Add complexportal

src/pyobo/sources/__init__.py

@@ -18,6 +18,7 @@
 
 CONVERTED = {
     'cgnc': 'cgnc',
+    'complexportal': 'complexportal',
     'ncbigene': 'entrez',
     'ec-code': 'expasy',
     'hgnc': 'hgnc',

src/pyobo/sources/complexportal.py

@@ -0,0 +1,175 @@
+# -*- coding: utf-8 -*-
+
+"""Converter for ComplexPortal."""
+
+import logging
+from typing import Iterable, List, Tuple
+
+import pandas as pd
+from tqdm import tqdm
+
+from pyobo import get_id_name_mapping
+from pyobo.path_utils import ensure_df
+from pyobo.struct import Obo, Reference, Synonym, Term, from_species, has_part
+
+logger = logging.getLogger(__name__)
+
+PREFIX = 'complexportal'
+VERSION = '2020-03-11'
+URL_BASE = f'ftp://ftp.ebi.ac.uk/pub/databases/intact/complex/{VERSION}/complextab'
+SPECIES = [
+    'arabidopsis_thaliana',
+    'bos_taurus',
+    'caenorhabditis_elegans',
+    'canis_familiaris',
+    'danio_rerio',
+    'drosophila_melanogaster',
+    'escherichia_coli',
+    'gallus_gallus',
+    'homo_sapiens',
+    'lymnaea_stagnalis',
+    'mus_musculus',
+    'oryctolagus_cuniculus',
+    'ovis_aries',
+    'pseudomonas_aeruginosa',
+    'rattus_norvegicus',
+    'saccharomyces_cerevisiae',
+    'schizosaccharomyces_pombe',
+    'sus_scrofa',
+    'torpedo_californica',
+    'torpedo_marmorata',
+    'xenopus_laevis',
+]
+URLS = [f'{URL_BASE}/{species}.tsv' for species in SPECIES]
+
+COLUMNS = [
+    'complexportal_id',
+    'name',
+    'aliases',
+    'taxonomy_id',
+    'members',
+    'confidence',
+    'experimental_evidence',
+    'goa',
+    'xrefs',
+    'definition',
+    'Complex properties',
+    'Complex assembly',
+    'Ligand',
+    'Disease',
+    'Agonist',
+    'Antagonist',
+    'Comment',
+    'Source',
+]
+DTYPE = {
+    'taxonomy_id': str,
+}
+
+
+def _parse_members(s) -> List[Tuple[Reference, str]]:
+    if pd.isna(s):
+        return []
+
+    rv = []
+    for member in s.split('|'):
+        entity_id, count = member.split('(')
+        count = count.rstrip(')')
+        if ':' in entity_id:
+            prefix, identifier = entity_id.split(':', 1)
+        else:
+            prefix, identifier = 'uniprot', entity_id
+        rv.append((Reference(prefix=prefix, identifier=identifier), count))
+    return rv
+
+
+def _parse_xrefs(s) -> List[Tuple[Reference, str]]:
+    if pd.isna(s):
+        return []
+
+    rv = []
+    for xref in s.split('|'):
+        entity_id, note = xref.split('(')
+        note = note.rstrip(')')
+        prefix, identifier = entity_id.split(':', 1)
+        rv.append((Reference(prefix=prefix, identifier=identifier), note))
+    return rv
+
+
+def get_obo() -> Obo:
+    """Get the ComplexPortal OBO."""
+    terms = list(get_terms())
+    return Obo(
+        ontology=PREFIX,
+        name='Complex Portal',
+        data_version=VERSION,
+        terms=terms,
+    )
+
+
+def get_df() -> pd.DataFrame:
+    """Get a combine ComplexPortal dataframe."""
+    dfs = [
+        ensure_df(PREFIX, url, version=VERSION, na_values={'-'}, names=COLUMNS, header=0, dtype=DTYPE)
+        for url in URLS
+    ]
+    return pd.concat(dfs)
+
+
+def get_terms() -> Iterable[Term]:
+    """Get ComplexPortal terms."""
+    df = get_df()
+
+    df['aliases'] = df['aliases'].map(lambda s: s.split('|') if pd.notna(s) else [])
+    df['members'] = df['members'].map(_parse_members)
+    df['xrefs'] = df['xrefs'].map(_parse_xrefs)
+
+    taxnomy_id_to_name = get_id_name_mapping('ncbitaxon')
+    df['taxonomy_name'] = df['taxonomy_id'].map(taxnomy_id_to_name.get)
+
+    slim_df = df[[
+        'complexportal_id',
+        'name',
+        'definition',
+        'aliases',
+        'xrefs',
+        'taxonomy_id',
+        'taxonomy_name',
+        'members',
+    ]]
+    it = tqdm(slim_df.values, total=len(slim_df.index))
+    unhandled_xref_type = set()
+    for complexportal_id, name, definition, aliases, xrefs, taxonomy_id, taxonomy_name, members in it:
+        synonyms = [
+            Synonym(name=alias)
+            for alias in aliases
+        ]
+        _xrefs = []
+        provenance = []
+        for reference, note in xrefs:
+            if note == 'identity':
+                _xrefs.append(reference)
+            elif note == 'see-also' and reference.prefix == 'pubmed':
+                provenance.append(reference)
+            elif (note, reference.prefix) not in unhandled_xref_type:
+                it.write(f'unhandled xref type: {note} / {reference.prefix}')
+                unhandled_xref_type.add((note, reference.prefix))
+
+        term = Term(
+            reference=Reference(prefix=PREFIX, identifier=complexportal_id, name=name),
+            name=name,
+            definition=definition.strip(),
+            synonyms=synonyms,
+            xrefs=_xrefs,
+            provenance=provenance,
+        )
+        term.append_relationship(from_species, Reference(prefix='taxonomy', identifier=taxonomy_id, name=taxonomy_name))
+
+        for reference, count in members:
+            term.append_relationship(has_part, reference)
+
+        yield term
+
+
+if __name__ == '__main__':
+    get_obo().write_default()