-)

(limited to 'workflows/pangenome-generate')

diff --git a/bh20simplewebuploader/main.py b/bh20simplewebuploader/main.py
index 73503b4..405544c 100644
--- a/bh20simplewebuploader/main.py
+++ b/bh20simplewebuploader/main.py
@@ -675,13 +675,27 @@ sparqlURL='http://sparql.genenetwork.org/sparql/'
 @app.route('/resource/<id>')
 def resource(id):
     """Get a COVID19 resource using identifier"""
+
     query=f"""
 PREFIX pubseq: <http://biohackathon.org/bh20-seq-schema#MainSchema/>
 PREFIX sio: <http://semanticscience.org/resource/>
 select distinct ?sample ?geoname ?date ?source ?geo ?sampletype ?institute ?sequenceuri
+where {{
 {{
    ?sample sio:SIO_000115 "{id}" .
    ?sequenceuri pubseq:sample ?sample .
+}}
+union
+{{
+   <http://collections.lugli.arvadosapi.com/c={id}/sequence.fasta> pubseq:sample ?sample .
+   ?sequenceuri pubseq:sample ?sample .
+}}
+union
+{{
+   <http://covid19.genenetwork.org/resource/{id}> pubseq:sample ?sample .
+   ?sequenceuri pubseq:sample ?sample .
+}}
+
    ?sample <http://purl.obolibrary.org/obo/GAZ_00000448> ?geo .
    ?geo rdfs:label ?geoname .
    ?sample <http://ncicb.nci.nih.gov/xml/owl/EVS/Thesaurus.owl#C25164> ?date .
diff --git a/workflows/pangenome-generate/collect-seqs.py b/workflows/pangenome-generate/collect-seqs.py
index af4a0dc..1a0807c 100644
--- a/workflows/pangenome-generate/collect-seqs.py
+++ b/workflows/pangenome-generate/collect-seqs.py
@@ -38,7 +38,7 @@ for item in validated:
     pdh = item["portable_data_hash"]
     with arvados.collection.CollectionReader(pdh, api_client=api, keep_client=keepclient) as col:
         with col.open("sequence.fasta", "rt") as fa:
-            subject = "http://collections.lugli.arvadosapi.com/c=%s/sequence.fasta" % pdh
+            subject = "http://covid19.genenetwork.org/resource/%s" % pdh
             label = fa.readline().strip()
             merged_metadata.write("<%s> <http://biohackathon.org/bh20-seq-schema/original_fasta_label> \"%s\" .\n" % (subject, label[1:].replace('"', '\\"')))
             skip = (subject in blacklist or label[1:] in blacklist)
-- 
cgit v1.2.3