Merge pull request #76 from Knowledge-Graph-Hub/sars_cov2_gene_annota…

…tions

SARS-CoV-2 gene annotations

download.yaml

@@ -92,3 +92,18 @@
 -
   url: https://s3.pgkb.org/data/genes.zip
   local_name: pharmgkb_genes.zip
+
+#
+# SARS-CoV-2 gene annotations from Uniprot
+#
+
+-
+  # Gene Product Association Data for SARS-CoV-2 genes in GPAD format
+  # http://geneontology.org/docs/gene-product-association-data-gpad-format/
+  url: ftp://ftp.ebi.ac.uk/pub/contrib/goa/uniprot_sars-cov-2.gpa
+  local_name: uniprot_sars-cov-2.gpa
+-
+  # Gene Product information for SARS-CoV-2 genes in GPI format
+  # http://geneontology.org/docs/gene-product-information-gpi-format/
+  url: ftp://ftp.ebi.ac.uk/pub/contrib/goa/uniprot_sars-cov-2.gpi
+  local_name: uniprot_sars-cov-2.gpi

kg_covid_19/transform.py

@@ -5,6 +5,8 @@
 
 from kg_covid_19.transform_utils.drug_central.drug_central import DrugCentralTransform
 from kg_covid_19.transform_utils.hpo.hpo import HpoTransform
+from kg_covid_19.transform_utils.\
+    sars_cov_2_gene_annot.sars_cov_2_gene_annot import SARSCoV2GeneAnnot
 from kg_covid_19.transform_utils.scibite_cord import ScibiteCordTransform
 from kg_covid_19.transform_utils.pharmgkb import PharmGKB
 from kg_covid_19.transform_utils.string_ppi import StringTransform
@@ -19,7 +21,8 @@
     'TTDTransform': TTDTransform,
     'StringTransform': StringTransform,
     'ScibiteCordTransform': ScibiteCordTransform,
-    'PharmGKB': PharmGKB
+    'PharmGKB': PharmGKB,
+    'SARSCoV2GeneAnnot': SARSCoV2GeneAnnot
 }
 
 

kg_covid_19/transform_utils/example_transform/example_transform_script.py

@@ -13,18 +13,17 @@
 Input: any file in data/raw/ (that was downloaded by placing a URL in incoming.txt/yaml and running `run.py download`
 Output: transformed data in data/raw/[source name]:
 
-Either TSV, output these two files:
+Output these two files:
 - nodes.tsv
 - edges.tsv
-
-Or JSON, all in one file: nodes_edges.json
 """
 
 
 class YourTransform(Transform):
 
-    def __init__(self):
-        super().__init__(source_name="some_unique_name")
+    def __init__(self, input_dir: str = None, output_dir: str = None):
+        source_name = "some_unique_name"
+        super().__init__(source_name, input_dir, output_dir)
 
     def run(self):
         # replace with downloaded data of for this source

kg_covid_19/transform_utils/sars_cov_2_gene_annot/__init__.py

@@ -0,0 +1,5 @@
+from .sars_cov_2_gene_annot import SARSCoV2GeneAnnot
+
+__all__ = [
+    "SARSCoV2GeneAnnot"
+]

kg_covid_19/transform_utils/sars_cov_2_gene_annot/sars_cov_2_gene_annot.py

@@ -0,0 +1,210 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+import logging
+import os
+from typing import Generator, TextIO, List
+
+from kg_covid_19.utils.transform_utils import get_item_by_priority, ItemInDictNotFound
+
+from kg_covid_19.transform_utils.transform import Transform
+from kg_covid_19.utils import write_node_edge_item
+
+"""Parse the GPA and GPI files for SARS-CoV-2 gene annotations, including GO annotations
+and more. Make a node for each gene using GPI lines, and make an edge for 
+each gene -> annotation described in GPA 
+"""
+
+
+class SARSCoV2GeneAnnot(Transform):
+
+    def __init__(self, input_dir: str = None, output_dir: str = None):
+        source_name = "sars_cov_2_gene_annot"
+        super().__init__(source_name, input_dir, output_dir)
+
+        self.node_header = ['id', 'name', 'category', 'synonym', 'taxon']
+        self.edge_header = ['subject', 'edge_label', 'object', 'relation',
+                            'DB_References', 'ECO_code', 'With', 'Interacting_taxon_ID',
+                            'Date', 'Assigned_by', 'Annotation_Extension',
+                            'Annotation_Properties']
+
+        self.protein_node_type = "biolink:Protein"
+        self.ncbi_taxon_prefix = "NCBITaxon:"
+
+        # translate edge labels to RO term, for the 'relation' column in edge
+        self.edge_label_prefix = "biolink:"  # prepend to edge label
+        self.edge_label_to_RO_term: dict = {
+            'enables': 'RO:0002327',
+            'involved_in': 'RO:0002331',
+            'part_of': 'BFO:0000050'
+        }
+
+    def run(self):
+
+        # file housekeeping
+        os.makedirs(self.output_dir, exist_ok=True)
+
+        gpi_file = os.path.join(self.input_base_dir, "uniprot_sars-cov-2.gpi")
+        gpa_file = os.path.join(self.input_base_dir, "uniprot_sars-cov-2.gpa")
+
+        with open(self.output_node_file, 'w') as node, \
+                open(self.output_edge_file, 'w') as edge:
+
+            # write headers
+            node.write("\t".join(self.node_header) + "\n")
+            edge.write("\t".join(self.edge_header) + "\n")
+
+            with open(gpi_file, 'r') as gpi_fh:
+                for rec in _gpi12iterator(gpi_fh):
+                    node_data = self.gpi_to_gene_node_data(rec)
+                    write_node_edge_item(node, self.node_header, node_data)
+
+            with open(gpa_file, 'r') as gpa_fh:
+                for rec in _gpa11iterator(gpa_fh):
+                    edge_data = self.gpa_to_edge_data(rec)
+                    write_node_edge_item(edge, self.edge_header, edge_data)
+
+    def gpa_to_edge_data(self, rec: dict) -> list:
+        """given a parsed gpa entry, return an edge with the annotations
+
+        :param rec: record from gpa iterator
+        :return:
+        """
+        subj: str = self._rec_to_id(rec)
+        edge_label: str = get_item_by_priority(rec, ['Qualifier'])[0]
+        obj: str = get_item_by_priority(rec, ['GO_ID'])
+        try:
+            relation: str = self.edge_label_to_RO_term[edge_label]
+        except KeyError:
+            relation = ''
+
+        edge_data = [subj, self.edge_label_prefix + edge_label, obj, relation]
+        # all the others
+        for key in ['DB:Reference', 'ECO_Evidence_code', 'With', 'Interacting_taxon_ID',
+                    'Date', 'Assigned_by', 'Annotation_Extension',
+                    'Annotation_Properties']:
+            try:
+                item = get_item_by_priority(rec, [key])
+                if type(item) is list:
+                    item = item[0]
+            except (ItemInDictNotFound, IndexError):
+                item = ''
+            edge_data.append(item)
+        return edge_data
+
+    def _rec_to_id(self, rec: dict) -> str:
+        try:
+            this_id: str = get_item_by_priority(rec, ['DB']) + ":" + \
+                           get_item_by_priority(rec, ['DB_Object_ID'])
+        except ItemInDictNotFound:
+            logging.error("Can't make ID for record: %s", "\t".join(rec))
+            this_id = ''
+        return this_id
+
+    def gpi_to_gene_node_data(self, rec: dict) -> list:
+        """given a parsed gpi entry, return a node that can be passed to
+        write_node_edge_item()
+
+        :param rec: record from gpi iterator
+        :return: list of node items, one for each thing in self.node_header
+        """
+        # ['id', 'name', 'category', 'synonym', 'taxon']
+        id: str = self._rec_to_id(rec)
+
+        try:
+            name_list = get_item_by_priority(rec, ['DB_Object_Name'])
+            if name_list is not None and len(name_list) > 0:
+                name = name_list[0]
+            else:
+                name = ''
+        except (IndexError, ItemInDictNotFound):
+            name = ''
+
+        category = self.protein_node_type
+        try:
+            synonym = get_item_by_priority(rec, ['DB_Object_Synonym'])[0]
+        except (IndexError, ItemInDictNotFound):
+            synonym = ''
+        taxon = get_item_by_priority(rec, ['Taxon'])
+        return [id, name, category, synonym, taxon]
+
+
+def _gpi12iterator(handle: TextIO) -> Generator:
+    # Partly cribbed from Biopython GPI 1.1 parser
+    # There is no GPI 1.2 parser yet, so I made this
+    """Read GPI 1.2 format files (PRIVATE).
+
+    This iterator is used to read a gp_information.goa_uniprot
+    file which is in the GPI 1.2 format.
+    """
+    logging.getLogger().setLevel(logging.WARNING)
+    # GPI version 1.2
+    GPI11FIELDS = [
+        "DB",
+        "DB_Object_ID",
+        "DB_Object_Symbol",
+        "DB_Object_Name",
+        "DB_Object_Synonym",
+        "DB_Object_Type",
+        "Taxon",
+        "Parent_Object_ID",
+        "DB_Xref",
+        "Properties",
+    ]
+    for inline in handle:
+        if inline[0] == "!":
+            continue
+        inrec: List[str] = inline.rstrip("\n").split("\t")
+        if len(inrec) == 1:
+            continue
+        # DB_Object_Name
+        inrec[2] = inrec[2].split("|")  # type: ignore
+        # DB_Object_Synonym(s)
+        inrec[3] = inrec[3].split("|")  # type: ignore
+        try:
+            # DB_Xref(s)
+            inrec[7] = inrec[7].split("|")  # type: ignore
+        except IndexError:
+            logging.debug("No index for DB_Xref for this record")
+        try:
+            # Properties
+            inrec[8] = inrec[8].split("|")  # type: ignore
+        except IndexError:
+            logging.debug("No index for Properties for this record")
+
+        yield dict(zip(GPI11FIELDS, inrec))
+
+
+# from biopython, to avoid dependency
+def _gpa11iterator(handle):
+    """Read GPA 1.1 format files (PRIVATE).
+
+    This iterator is used to read a gp_association.goa_uniprot
+    file which is in the GPA 1.1 format. Do not call directly. Rather
+    use the gpa_iterator function
+    """
+    # GPA version 1.1
+    GPA11FIELDS = [
+        "DB",
+        "DB_Object_ID",
+        "Qualifier",
+        "GO_ID",
+        "DB:Reference",
+        "ECO_Evidence_code",
+        "With",
+        "Interacting_taxon_ID",
+        "Date",
+        "Assigned_by",
+        "Annotation Extension",
+        "Annotation_Properties",
+    ]
+    for inline in handle:
+        if inline[0] == "!":
+            continue
+        inrec = inline.rstrip("\n").split("\t")
+        if len(inrec) == 1:
+            continue
+        inrec[2] = inrec[2].split("|")  # Qualifier
+        inrec[4] = inrec[4].split("|")  # DB:Reference(s)
+        inrec[6] = inrec[6].split("|")  # With
+        inrec[10] = inrec[10].split("|")  # Annotation extension
+        yield dict(zip(GPA11FIELDS, inrec))

kg_covid_19/utils/transform_utils.py

@@ -74,7 +74,7 @@ def write_node_edge_item(fh: Any, header: List, data: List, sep: str = '\t'):
         raise Exception('Header and data are not the same length.')
     try:
         fh.write(sep.join(data) + "\n")
-    except:
+    except IOError:
         logging.warning("Can't write data for {}".format(data))
 
 

merge.yaml

@@ -36,6 +36,12 @@ target:
          - data/transformed/SciBite-CORD-19/nodes.tsv
          - data/transformed/SciBite-CORD-19/edges.tsv
 
+   sars-cov-2-gene-annot:
+      type: tsv
+      filename:
+         - data/transformed/sars_cov_2_gene_annot/nodes.tsv
+         - data/transformed/sars_cov_2_gene_annot/edges.tsv
+
 destination:
    type: tsv
    filename: merged-kg

tests/resources/uniprot_sars-cov-2_SNIPPET.gpa

@@ -0,0 +1,22 @@
+!gpa-version: 1.1
+!Columns:
+!
+!   name                  required? cardinality   GAF column #
+!   DB                    required  1             1
+!   DB_Object_ID          required  1             2 / 17
+!   Qualifier             required  1 or greater  4
+!   GO ID                 required  1             5
+!   DB:Reference(s)       required  1 or greater  6
+!   ECO evidence code     required  1             7 + 6 (GO evidence code + reference)
+!   With                  optional  0 or greater  8
+!   Interacting taxon ID  optional  0 or 1        13
+!   Date                  required  1             14
+!   Assigned_by           required  1             15
+!   Annotation Extension  optional  0 or greater  16
+!   Annotation Properties optional  0 or 1        n/a
+!
+!Generated: 2020-03-26 11:33
+!GO-version: http://purl.obolibrary.org/obo/go/releases/2020-03-24/extensions/go-plus.owl
+!
+UniProtKB	P0DTC1	enables	GO:0003723	GO_REF:0000043	ECO:0000322	UniProtKB-KW:KW-0694		20200321	UniProt		go_evidence=IEA
+UniProtKB	P0DTC1	involved_in	GO:0090305	GO_REF:0000108	ECO:0000366	GO:0004519		20200322	GOC		go_evidence=IEA

tests/resources/uniprot_sars-cov-2_SNIPPET.gpi

@@ -0,0 +1,23 @@
+!gpi-version: 1.2
+!
+!This file contains additional information for proteins in the UniProt KnowledgeBase (UniProtKB).
+!Protein accessions are represented in this file even if there is no associated GO annotation.
+!
+!Columns:
+!
+!   name                   required? cardinality   GAF column #  Example content
+!   DB                     required  1             1             UniProtKB
+!   DB_Object_ID           required  1             2/17          Q4VCS5-1
+!   DB_Object_Symbol       required  1             3             AMOT
+!   DB_Object_Name         optional  0 or greater  10            Angiomotin
+!   DB_Object_Synonym(s)   optional  0 or greater  11            AMOT|KIAA1071
+!   DB_Object_Type         required  1             12            protein
+!   Taxon                  required  1             13            taxon:9606
+!   Parent_Object_ID       optional  0 or 1        -             UniProtKB:Q4VCS5
+!   DB_Xref(s)             optional  0 or greater  -             WB:WBGene00000035
+!   Properties             optional  0 or greater  -             db_subset=Swiss-Prot|target_set=KRUK,BHFL
+!
+!Generated: 2020-03-24 15:52
+!
+UniProtKB	P0DTD2	P0DTD2	Protein 9b		protein	taxon:2697049
+UniProtKB	A0A663DJA2	ORF10	ORF10 protein	ORF10	protein	taxon:2697049