Merge branch 'add_domains_to_proteins' into wizard_form

ippisite/ippidb/models.py

@@ -188,6 +188,7 @@ class Protein(AutoFillableModel):
     entry_name = models.CharField('Entry name', max_length=30)
     organism = models.ForeignKey('Taxonomy', models.CASCADE)
     molecular_functions = models.ManyToManyField(MolecularFunction)
+    domains = models.ManyToManyField('Domain')
 
     @transaction.atomic
     def autofill(self):
@@ -208,6 +209,7 @@ class Protein(AutoFillableModel):
             taxonomy.save(autofill=True)
         self.organism = taxonomy
         super(Protein, self).save()
+
         for go_id in info['molecular_functions']:
             try:
                 mol_function = MolecularFunction.objects.get(go_id=go_id)
@@ -217,6 +219,15 @@ class Protein(AutoFillableModel):
                 mol_function.save(autofill=True)
             self.molecular_functions.add(mol_function)
 
+        for domain_id in info['domains']:
+            try:
+                domain = Domain.objects.get(pfam_acc=domain_id)
+            except MolecularFunction.DoesNotExist:
+                domain = Domain()
+                domain.pfam_acc = domain_id
+                domain.save(autofill=True)
+            self.domains.add(domain)
+
     def is_autofill_done(self):
         return len(self.gene_name) > 0
 
@@ -231,7 +242,7 @@ class Domain(AutoFillableModel):
     pfam_acc = models.CharField('Pfam Accession', max_length=10, unique=True)
     pfam_id = models.CharField('Pfam Family Identifier', max_length=20)
     pfam_description = models.CharField('Pfam Description', max_length=100)
-    domain_family = models.CharField('Domain family', max_length=25)
+    domain_family = models.CharField('Domain family', max_length=25, blank=True, default="")
 
     # TODO: what is this field? check database
     # contents

ippisite/ippidb/tests.py

@@ -4,15 +4,18 @@ iPPI-DB unit tests
 
 import re
 
+from django.core.management import call_command
 from django.test import TestCase
 from django.urls import reverse
-from django.core.management import call_command
 from openbabel import vectorUnsignedInt, OBFingerprint
 
+from ippidb import ws
+from ippidb.ws import get_uniprot_info
 from .models import Compound, CompoundTanimoto, create_tanimoto
-from .models import DrugBankCompound, DrugbankCompoundTanimoto
+from .models import DrugBankCompound
 from .utils import FingerPrinter, mol2smi, smi2mol, smi2inchi, smi2inchikey
 
+
 class MolSmiTestCase(TestCase):
     """
     Test MOL to SMILES and SMILES to MOL format conversion functions
@@ -322,3 +325,71 @@ class QuestionCompoundViews(TestCase):
         url = reverse('compound_list')
         response = self.client.get(url)
         self.assertEqual(response.status_code, 200)
+
+
+class TestGetUniprotInfo(TestCase):
+    """
+    Test retrieving information for a uniprot entry
+    """
+
+    def test_get_uniprot_info(self):
+        resp = get_uniprot_info('Q15286')
+        self.assertEqual(resp['recommended_name'],'Ras-related protein Rab-35')
+        self.assertEqual(resp['organism'],9606)
+        self.assertEqual(resp['gene_id'],11021)
+        exp_gene_names = [{'name': 'RAB35', 'type': 'primary'}, {'name': 'RAB1C', 'type': 'synonym'}, {'name': 'RAY', 'type': 'synonym'}]
+        self.assertEqual(sorted(resp['gene_names'], key=lambda k: k['name']), sorted(exp_gene_names, key=lambda k: k['name']))
+        self.assertEqual(resp['entry_name'],'RAB35_HUMAN')
+        self.assertEqual(resp['short_name'],'RAB35')
+        exp_molecular_functions = ['GO_0003924', 'GO_0005525', 'GO_0005546', 'GO_0019003']
+        self.assertEqual(sorted(resp['molecular_functions']),sorted(exp_molecular_functions))
+        exp_cellular_localisations = ['GO_0000139', 'GO_0005829', 'GO_0005886', 'GO_0005905', 'GO_0010008',
+                                      'GO_0030665', 'GO_0031253', 'GO_0042470', 'GO_0045171', 'GO_0045334',
+                                      'GO_0055038', 'GO_0070062', 'GO_0098993']
+        self.assertEqual(sorted(resp['cellular_localisations']),sorted(exp_cellular_localisations))
+        exp_biological_processes = ['GO_0000281', 'GO_0006886', 'GO_0008104', 'GO_0016197',
+                                    'GO_0019882', 'GO_0031175', 'GO_0032456', 'GO_0032482',
+                                    'GO_0036010', 'GO_0048227', 'GO_1990090']
+        self.assertEqual(sorted(resp['biological_processes']),sorted(exp_biological_processes))
+        exp_accessions = ['Q15286', 'B2R6E0', 'B4E390']
+        self.assertEqual(sorted(resp['accessions']),sorted(exp_accessions))
+        exp_citations = [
+            {'doi': '10.1006/bbrc.1994.2889', 'pmid': '7811277'},
+            {'doi': '10.1038/ng1285', 'pmid': '14702039'},
+            {'doi': '10.1038/nature04569', 'pmid': '16541075'},
+            {'doi': '10.1101/gr.2596504', 'pmid': '15489334'},
+            {'doi': '10.1016/j.cub.2006.07.020', 'pmid': '16950109'},
+            {'doi': '10.1021/pr060363j', 'pmid': '17081065'},
+            {'doi': '10.1074/jbc.M109.050930', 'pmid': '20154091'},
+            {'doi': '10.1186/1752-0509-5-17', 'pmid': '21269460'},
+            {'doi': '10.1038/nature10335', 'pmid': '21822290'},
+            {'doi': '10.1038/emboj.2012.16', 'pmid': '22307087'},
+            {'doi': '10.1111/j.1600-0854.2011.01294.x', 'pmid': '21951725'},
+            {'doi': '10.1021/pr300630k', 'pmid': '23186163'},
+            {'doi': '10.1002/pmic.201400617', 'pmid': '25944712'},
+            {'doi': '10.1073/pnas.1110415108', 'pmid': '22065758'}
+            ]
+        self.assertEqual(sorted(exp_citations, key=lambda k: k['pmid']), sorted(resp['citations'], key=lambda k: k['pmid']))
+        exp_alternative_names = [{'full': 'GTP-binding protein RAY'}, {'full': 'Ras-related protein Rab-1C'}]
+        self.assertEqual(sorted(exp_alternative_names, key=lambda k: k['full']), sorted(resp['alternative_names'], key=lambda k: k['full']))
+
+    def test_get_uniprot_info_domains(self):
+        resp = get_uniprot_info('O00255')
+        exp_domains = ['PF05053']
+        self.assertEqual(sorted(resp['domains']), sorted(exp_domains))
+
+
+class TestGetPDBUniProtMapping(TestCase):
+    """
+    Test retrieving protein for a PDB entry
+    """
+
+    def test_find_info(self):
+        target = sorted(['Q03164', 'O00255'])
+        resp = ws.get_pdb_uniprot_mapping('3u85')
+        resp = sorted(resp)
+        self.assertEqual(resp, target)
+        self.assertEqual(len(resp), len(set(resp)))
+
+    def test_entry_not_found(self):
+        self.assertRaises(ws.EntryNotFoundError, ws.get_pdb_uniprot_mapping, 'Xu85')

ippisite/ippidb/ws.py

 """
 iPPI-DB web-service client utility functions
 """
+
 import json
 import xml.etree.ElementTree as ET
 
@@ -10,7 +11,14 @@ from bioservices.uniprot import UniProt
 from bs4 import BeautifulSoup
 
 
-class BibliographicalEntryNotFound(Exception):
+class EntryNotFoundError(Exception):
+
+    def __init__(self, entry_id):
+        self.entry_id = entry_id
+        super().__init__('%s not found' % entry_id)
+
+
+class BibliographicalEntryNotFound(EntryNotFoundError):
     pass
 
 
@@ -182,30 +190,83 @@ def get_uniprot_info(uniprot_id):
     """
     uniprot_client = UniProt()
     ns = {'u': 'http://uniprot.org/uniprot'}
-    resp = uniprot_client.retrieve(uniprot_id)
-    f = open('/tmp/'+uniprot_id+'.xml', 'w')
-    f.write(str(resp))
-    f.close()
-    recommended_name = resp.root.findall(
-        'u:entry/u:protein/u:recommendedName/u:fullName', ns)[0].text
+    try:
+        resp = uniprot_client.retrieve(uniprot_id)
+    except TypeError:
+        raise EntryNotFoundError(uniprot_id)
+    if resp.root=='':
+        raise EntryNotFoundError(uniprot_id)
+    try:
+        recommended_name = resp.root.findall(
+            'u:entry/u:protein/u:recommendedName/u:fullName', ns)[0].text
+    except:
+        recommended_name = None
+    try:
+        recommended_short_name = resp.root.findall(
+            'u:entry/u:protein/u:recommendedName/u:shortName', ns)[0].text
+    except:
+        recommended_short_name = None
     organism = resp.root.findall(
         'u:entry/u:organism/u:dbReference[@type="NCBI Taxonomy"]', ns)[0].attrib['id']
-    gene = resp.root.findall(
-        'u:entry/u:gene/u:name[@type="primary"]', ns)[0].text
+    gene_names = []
+    for el in resp.root.findall('u:entry/u:gene/u:name', ns):
+        gene_name = {'name':el.text, 'type':el.attrib['type']}
+        gene_names.append(gene_name)
+    try:
+        gene_id = resp.root.findall(
+            'u:entry/u:dbReference[@type="GeneID"]', ns)[0].attrib['id']
+    except IndexError:
+        gene_id = None
     entry_name = resp.root.findall('u:entry/u:name', ns)[0].text
     go_els = resp.root.findall('u:entry/u:dbReference[@type="GO"]', ns)
+    accessions = [el.text for el in resp.root.findall('u:entry/u:accession', ns)]
     molecular_functions = []
+    cellular_localisations = []
+    biological_processes = []
     for go_el in go_els:
         term_property_value = go_el.findall(
             'u:property[@type="term"]', ns)[0].attrib['value']
         if term_property_value[0:2] == 'F:':
             molecular_functions.append('GO_' + go_el.attrib['id'][3:])
+        if term_property_value[0:2] == 'C:':
+            cellular_localisations.append('GO_' + go_el.attrib['id'][3:])
+        if term_property_value[0:2] == 'P:':
+            biological_processes.append('GO_' + go_el.attrib['id'][3:])
+    citations = []
+    for el in resp.root.findall('u:entry/u:reference', ns):
+        try:
+            doi = el.findall('u:citation/u:dbReference[@type="DOI"]', ns)[0].attrib['id']
+            pmid = el.findall('u:citation/u:dbReference[@type="PubMed"]', ns)[0].attrib['id']
+            citations.append({'doi':doi, 'pmid':pmid})
+        except IndexError:
+            continue
+    alternative_names = []
+    for el in resp.root.findall('u:entry/u:protein/u:alternativeName', ns):
+        alternative_name = {'full':el.findall('u:fullName',ns)[0].text}
+        if el.findall('u:shortName', ns):
+            alternative_name['short'] = el.findall('u:shortName', ns)[0].text
+        alternative_names.append(alternative_name)
+
+    db_references = resp.root.findall('u:entry/u:dbReference[@type="Pfam"]', ns)
+    domains = []
+    for db_reference in db_references:
+        name = db_reference.attrib["id"]
+        domains.append(name)
+
     return {'recommended_name': recommended_name,
+            'recommended_short_name': recommended_short_name,
             'organism': int(organism),
-            'gene': gene,
+            'gene_id': int(gene_id) if gene_id else None,
+            'accessions': accessions,
+            'gene_names': gene_names,
             'entry_name': entry_name,
             'short_name': entry_name.split('_')[0],
-            'molecular_functions': molecular_functions
+            'molecular_functions': molecular_functions,
+            'domains': domains,
+            'cellular_localisations': cellular_localisations,
+            'biological_processes': biological_processes,
+            'citations': citations,
+            'alternative_names': alternative_names
             }
 
 
@@ -272,6 +333,8 @@ def get_pdb_uniprot_mapping(pdb_id):
     pdb_id = pdb_id.lower()
     resp = requests.get(
         'https://www.ebi.ac.uk/pdbe/api/mappings/uniprot/{}'.format(pdb_id.lower()))
+    if resp.status_code != 200:
+        raise EntryNotFoundError(pdb_id)
     uniprot_ids = list(resp.json()[pdb_id]['UniProt'].keys())
     return uniprot_ids