Merge branch 'features/pubmed' into 'develop'

Features/pubmed

See merge request SOaAS/dbxref!16

dbxref/providers.yaml

@@ -74,10 +74,16 @@
     type: 'internal'
     location: 'dbxref.retrieve.rfam'
 - name: Pubmed
-  prefixes: ["Pubmed"]
+  prefixes: ["Pubmed", "PM"]
   resources:
-      html: ["http://www.ncbi.nlm.nih.gov/pubmed/%i"]
+      html: ['http://www.ncbi.nlm.nih.gov/pubmed/%i']
+      text: ['https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=pubmed&id=%i']
+      xml: ['https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=pubmed&id=%i&retmode=xml']
+      json: ['https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esummary.fcgi?db=pubmed&id=%i&retmode=json']
   check_existence: "http://www.ncbi.nlm.nih.gov/pubmed/%i"
+  retriever:
+    type: 'internal'
+    location: 'dbxref.retrieve.pubmed'
 - name: Protein Families
   prefixes: ["PFAM"]
   resources:

dbxref/retrieve/pubmed.py

+#!/usr/bin/env python3
+import dbxref.resolver
+import requests
+import logging
+import json
+import argparse
+
+logger = logging.getLogger(__name__)
+
+
+def main():
+    """main()method for script usage"""
+    parser = argparse.ArgumentParser(description="Retrieve Pubmed json documents and parse into dbxref json format")
+    parser.add_argument("--basics", "-b", action="store_true", help="Include basic information such as title, language,"
+                                                                    " dbxref-id, and day of publishment on pubmed.")
+    parser.add_argument("dbxrefs", nargs=argparse.REMAINDER)
+    args = parser.parse_args()
+
+    if None not in (args.basics):
+        args.basics = True
+
+    dbxrefs = dbxref.resolver.convert_to_dbxrefs(args.dbxrefs)
+    documents = retrieve(dbxrefs, basics=args.basics)
+    print(json.dumps(documents, sort_keys=True, indent=4))
+
+def _get(result, field, mandatory=False, default="", transform=lambda x: x):
+  """Retrieve a given field if available, return default or exception otherwise. Result may be manipulated by transformation function"""
+  if field in result:
+    return transform(result[field])
+  else:
+    if mandatory:
+      raise KeyError("Field '"+field+"' not found in dictionary")
+    else:
+      return default
+
+def find_id(list, type):
+  """Find id of given type in pubmed islist"""
+  matches = [x for x in list if x['idtype'] == type]
+  if matches:
+    return matches[0]["value"]
+  else:
+    raise KeyError("Id of type '" + type + "' not found in idlist.")
+
+def join_authors(list):
+  """Joins pubmed entry authors to a single string"""
+  return ", ".join([x["name"] for x in list])
+
+def retrieve(dbxrefs, basics=True):
+    """Retrieve Pubmed json documents and parse into dbxref json format"""
+    resolved = dbxref.resolver.resolve(dbxrefs, check_existence=False)
+    documents = []
+    for entry in resolved:
+        # Construct URL for retrival
+        json_url = entry["locations"]["json"][0]
+        logger.debug("URL: %s", json_url)
+        r = requests.get(json_url)
+        logger.debug("Content: %s", r.text)
+        pm = json.loads(r.text)
+        output = {"id": entry["dbxref"]}
+        entry_id = dbxrefs[0]["id"]
+        # Parse basic information
+        result = pm["result"][entry_id]
+        if basics:
+            output["publication-date"] = _get(result, "epubdate")
+            output["dbxref"] = "Pubmed:" + _get(result, "uid")
+            output["title"] = _get(result, "title")
+            output["language"] = _get(result, "lang", transform=lambda x: ", ".join(x))
+            output["authors"] = _get(result, "authors", transform=lambda x: join_authors(x))
+            output["source"] = _get(result, "source")
+            output["volume"] = _get(result, "volume")
+            output["issue"] = _get(result, "issue")
+            output["doi"] = _get(result, "articleids", transform=lambda x: find_id(x, "doi"))
+        documents.append(output)
+    return documents
+
+
+if __name__ == '__main__':
+    main()

docs/pubmed.rst

+.. highlight:: yaml
+
+Pubmed
+=================
+
+Retrieve Pubmed json documents for dbxrefs and convert them into dbxref json format.
+
+Options
+-------
+
+  * ``--basics`` - Include basic information such as title, language, dbxref-id, and day of publishment on pubmed.
+  * ``--references`` - Include reference information such as journal name, DOI, authors and day of publishment.
+  * ``--article_ids`` - Include article-IDs.
+
+Input
+-----
+
+example: ``PM:19393038``
+
+
+Output
+------
+
+output scheme::
+
+[
+    {
+        "article_IDs": [
+            "pubmed: 19393038",
+            "pii: gb-2009-10-4-r42",
+            "doi: 10.1186/gb-2009-10-4-r42",
+            "pmc: PMC2688933",
+            "rid: 19393038",
+            "eid: 19393038",
+            "pmcid: pmc-id: PMC2688933;"
+        ],
+        "dbxref_id": "PM:19393038",
+        "epublic-date": "24.Apr.2009",
+        "language": [
+            "eng"
+        ],
+        "references": {
+            "DOI": "10.1186/gb-2009-10-4-r42",
+            "authors": [
+                "Zimin, AV",
+                "Delcher, AL",
+                "Florea, L",
+                "Kelley, DR",
+                "Schatz, MC",
+                "Puiu, D",
+                "Hanrahan, F",
+                "Pertea, G",
+                "Van, Tassell CP",
+                "Sonstegard, TS",
+                "Mar\u00e7ais, G",
+                "Roberts, M",
+                "Subramanian, P",
+                "Yorke, JA",
+                "Salzberg, SL"
+            ],
+            "journal": "Genome biology",
+            "pubdate": "2009/01/01 00:00"
+        },
+        "title": "A whole-genome assembly of the domestic cow, Bos taurus."
+    }
+]

tests/test_pubmed.py

+import unittest
+from dbxref.retrieve import pubmed
+
+
+class TestPubmed(unittest.TestCase):
+
+    def test_output(self):
+        documents = pubmed.retrieve([{"db": "Pubmed", "id": "19393038"}], basics=True)
+        self.assertTrue(documents)
+
+
+if __name__ == "__main__":
+    unittest.main()