diff --git a/ippisite/ippidb/admin.py b/ippisite/ippidb/admin.py
index 6253bfa4ae4ae5033aea3b87f0360a72e6344892..886f7d3395b8bdd61b2b4f650564ad2390010aea 100644
--- a/ippisite/ippidb/admin.py
+++ b/ippisite/ippidb/admin.py
@@ -70,6 +70,9 @@ class TextActivityDescription(admin.ModelAdmin):
@admin.register(Ppi)
class PpiModelAdmin(admin.ModelAdmin):
filter_horizontal = ('diseases',)
+ list_display = ('pdb_id', 'name', 'symmetry', 'family')
+ list_filter = ('diseases',)
+ search_fields = ('pdb_id', 'name', 'symmetry', 'family__name', 'diseases__name')
@admin.register(ProteinDomainComplex)
diff --git a/ippisite/ippidb/management/commands/import_v1_data.py b/ippisite/ippidb/management/commands/import_v1_data.py
index 6386137e879898b8146389dd9c2ef2f49418f439..04bc53a12db29aad5f03b7f74e1acb93acec9276 100644
--- a/ippisite/ippidb/management/commands/import_v1_data.py
+++ b/ippisite/ippidb/management/commands/import_v1_data.py
@@ -1,9 +1,13 @@
import glob
import re
+import copy
from django.utils import timezone
+from django.db import transaction
from django.core.management import BaseCommand, CommandError
import mysql.connector
+import requests_cache
+import click
from ippidb.models import Bibliography, Protein, Taxonomy, MolecularFunction, \
Domain, ProteinDomainBoundComplex, ProteinDomainPartnerComplex, Symmetry, Ppi, PpiComplex, Disease, \
@@ -24,732 +28,868 @@ class MyConverter(mysql.connector.conversion.MySQLConverter):
return[to_unicode(col) for col in row]
+class ImportTask(object):
+
+ description = "Abstract import task"
+
+ option = ""
+
+ target_classes = {}
+
+ main_class = None
+
+ depends_on = []
+
+ def __init__(self, command, traceback=False, stop_on_fail=False, progress_bar=False):
+ self.out_stream = command.stdout
+ self.err_stream = command.stderr
+ self.style = command.style
+ self.traceback = traceback
+ self.stop_on_fail = stop_on_fail
+ self.progress_bar = progress_bar
+ self.done = False
+
+ def _flush_target_models(self):
+ for target_class in self.target_classes:
+ self.out_stream.write(
+ 'Flushing {target_class} models...'.format(target_class=target_class.__name__))
+ target_class.objects.all().delete()
+ self.out_stream.write(self.style.SUCCESS(
+ 'Successfully flushed {target_class} models!'.format(target_class=target_class.__name__)))
+
+ def check_final_count(self):
+ if self.main_class:
+ count = self.main_class.objects.count()
+ if count==self.source_count:
+ self.out_stream.write(self.style.SUCCESS(
+ 'rows count ok for model {}: expected {}, counted {}'.format(self.main_class.__name__, self.source_count, count)))
+ else:
+ message = 'rows count not ok for model {}: expected {}, counted {}'.format(
+ self.main_class.__name__, self.source_count, count)
+ if self.stop_on_fail:
+ raise CommandError(message)
+ else:
+ self.out_stream.write(self.style.ERROR(message))
+
+ def migrate_row(self,row):
+ raise NotImplementedError()
+
+ def open_data_source(self):
+ raise NotImplementedError()
+
+ def post_process(self):
+ pass
+
+ def _process_rows(self, rows):
+ for row in rows:
+ try:
+ new_object = None
+ with transaction.atomic():
+ new_object = self.migrate_row(row)
+ except Exception as e:
+ if self.traceback:
+ import traceback
+ self.err_stream.write(self.style.NOTICE(traceback.format_exc()))
+ if self.stop_on_fail:
+ raise CommandError(
+ 'Failed inserting {}'.format(new_object))
+ else:
+ self.out_stream.write(self.style.ERROR(
+ 'Failed inserting {}'.format(new_object)))
+ else:
+ if self.progress_bar == False:
+ self.out_stream.write(self.style.SUCCESS(
+ 'Successfully inserted {}'.format(new_object)))
+
+ def count_source(self):
+ self.source_count = len(self.rows)
+
+ def _run_import_loop(self):
+ self.open_data_source()
+ self.count_source()
+ if self.progress_bar is True:
+ with click.progressbar(self.rows,
+ label='Importing ' + self.description + ' (' + str(self.source_count) + ' rows to process).') as rows_list:
+ self._process_rows(rows_list)
+ else:
+ self._process_rows(self.rows)
+ self.post_process()
+ self.check_final_count()
+ self.done = True
+
+ def check(self):
+ self.open_data_source()
+ self.count_source()
+ self.check_final_count()
+
+ def run(self):
+ self._flush_target_models()
+ self._run_import_loop()
+
+class MysqlImportTask(ImportTask):
+
+ description = "Abstract MySQL import task"
+
+ outer_sql = ""
+
+ def set_mysql_conn(self, conn):
+ self.conn = conn
+
+ def get_cursor(self):
+ return self.conn.cursor()
+
+ def open_data_source(self):
+ cursor = self.get_cursor()
+ cursor.execute(self.outer_sql)
+ self.rows = cursor.fetchall()
+
+class ListImportTask(ImportTask):
+
+ description = "Abstract Python list import task"
+
+ DATA = []
+
+ def open_data_source(self):
+ self.rows = self.DATA
+
+class SymmetriesImportTask(ListImportTask):
+
+ description = "Symmetries import"
+
+ option = "symmetries"
+
+ target_classes = [Symmetry]
+
+ main_class = Symmetry
+
+ DATA = [
+ ['AS', 'asymmetric'],
+ ['C2', 'C2 symmetry'],
+ ['D2', 'D2 symmetry'],
+ ['C3', 'C3 symmetry'],
+ ['D3', 'D3 symmetry'],
+ ['C4', 'C4 symmetry'],
+ ['D4', 'D4 symmetry'],
+ ['C5', 'C5 symmetry'],
+ ['D5', 'D5 symmetry']
+ ]
+
+ def migrate_row(self, row):
+ symmetry = Symmetry()
+ symmetry.code = row[0]
+ symmetry.description = row[1]
+ symmetry.save()
+ return symmetry
+
+
+class ProteinsImportTask(MysqlImportTask):
+
+ description = "Proteins import"
+
+ outer_sql = "SELECT * FROM protein"
+
+ option = "prot"
+
+ target_classes = [Protein]
+
+ main_class = Protein
+
+ def migrate_row(self, row):
+ p = Protein()
+ p.id = row[0]
+ p.uniprot_id = row[1]
+ p.save(autofill=True)
+ return p
+
+
+class BibliographyImportTask(MysqlImportTask):
+
+ description = "Bibliography references import"
+
+ outer_sql = "SELECT * FROM biblio"
+
+ option = "bib"
+
+ target_classes = [Bibliography]
+
+ main_class = Bibliography
+
+ def migrate_row(self, row):
+ b = Bibliography()
+ b.id = row[0]
+ if row[1] == 'article':
+ b.source = 'PM'
+ else:
+ b.source = 'PT'
+ b.id_source = row[2]
+ b.cytotox = row[6]=='Y'
+ b.in_silico = row[7]=='Y'
+ b.in_vitro = row[8]=='Y'
+ b.in_vivo = row[9]=='Y'
+ b.in_cellulo = row[10]=='Y'
+ b.pharmacokinetic = row[11]=='Y'
+ b.xray = row[12]=='Y'
+ b.save(autofill=True)
+ return b
+
+
+class DomainImportTask(MysqlImportTask):
+
+ description = "Domains import"
+
+ outer_sql= "SELECT * FROM domain"
+
+ option = "dom"
+
+ target_classes = [Domain, Taxonomy, MolecularFunction]
+
+ main_class = Domain
+
+ def migrate_row(self, row):
+ d = Domain()
+ d.id = row[0]
+ d.pfam_acc = row[2]
+ d.domain_family = row[4]
+ d.save(autofill=True)
+ return d
+
+
+class CompoundImportTask(MysqlImportTask):
+
+ description = "Compound import"
+
+ outer_sql= "SELECT * FROM compound"
+
+ option = "com"
+
+ target_classes = [Compound]
+
+ main_class = Compound
+
+ def migrate_row(self, row):
+ compound = Compound()
+ compound.id = row[0]
+ compound.canonical_smile = row[1]
+ compound.is_macrocycle = (row[4] == 'Y')
+ compound.aromatic_ratio = row[5]
+ compound.balaban_index = row[6]
+ compound.fsp3 = row[7] # Csp3Ratio
+ compound.gc_molar_refractivity = row[
+ 10] # GCMolarRefractivity
+ compound.log_d = row[13] # LogD
+ compound.a_log_p = row[14] # ALogP
+ compound.mean_atom_vol_vdw = row[15] # MeanAtomVolVdW
+ compound.molecular_weight = row[16] # MolecularWeight
+ compound.nb_acceptor_h = row[17] # NbAcceptorH
+ compound.nb_aliphatic_amines = row[
+ 18] # NbAliphaticsAmines
+ compound.nb_aromatic_bonds = row[19] # NbAromaticBonds
+ compound.nb_aromatic_ether = row[20] # NbAromaticsEther
+ compound.nb_aromatic_sssr = row[21] # NbAromaticsSSSR
+ compound.nb_atom = row[22] # NbAtom
+ compound.nb_atom_non_h = row[23] # NbAtomNonH
+ compound.nb_benzene_like_rings = row[24] # NbBenzLikeRings
+ compound.nb_bonds = row[25] # NbBonds
+ compound.nb_bonds_non_h = row[26] # NbBondsNonH
+ compound.nb_br = row[27] # NbBr
+ compound.nb_c = row[28] # NbC
+ compound.nb_chiral_centers = row[29] # NbChiralCenters
+ compound.nb_circuits = row[30] # NbCircuits
+ compound.nb_cl = row[31] # NbCl
+ compound.nb_csp2 = row[32] # NbCsp2
+ compound.nb_csp3 = row[33] # NbCsp3
+ compound.nb_donor_h = row[34] # NbDonorH
+ compound.nb_double_bonds = row[35] # NbDoubleBonds
+ compound.nb_f = row[36] # NbF
+ compound.nb_i = row[37] # NbI
+ compound.nb_multiple_bonds = row[38] # NbMultBonds
+ compound.nb_n = row[39] # NbN
+ compound.nb_o = row[40] # NbO
+ compound.nb_rings = row[41] # NbRings
+ compound.nb_rotatable_bonds = row[42] # NbRotatableBonds
+ compound.randic_index = row[44] # RandicIndex
+ compound.rdf070m = row[45] # RDF070m
+ compound.rotatable_bond_fraction = row[
+ 46] # RotatableBondFraction
+ compound.sum_atom_polar = row[47] # SumAtomPolar
+ compound.sum_atom_vol_vdw = row[48] # SumAtomVolVdW
+ compound.tpsa = row[51] # TPSA
+ compound.ui = row[52] # Ui
+ compound.wiener_index = row[54] # WienerIndex
+ if row[55] != 'N':
+ compound.common_name = row[55] # CmpdNameSh
+ compound.pubchem_id = row[56] # IdPubchem
+ if row[57] != 'N':
+ compound.chemspider_id = row[57] # IdPubchem
+ compound.chembl_id = row[58]
+ compound.iupac_name = row[59]
+ compound.save(autofill=True)
+ return compound
+
+
+class RefCompoundBiblioImportTask(MysqlImportTask):
+
+ description = "RefCompoundBiblio import"
+
+ outer_sql= """
+ select r.CmpdNameInBiblio, c.CanonicalSmile, b.IDSource
+ from refCmpdBiblio as r inner join compound as c
+ on r.IDCompound=c.IDCompound
+ inner join biblio as b on r.IDBiblio=b.IDBiblio;"""
+
+ option = "rcb"
+
+ target_classes = [RefCompoundBiblio]
+
+ main_class = RefCompoundBiblio
+
+ depends_on = [BibliographyImportTask, CompoundImportTask]
+
+ def migrate_row(self, row):
+ c = Compound.objects.get(canonical_smile=row[1])
+ b = Bibliography.objects.get(id_source=row[2])
+ r = RefCompoundBiblio()
+ r.compound_id = c.id
+ r.bibliography_id = b.id
+ r.compound_name = '-'.join(re.split('[-.]',row[0])[1:])
+ # mysql format for this field is [PMID/WIPOID]-name: we remove the first part
+ # sometimes there is a . instead of the dash
+ # sometimes we have more than one -
+ r.save()
+ return r
+
+
+class PpiImportTask(MysqlImportTask):
+
+ description = "PPIs import"
+
+ outer_sql= """
+ select distinct protein.NumUniprot, domain.PfamNumAccession, complexe.NbCopy, cmpdAction.IDComplexeBound, bindingSiteEvidence.CodePDB,
+ 'part1', ppi.IDPPI, disease.Disease, complexe.IDComplexe, ppi.Family from bindingSite inner join ppi on (bindingSite.IDBindingSite=ppi.IDBindingSite1)
+ inner join complexe on (ppi.IDComplexe1=complexe.IDComplexe) left outer join cmpdAction on (complexe.IDComplexe=cmpdAction.IDComplexeBound)
+ inner join protein on (bindingSite.IDProtein=protein.IDProtein) inner join domain on (bindingSite.IDDomain=domain.IDDomain)
+ inner join disease on (disease.IDPPI=ppi.IDPPI) left outer join bindingSiteEvidence on (ppi.IDPPI=bindingSiteEvidence.IDPPI)
+ union
+ select distinct protein.NumUniprot, domain.PfamNumAccession , complexe.NbCopy, cmpdAction.IDComplexeBound, null,
+ 'part2', ppi.IDPPI, disease.Disease, complexe.IDComplexe, ppi.Family from bindingSite inner join ppi on (bindingSite.IDBindingSite=ppi.IDBindingSite2)
+ inner join complexe on (ppi.IDComplexe2=complexe.IDComplexe) left outer join cmpdAction on (complexe.IDComplexe=cmpdAction.IDComplexeBound)
+ inner join protein on (bindingSite.IDProtein=protein.IDProtein) inner join domain on (bindingSite.IDDomain=domain.IDDomain)
+ inner join disease on (disease.IDPPI=ppi.IDPPI)
+ """
+
+ option = "ppi"
+
+ target_classes = [ProteinDomainBoundComplex, ProteinDomainPartnerComplex, Disease, Ppi, PpiFamily, PpiComplex, CompoundAction]
+
+ main_class = Ppi
+
+ depends_on = [ProteinsImportTask, DomainImportTask, CompoundImportTask]
+
+ def migrate_row(self, row):
+ cursor_aux = self.get_cursor()
+ cursor_aux2 = self.get_cursor()
+ # create or retrieve Ppi object
+ if row[5] == 'part1':
+ ppi = Ppi()
+ ppi.id = row[6]
+ disease, created = Disease.objects.get_or_create(
+ name=row[7])
+ ppi_family, created = PpiFamily.objects.get_or_create(
+ name=row[9].strip())
+ ppi.family = ppi_family
+ ppi.pdb_id = row[4]
+ ppi.pockets_nb = 1
+ ppi.symmetry = Symmetry.objects.get(code='AS')
+ ppi.save()
+ ppi.diseases.add(disease)
+ ppi.save()
+ else:
+ ppi = Ppi.objects.get(id=row[6])
+ # create a complex
+ if row[3] is None:
+ c = ProteinDomainPartnerComplex()
+ else:
+ c = ProteinDomainBoundComplex()
+ c.id = row[8]
+ protein = Protein.objects.get(uniprot_id=row[0])
+ c.protein = protein
+ domain = Domain.objects.get(pfam_acc=row[1])
+ c.domain = domain
+ c.ppc_copy_nb = row[2]
+ if isinstance(c, ProteinDomainBoundComplex):
+ c.ppp_copy_nb_per_p = 1
+ c.save()
+ # create the PpiComplex object
+ ppi_complex = PpiComplex()
+ ppi_complex.ppi = ppi
+ ppi_complex.complex = c
+ ppi_complex.cc_nb = 1
+ ppi_complex.save()
+ if row[3] is not None:
+ sql_ca_string = '''select distinct c.CanonicalSmile from cmpdAction as ca
+ inner join compound as c on ca.IDCompound=c.IDCompound where ca.IDComplexeBound='''\
+ + str(row[3])
+ cursor_aux.execute(sql_ca_string)
+ self.rows_aux = cursor_aux.fetchall()
+ for row_aux in self.rows_aux:
+ canonical_smile = row_aux[0]
+ sql_bse_string = '''select CodePDB from bindingSiteEvidence where IDPPI=''' + str(ppi.id)
+ cursor_aux2.execute(sql_bse_string)
+ row_aux2 = cursor_aux2.fetchone()
+ if row_aux2 is None:
+ pdb_id_ca = None
+ else:
+ pdb_id_ca = row_aux2[0]
+ c = Compound.objects.get(canonical_smile=canonical_smile)
+ ca = CompoundAction()
+ ca.compound = c
+ ca.ppi = ppi
+ ca.activation_mode = 'U'
+ ca.nb_copy_compounds = 1
+ ca.pdb_id = pdb_id_ca
+ ca.save()
+ return ppi
+
+ def post_process(self):
+ # add names to PPIs automatically once the PPI complexes which the names
+ # depend on have been inserted
+ for ppi in Ppi.objects.all():
+ ppi.save(autofill=True)
+
+
+ def check_final_count(self):
+ self.source_count = int(self.source_count/2)
+ super().check_final_count()
+
+class AdditionalCasImportTask(ListImportTask):
+
+ description = "Additional Compound Actions import (hardcoded because not in the source file)"
+
+ DATA = [
+ {'compound_id':15, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':36, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':57, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':69, 'ppi_name': 'IL2 / IL2RA'},
+ {'compound_id':78, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':110, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':111, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':120, 'ppi_name': 'XIAP / DBLOH'},
+ {'compound_id':124, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':144, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':171, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':213, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':221, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':238, 'ppi_name': 'IL2 / IL2RA'},
+ {'compound_id':272, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':305, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':316, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':354, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':355, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':399, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':406, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':409, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':412, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':458, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':463, 'ppi_name': 'IL2 / IL2RA'},
+ {'compound_id':465, 'ppi_name': 'XIAP / DBLOH'},
+ {'compound_id':469, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':477, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':485, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':487, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':490, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':538, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':563, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':585, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':590, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':626, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':666, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':683, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':702, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':705, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':714, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':734, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':769, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':786, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':801, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':814, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':817, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':886, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':971, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':977, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':999, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1011, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1026, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1031, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1055, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1056, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1069, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1115, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1122, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1126, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1153, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1155, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1160, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1202, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1221, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1251, 'ppi_name': 'XIAP / DBLOH'},
+ {'compound_id':1268, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1280, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1296, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1297, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1310, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1313, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1317, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1376, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1398, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1402, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1403, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1407, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1434, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1453, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1477, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1481, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1504, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1514, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1553, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1594, 'ppi_name': 'BCL2 / BAK'},
+ {'compound_id':1648, 'ppi_name': 'MDM2 / P53'},
+ {'compound_id':1649, 'ppi_name': 'MDM2 / P53'},
+ ]
+
+ option = "acas"
+
+ target_classes = [CompoundAction]
+
+ main_class = CompoundAction
+
+ depends_on = [CompoundImportTask, PpiImportTask]
+
+ def migrate_row(self, row):
+ ca = CompoundAction()
+ ca.compound = Compound.objects.get(id=row['compound_id'])
+ ca.ppi = Ppi.objects.get(name=row['ppi_name'])
+ ca.activation_mode = 'U'
+ ca.nb_copy_compounds = 1
+ ca.save()
+ return ca
+
+
+class TestActivityDescriptionImportTask(MysqlImportTask):
+
+ description = "TestActivityDescription import"
+
+ outer_sql= """
+ SELECT * FROM testActivityDescription
+ """
+
+ option = "tad"
+
+ target_classes = [TestActivityDescription]
+
+ main_class = TestActivityDescription
+
+ depends_on = [BibliographyImportTask, PpiImportTask]
+
+ def migrate_row(self, row):
+ tad = TestActivityDescription()
+ tad.id = row[0]
+ cursor_biblio = self.get_cursor()
+ cursor_biblio.execute(
+ """select IDSource from biblio where IDBiblio={}""".format(row[2]))
+ biblio_row = cursor_biblio.fetchone()
+ biblio = Bibliography.objects.get(id_source=biblio_row[0])
+ tad.biblio = biblio
+ tad.protein_domain_bound_complex = ProteinDomainBoundComplex.objects.get(id=row[1])
+ tad.ppi = Ppi.objects.get(id=row[3])
+ tad.test_name = row[4]
+ tad.is_primary = row[5]=='Y'
+ if row[6]=='Unspecified':
+ tad.protein_bound_construct = 'U'
+ elif row[6]=='Full length':
+ tad.protein_bound_construct = 'F'
+ tad.test_type = row[7].upper()
+ tad.test_modulation_type = row[8][0]
+ tad.nb_active_compounds = row[9]
+ if row[16] is not None:
+ tad.cell_line, created = CellLine.objects.get_or_create(
+ name=row[16])
+ tad.save()
+ return tad
+
+class CompoundActivityResultImportTask(MysqlImportTask):
+
+ description = "CompoundActivityResult import"
+
+ outer_sql= """
+ select c.CanonicalSmile, r.IDTestActivity, r.ActivityType,
+ r.Activity, r.PourcentInhib from cmpdActiveResult as r
+ inner join compound as c on r.IDCompound = c.IDCompound;
+ """
+
+ option = "car"
+
+ target_classes = [CompoundActivityResult]
+
+ main_class = CompoundActivityResult
+
+ depends_on = [TestActivityDescriptionImportTask, CompoundImportTask]
+
+ def migrate_row(self, row):
+ car = CompoundActivityResult()
+ car.test_activity_description = TestActivityDescription.objects.get(id=row[1])
+ car.compound = Compound.objects.get(canonical_smile=row[0])
+ car.activity_type = row[2]
+ car.activity = row[3]
+ car.inhibition_percentage = row[4]
+ car.modulation_type = 'I' # because previous DB is only about inhibitors
+ car.save()
+ return car
+
+
+class TestCytotoxDescriptionImportTask(MysqlImportTask):
+
+ description = "TestCytotoxDescription import"
+
+ outer_sql= """
+ SELECT * FROM testCytotoxDescription
+ """
+
+ option = "tcd"
+
+ target_classes = [TestCytotoxDescription]
+
+ main_class = TestCytotoxDescription
+
+ depends_on = [BibliographyImportTask, CompoundImportTask]
+
+ def migrate_row(self, row):
+ tcd = TestCytotoxDescription()
+ tcd.id = row[0]
+ cursor_biblio = self.get_cursor()
+ cursor_biblio.execute(
+ """select IDSource from biblio where IDBiblio={}""".format(row[1]))
+ biblio_row = cursor_biblio.fetchone()
+ biblio = Bibliography.objects.get(id_source=biblio_row[0])
+ tcd.biblio = biblio
+ tcd.test_name = row[2]
+ tcd.compound_concentration = row[4]
+ if row[3] is not None:
+ tcd.cell_line, created = CellLine.objects.get_or_create(
+ name=row[3])
+ tcd.save()
+ return tcd
+
+class CompoundCytotoxicityResultImportTask(MysqlImportTask):
+
+ description = "CompoundCytotoxicityResult import"
+
+ outer_sql= """
+ select c.CanonicalSmile, r.IDTestCytotox, r.IDTestCytotox
+ from cmpdCytotoxResult as r inner join compound as c
+ on r.IDCompound = c.IDCompound;
+ """
+
+ option = "ccr"
+
+ target_classes = [CompoundCytotoxicityResult]
+
+ main_class = CompoundCytotoxicityResult
+
+ depends_on = [CompoundImportTask, TestCytotoxDescriptionImportTask]
+
+ def migrate_row(self, row):
+ ccr = CompoundCytotoxicityResult()
+ ccr.test_cytotoxicity_description = TestCytotoxDescription.objects.get(id=row[1])
+ ccr.compound = Compound.objects.get(canonical_smile=row[0])
+ ccr.toxicity = row[2] == 'Y'
+ ccr.save()
+ return ccr
+
+
+class TestPKDescriptionImportTask(MysqlImportTask):
+
+ description = "TestPKDescription import"
+
+ outer_sql= """
+ SELECT * FROM testPKDescription
+ """
+
+ option = "tpd"
+
+ target_classes = [TestPKDescription]
+
+ main_class = TestPKDescription
+
+ depends_on = [BibliographyImportTask, CompoundImportTask]
+
+ def migrate_row(self, row):
+ tpd = TestPKDescription()
+ tpd.id = row[0]
+ cursor_biblio = self.get_cursor()
+ cursor_biblio.execute(
+ """select IDSource from biblio where IDBiblio={}""".format(row[1]))
+ biblio_row = cursor_biblio.fetchone()
+ biblio = Bibliography.objects.get(id_source=biblio_row[0])
+ tpd.biblio = biblio
+ tpd.test_name = row[2]
+ #tcd.compound_concentration = row[4]
+ try:
+ taxonomy = Taxonomy.objects.get(taxonomy_id=10090)
+ except Taxonomy.DoesNotExist:
+ taxonomy = Taxonomy()
+ taxonomy.taxonomy_id = 10090
+ # dirty hack: all organisms in this table are "mice",
+ # hence assuming Mus musculus
+ taxonomy.save(autofill=True)
+ tpd.organism = taxonomy
+ tpd.administration_mode = row[4]
+ tpd.concentration = row[5]
+ tpd.dose = row[6]
+ tpd.dose_interval = row[7]
+ tpd.save()
+ return tpd
+
+
+class CompoundPKResultImportTask(MysqlImportTask):
+
+ description = "CompoundPKResult import"
+
+ outer_sql= """
+ select c.CanonicalSmile, r.IDTestPK, r.Tolerated, r.AUC, r.Clearance,
+ r.Cmax, r.OralBioavailability, r.Tdemi, r.Tmax, r.VolDistribution
+ from cmpdPKResult as r inner join compound as c
+ on r.IDCompound = c.IDCompound;
+ """
+
+ option = "cpr"
+
+ target_classes = [CompoundPKResult]
+
+ main_class = CompoundPKResult
+
+ depends_on = [CompoundImportTask, TestPKDescriptionImportTask]
+
+ def migrate_row(self, row):
+ cpr = CompoundPKResult()
+ cpr.test_pk_description = TestPKDescription.objects.get(id=row[1])
+ cpr.compound = Compound.objects.get(canonical_smile=row[0])
+ cpr.tolerated = row[2] == 'Y'
+ cpr.auc = row[3]
+ cpr.clearance = row[4]
+ cpr.c_max = row[5]
+ cpr.oral_bioavailability = row[6]
+ cpr.t_demi = row[7]
+ cpr.t_max = row[8]
+ cpr.voldistribution = row[9]
+ cpr.save()
+ return cpr
class Command(BaseCommand):
+ help = "Import iPPI-DB data from the MySQL database"
- help = "Import data from the local v1 database"
+ task_classes = [ProteinsImportTask,
+ BibliographyImportTask,
+ DomainImportTask,
+ CompoundImportTask,
+ SymmetriesImportTask,
+ PpiImportTask,
+ RefCompoundBiblioImportTask,
+ AdditionalCasImportTask,
+ TestActivityDescriptionImportTask,
+ CompoundActivityResultImportTask,
+ TestCytotoxDescriptionImportTask,
+ CompoundCytotoxicityResultImportTask,
+ TestPKDescriptionImportTask,
+ CompoundPKResultImportTask
+ ]
def add_arguments(self, parser):
+ task_choices = [task_class.option for task_class in self.task_classes]
+ task_choices.append('all')
+ task_help = 'Import task to be run.\n '
+ task_help += ',\n \n'.join([task_class.option + ': ' + task_class.description for task_class in self.task_classes])
+ task_help += ',\n \nall: import everything.'
parser.add_argument(
- '--bibliographies',
- action='store_true',
- dest='bibliographies',
- default=False,
- help='Flush and migrate bibliographies',
+ 'task',
+ type=str,
+ choices=task_choices,
+ help=task_help,
)
parser.add_argument(
- '--proteins',
+ '--all',
action='store_true',
- dest='proteins',
+ dest='all',
default=False,
- help='Flush and migrate proteins',
+ help='import everything',
)
parser.add_argument(
- '--domains',
+ '--wscache',
action='store_true',
- dest='domains',
+ dest='wscache',
default=False,
- help='Flush and migrate domains',
+ help='use web services cache',
)
parser.add_argument(
- '--symmetries',
+ '--errortb',
action='store_true',
- dest='symmetries',
+ dest='errortb',
default=False,
- help='Flush and create symmetries',
+ help='show tracebacks on errors',
)
parser.add_argument(
- '--ppi',
- action='store_true',
- dest='ppi',
- default=False,
- help='Flush and migrate ppis, complexes, and tests',
- )
- parser.add_argument(
- '--compound',
+ '--stoponfail',
action='store_true',
- dest='compound',
+ dest='stoponfail',
default=False,
- help='Flush and migrate compounds',
+ help='stop on first error',
)
parser.add_argument(
- '--compound-biblio',
+ '--progressbar',
action='store_true',
- dest='compoundbiblio',
+ dest='progress_bar',
default=False,
- help='Flush and migrate compounds-bibliography',
+ help='show progressbar instead of logging inserted entries',
)
parser.add_argument(
- '--stoponfail',
+ '--check',
action='store_true',
- dest='stoponfail',
+ dest='check',
default=False,
- help='Stop on fail',
+ help='only check instead of running import',
)
def handle(self, *args, **options):
conn = mysql.connector.connect(
converter_class=MyConverter, host="localhost", user="root", password="ippidb", database="ippidb")
- cursor = conn.cursor()
- cursor_aux = conn.cursor(buffered=True)
- cursor_aux2 = conn.cursor(buffered=True)
- if options['bibliographies']:
- cursor.execute("""SELECT * FROM biblio""")
- rows = cursor.fetchall()
- Bibliography.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed bibliography table'))
- for row in rows:
- try:
- b = Bibliography()
- if row[1] == 'article':
- b.source = 'PM'
- else:
- b.source = 'PT'
- b.id_source = row[2]
- b.cytotox = row[6]=='Y'
- b.in_silico = row[7]=='Y'
- b.in_vitro = row[8]=='Y'
- b.in_vivo = row[9]=='Y'
- b.in_cellulo = row[10]=='Y'
- b.pharmacokinetic = row[11]=='Y'
- b.xray = row[12]=='Y'
- b.save(autofill=True)
- except Exception as e:
- if options['stoponfail']:
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting {} {}'.format(row[1], row[2]))
- else:
- self.stdout.write(
- self.style.ERROR('Failed inserting {} {}'.format(row[1], row[2])))
- else:
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted {} {}'.format(row[1], row[2])))
- if options['proteins']:
- cursor.execute("""SELECT * FROM protein""")
- rows = cursor.fetchall()
- Protein.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed protein table'))
- Taxonomy.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed taxonomy table'))
- MolecularFunction.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed molecular function table'))
- for row in rows:
- try:
- p = Protein()
- p.uniprot_id = row[1]
- p.save(autofill=True)
- except Exception as e:
- if options['stoponfail']:
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting {} {}'.format(row[1], row[2]))
- else:
- self.stdout.write(
- self.style.ERROR('Failed inserting {} {}'.format(row[1], row[2])))
- else:
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted {} {}'.format(row[1], row[2])))
- if options['domains']:
- cursor.execute("""SELECT * FROM domain""")
- rows = cursor.fetchall()
- Domain.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed domain table'))
- for row in rows:
- try:
- p = Domain()
- p.pfam_acc = row[2]
- p.domain_family = row[4]
- p.save(autofill=True)
- except Exception as e:
- if options['stoponfail']:
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting {} {}'.format(row[1], row[2]))
- else:
- self.stdout.write(
- self.style.ERROR('Failed inserting {} {}'.format(row[1], row[2])))
- else:
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted {} {}'.format(row[1], row[2])))
- if options['symmetries']:
- Symmetry.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed symmetries table'))
- rows = [
- ['AS', 'asymmetric'],
- ['C2', 'C2 symmetry'],
- ['D2', 'D2 symmetry'],
- ['C3', 'C3 symmetry'],
- ['D3', 'D3 symmetry'],
- ['C4', 'C4 symmetry'],
- ['D4', 'D4 symmetry'],
- ['C5', 'C5 symmetry'],
- ['D5', 'D5 symmetry'],
- ]
- for row in rows:
- try:
- symmetry = Symmetry()
- symmetry.code = row[0]
- symmetry.description = row[1]
- symmetry.save()
- except Exception as e:
- if options['stoponfail']:
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting {} {}'.format(row[0], row[1]))
- else:
- self.stdout.write(
- self.style.ERROR('Failed inserting {} {}'.format(row[0], row[1])))
- else:
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted {} {}'.format(row[0], row[1])))
- if options['ppi']:
- sql_request_string = '''
-select distinct protein.NumUniprot, domain.PfamNumAccession, complexe.NbCopy, cmpdAction.IDComplexeBound, bindingSiteEvidence.CodePDB,
- 'part1', ppi.IDPPI, disease.Disease, complexe.IDComplexe, ppi.Family from bindingSite inner join ppi on (bindingSite.IDBindingSite=ppi.IDBindingSite1)
- inner join complexe on (ppi.IDComplexe1=complexe.IDComplexe) left outer join cmpdAction on (complexe.IDComplexe=cmpdAction.IDComplexeBound)
- inner join protein on (bindingSite.IDProtein=protein.IDProtein) inner join domain on (bindingSite.IDDomain=domain.IDDomain)
- inner join disease on (disease.IDPPI=ppi.IDPPI) left outer join bindingSiteEvidence on (ppi.IDPPI=bindingSiteEvidence.IDPPI)
-union
-select distinct protein.NumUniprot, domain.PfamNumAccession , complexe.NbCopy, cmpdAction.IDComplexeBound, null,
- 'part2', ppi.IDPPI, disease.Disease, complexe.IDComplexe, ppi.Family from bindingSite inner join ppi on (bindingSite.IDBindingSite=ppi.IDBindingSite2)
- inner join complexe on (ppi.IDComplexe2=complexe.IDComplexe) left outer join cmpdAction on (complexe.IDComplexe=cmpdAction.IDComplexeBound)
- inner join protein on (bindingSite.IDProtein=protein.IDProtein) inner join domain on (bindingSite.IDDomain=domain.IDDomain)
- inner join disease on (disease.IDPPI=ppi.IDPPI)
- '''
- cursor.execute(sql_request_string)
- rows = cursor.fetchall()
- ProteinDomainBoundComplex.objects.all().delete()
- ProteinDomainPartnerComplex.objects.all().delete()
- Disease.objects.all().delete()
- Ppi.objects.all().delete()
- PpiFamily.objects.all().delete()
- PpiComplex.objects.all().delete()
- CompoundAction.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed protein domain complex, PPI, PPI families, compound actions, and disease tables'))
- ppi_ids_mapping = {}
- protein_domain_bound_complexes_mapping = {}
- for row in rows:
- try:
- # create or retrieve Ppi object
- if row[5] == 'part1':
- ppi = Ppi()
- disease, created = Disease.objects.get_or_create(
- name=row[7])
- ppi_family, created = PpiFamily.objects.get_or_create(
- name=row[9].strip())
- ppi.family = ppi_family
- ppi.pdb_id = row[4]
- ppi.pockets_nb = 1
- ppi.symmetry = Symmetry.objects.get(code='AS')
- ppi.save()
- ppi.diseases.add(disease)
- ppi.save()
- ppi_ids_mapping[row[6]] = ppi.id
- else:
- ppi = Ppi.objects.get(id=ppi_ids_mapping[row[6]])
- # create a complex
- if row[3] is None:
- c = ProteinDomainPartnerComplex()
- else:
- c = ProteinDomainBoundComplex()
- protein = Protein.objects.get(uniprot_id=row[0])
- c.protein = protein
- domain = Domain.objects.get(pfam_acc=row[1])
- c.domain = domain
- c.ppc_copy_nb = row[2]
- if isinstance(c, ProteinDomainBoundComplex):
- c.ppp_copy_nb_per_p = 1
- c.save()
- # save ProteinDomainBoundComplex/source ID in 'complexe' table
- if row[3] is not None:
- protein_domain_bound_complexes_mapping[row[8]] = c
- # create the PpiComplex object
- ppi_complex = PpiComplex()
- ppi_complex.ppi = ppi
- ppi_complex.complex = c
- ppi_complex.cc_nb = 1
- ppi_complex.save()
- if row[3] is not None:
- sql_ca_string = '''select distinct c.CanonicalSmile from cmpdAction as ca
- inner join compound as c on ca.IDCompound=c.IDCompound where ca.IDComplexeBound='''\
- + str(row[3])
- cursor_aux.execute(sql_ca_string)
- rows_aux = cursor_aux.fetchall()
- for row_aux in rows_aux:
- if row_aux is None:
- self.stdout.write(sql_ca_string)
- continue
- canonical_smile = row_aux[0]
- sql_bse_string = '''select CodePDB from bindingSiteEvidence where IDPPI=''' + str(ppi.id)
- cursor_aux2.execute(sql_bse_string)
- row_aux2 = cursor_aux2.fetchone()
- if row_aux2 is None:
- pdb_id_ca = None
- else:
- pdb_id_ca = row_aux2[0]
- c = Compound.objects.get(canonical_smile=canonical_smile)
- ca = CompoundAction()
- ca.compound = c
- ca.ppi = ppi
- ca.activation_mode = 'U'
- ca.nb_copy_compounds = 1
- ca.pdb_id = pdb_id_ca
- ca.save()
- except Exception as e:
- if options['stoponfail']:
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting {} {}'.format(row[0], row[1]))
- else:
- self.stdout.write(
- self.style.ERROR('Failed inserting {} {}'.format(row[0], row[1])))
- else:
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted {} {}'.format(row[0], row[1])))
- # add names to PPIs automatically once the PPI complexes which the names
- # depend on have been inserted
- for ppi in Ppi.objects.all():
- ppi.save(autofill=True)
- additional_cas = [
- {'compound_id':15, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':36, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':57, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':69, 'ppi_name': 'IL2 / IL2RA'},
- {'compound_id':78, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':110, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':111, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':120, 'ppi_name': 'XIAP / DBLOH'},
- {'compound_id':124, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':144, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':171, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':213, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':221, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':238, 'ppi_name': 'IL2 / IL2RA'},
- {'compound_id':272, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':305, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':316, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':354, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':355, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':399, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':406, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':409, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':412, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':458, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':463, 'ppi_name': 'IL2 / IL2RA'},
- {'compound_id':465, 'ppi_name': 'XIAP / DBLOH'},
- {'compound_id':469, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':477, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':485, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':487, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':490, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':538, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':563, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':585, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':590, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':626, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':666, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':683, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':702, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':705, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':714, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':734, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':769, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':786, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':801, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':814, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':817, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':886, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':971, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':977, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':999, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1011, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1026, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1031, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1055, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1056, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1069, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1115, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1122, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1126, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1153, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1155, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1160, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1202, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1221, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1251, 'ppi_name': 'XIAP / DBLOH'},
- {'compound_id':1268, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1280, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1296, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1297, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1310, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1313, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1317, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1376, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1398, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1402, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1403, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1407, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1434, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1453, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1477, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1481, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1504, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1514, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1553, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1594, 'ppi_name': 'BCL2 / BAK'},
- {'compound_id':1648, 'ppi_name': 'MDM2 / P53'},
- {'compound_id':1649, 'ppi_name': 'MDM2 / P53'},
- ]
- for ca_props in additional_cas:
- try:
- ca = CompoundAction()
- ca.compound = Compound.objects.get(id=ca_props['compound_id'])
- ca.ppi = Ppi.objects.get(name=ca_props['ppi_name'])
- ca.activation_mode = 'U'
- ca.nb_copy_compounds = 1
- ca.save()
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted CompoundAction {}'.format(ca)))
- except Exception as e:
- if options['stoponfail']:
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting CompoundAction for compound {} and ppi {}'.format(ca_props['compound_id'], ca_props['ppi_name']))
- else:
- self.stdout.write(
- self.style.ERROR('Failed inserting CompoundAction for compound {} and ppi {}'.format(ca_props['compound_id'], ca_props['ppi_name'])))
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting CompoundAction for compound {} and ppi {}'.format(ca_props['compound_id'], ca_props['ppi_name']))
- else:
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted CompoundAction {}'.format(ca)))
- CellLine.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed cell lines table'))
- cursor.execute("""SELECT * FROM testActivityDescription""")
- rows = cursor.fetchall()
- TestActivityDescription.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed test activity descriptions table'))
- tad_id_mapping = {}
- for row in rows:
- try:
- tad = TestActivityDescription()
- cursor.execute(
- """select IDSource from biblio where IDBiblio={}""".format(row[2]))
- biblio_row = cursor.fetchone()
- biblio = Bibliography.objects.get(id_source=biblio_row[0])
- tad.biblio = biblio
- tad.protein_domain_bound_complex = protein_domain_bound_complexes_mapping[row[1]]
- tad.ppi = Ppi.objects.get(id=ppi_ids_mapping[row[3]])
- tad.test_name = row[4]
- tad.is_primary = row[5]=='Y'
- if row[6]=='Unspecified':
- tad.protein_bound_construct = 'U'
- elif row[6]=='Full length':
- tad.protein_bound_construct = 'F'
- tad.test_type = row[7].upper()
- tad.test_modulation_type = row[8][0]
- tad.nb_active_compounds = row[9]
- if row[16] is not None:
- tad.cell_line, created = CellLine.objects.get_or_create(
- name=row[16])
- tad.save()
- tad_id_mapping[row[0]] = tad.id
- except Exception as e:
- if options['stoponfail']:
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting {} {}'.format(row[1], row[2]))
- else:
- self.stdout.write(
- self.style.ERROR('Failed inserting {} {}'.format(row[1], row[2])))
- else:
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted {}'.format(row[2])))
- cursor.execute("""select c.CanonicalSmile, r.IDTestActivity, r.ActivityType, r.Activity, r.PourcentInhib from cmpdActiveResult as r inner join compound as c on r.IDCompound = c.IDCompound;""")
- rows = cursor.fetchall()
- CompoundActivityResult.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed compound activity result table'))
- for row in rows:
- try:
- car = CompoundActivityResult()
- car.test_activity_description = TestActivityDescription.objects.get(id=tad_id_mapping[row[1]])
- car.compound = Compound.objects.get(canonical_smile=row[0])
- car.activity_type = row[2]
- car.activity = row[3]
- car.inhibition_percentage = row[4]
- car.modulation_type = 'I' # because previous DB is only about inhibitors
- car.save()
- except Exception as e:
- if options['stoponfail']:
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting {} {}'.format(row[1], row[2]))
- else:
- self.stdout.write(
- self.style.ERROR('Failed inserting {} {}'.format(row[1], row[2])))
- else:
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted {}'.format(str(car))))
- cursor.execute("""SELECT * FROM testCytotoxDescription""")
- rows = cursor.fetchall()
- TestCytotoxDescription.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed test cytotoxicity descriptions table'))
- tcd_id_mapping = {}
- for row in rows:
- try:
- tcd = TestCytotoxDescription()
- cursor.execute(
- """select IDSource from biblio where IDBiblio={}""".format(row[1]))
- biblio_row = cursor.fetchone()
- biblio = Bibliography.objects.get(id_source=biblio_row[0])
- tcd.biblio = biblio
- tcd.test_name = row[2]
- tcd.compound_concentration = row[4]
- if row[3] is not None:
- tcd.cell_line, created = CellLine.objects.get_or_create(
- name=row[3])
- tcd.save()
- tcd_id_mapping[row[0]] = tcd.id
- except Exception as e:
- if options['stoponfail']:
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting {} {}'.format(row[1], row[2]))
- else:
- self.stdout.write(
- self.style.ERROR('Failed inserting {} {}'.format(row[1], row[2])))
- else:
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted {}'.format(row[2])))
- cursor.execute("""select c.CanonicalSmile, r.IDTestCytotox, r.IDTestCytotox from cmpdCytotoxResult as r inner join compound as c on r.IDCompound = c.IDCompound;""")
- rows = cursor.fetchall()
- CompoundCytotoxicityResult.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed compound cytotoxicity result table'))
- for row in rows:
- try:
- ccr = CompoundCytotoxicityResult()
- ccr.test_cytotoxicity_description = TestCytotoxDescription.objects.get(id=tcd_id_mapping[row[1]])
- ccr.compound = Compound.objects.get(canonical_smile=row[0])
- ccr.toxicity = row[2] == 'Y'
- ccr.save()
- except Exception as e:
- if options['stoponfail']:
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting {} {}'.format(row[1], row[2]))
- else:
- import traceback
- self.stderr.write(traceback.format_exc())
- self.stdout.write(
- self.style.ERROR('Failed inserting {} {}'.format(row[1], row[2])))
- else:
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted {}'.format(str(ccr))))
- cursor.execute("""SELECT * FROM testPKDescription""")
- rows = cursor.fetchall()
- TestPKDescription.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed test PK descriptions table'))
- tpd_id_mapping = {}
- for row in rows:
- try:
- tpd = TestPKDescription()
- cursor.execute(
- """select IDSource from biblio where IDBiblio={}""".format(row[1]))
- biblio_row = cursor.fetchone()
- biblio = Bibliography.objects.get(id_source=biblio_row[0])
- tpd.biblio = biblio
- tpd.test_name = row[2]
- #tcd.compound_concentration = row[4]
- try:
- taxonomy = Taxonomy.objects.get(taxonomy_id=10090)
- except Taxonomy.DoesNotExist:
- taxonomy = Taxonomy()
- taxonomy.taxonomy_id = 10090
- # dirty hack: all organisms in this table are "mice",
- # hence assuming Mus musculus
- taxonomy.save(autofill=True)
- tpd.organism = taxonomy
- tpd.administration_mode = row[4]
- tpd.concentration = row[5]
- tpd.dose = row[6]
- tpd.dose_interval = row[7]
- tpd.save()
- tpd_id_mapping[row[0]] = tpd.id
- except Exception as e:
- if options['stoponfail']:
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting {} {}'.format(row[1], row[2]))
- else:
- self.stdout.write(
- self.style.ERROR('Failed inserting {} {}'.format(row[1], row[2])))
- else:
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted {}'.format(row[2])))
- cursor.execute("""select c.CanonicalSmile, r.IDTestPK, r.Tolerated, r.AUC, r.Clearance, r.Cmax, r.OralBioavailability, r.Tdemi, r.Tmax, r.VolDistribution from cmpdPKResult as r inner join compound as c on r.IDCompound = c.IDCompound;""")
- rows = cursor.fetchall()
- CompoundPKResult.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed compound cytotoxicity PK table'))
- for row in rows:
- try:
- cpr = CompoundPKResult()
- cpr.test_pk_description = TestPKDescription.objects.get(id=tpd_id_mapping[row[1]])
- cpr.compound = Compound.objects.get(canonical_smile=row[0])
- cpr.tolerated = row[2] == 'Y'
- cpr.auc = row[3]
- cpr.clearance = row[4]
- cpr.c_max = row[5]
- cpr.oral_bioavailability = row[6]
- cpr.t_demi = row[7]
- cpr.t_max = row[8]
- cpr.voldistribution = row[9]
- cpr.save()
- except Exception as e:
- if options['stoponfail']:
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting {} {}'.format(row[1], row[2]))
- else:
- import traceback
- self.stderr.write(traceback.format_exc())
- self.stdout.write(
- self.style.ERROR('Failed inserting {} {}'.format(row[1], row[2])))
- else:
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted {}'.format(str(cpr))))
- if options['compound']:
- cursor.execute("""SELECT * FROM compound""")
- rows = cursor.fetchall()
- Compound.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed compound table'))
- for row in rows:
- try:
- compound = Compound()
- compound.id = row[0]
- compound.canonical_smile = row[1]
- compound.is_macrocycle = (row[4] == 'Y')
- compound.aromatic_ratio = row[5]
- compound.balaban_index = row[6]
- compound.fsp3 = row[7] # Csp3Ratio
- compound.gc_molar_refractivity = row[
- 10] # GCMolarRefractivity
- compound.log_d = row[13] # LogD
- compound.a_log_p = row[14] # ALogP
- compound.mean_atom_vol_vdw = row[15] # MeanAtomVolVdW
- compound.molecular_weight = row[16] # MolecularWeight
- compound.nb_acceptor_h = row[17] # NbAcceptorH
- compound.nb_aliphatic_amines = row[
- 18] # NbAliphaticsAmines
- compound.nb_aromatic_bonds = row[19] # NbAromaticBonds
- compound.nb_aromatic_ether = row[20] # NbAromaticsEther
- compound.nb_aromatic_sssr = row[21] # NbAromaticsSSSR
- compound.nb_atom = row[22] # NbAtom
- compound.nb_atom_non_h = row[23] # NbAtomNonH
- compound.nb_benzene_like_rings = row[24] # NbBenzLikeRings
- compound.nb_bonds = row[25] # NbBonds
- compound.nb_bonds_non_h = row[26] # NbBondsNonH
- compound.nb_br = row[27] # NbBr
- compound.nb_c = row[28] # NbC
- compound.nb_chiral_centers = row[29] # NbChiralCenters
- compound.nb_circuits = row[30] # NbCircuits
- compound.nb_cl = row[31] # NbCl
- compound.nb_csp2 = row[32] # NbCsp2
- compound.nb_csp3 = row[33] # NbCsp3
- compound.nb_donor_h = row[34] # NbDonorH
- compound.nb_double_bonds = row[35] # NbDoubleBonds
- compound.nb_f = row[36] # NbF
- compound.nb_i = row[37] # NbI
- compound.nb_multiple_bonds = row[38] # NbMultBonds
- compound.nb_n = row[39] # NbN
- compound.nb_o = row[40] # NbO
- compound.nb_rings = row[41] # NbRings
- compound.nb_rotatable_bonds = row[42] # NbRotatableBonds
- compound.randic_index = row[44] # RandicIndex
- compound.rdf070m = row[45] # RDF070m
- compound.rotatable_bond_fraction = row[
- 46] # RotatableBondFraction
- compound.sum_atom_polar = row[47] # SumAtomPolar
- compound.sum_atom_vol_vdw = row[48] # SumAtomVolVdW
- compound.tpsa = row[51] # TPSA
- compound.ui = row[52] # Ui
- compound.wiener_index = row[54] # WienerIndex
- if row[55] != 'N':
- compound.common_name = row[55] # CmpdNameSh
- compound.pubchem_id = row[56] # IdPubchem
- if row[57] != 'N':
- compound.chemspider_id = row[57] # IdPubchem
- compound.chembl_id = row[58]
- compound.iupac_name = row[59]
- compound.save(autofill=True)
- except Exception as e:
- if options['stoponfail']:
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting {} {}'.format(row[1], row[2]))
- else:
- self.stdout.write(
- self.style.ERROR('Failed inserting {} {}'.format(row[1], row[2])))
- else:
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted {} {}'.format(row[1], row[2])))
- if options['compoundbiblio']:
- cursor.execute("""select r.CmpdNameInBiblio, c.CanonicalSmile, b.IDSource from refCmpdBiblio as r inner join compound as c on r.IDCompound=c.IDCompound inner join biblio as b on r.IDBiblio=b.IDBiblio;""")
- rows = cursor.fetchall()
- RefCompoundBiblio.objects.all().delete()
- self.stdout.write(
- self.style.SUCCESS('Successfully flushed compound-biblio table'))
- for row in rows:
- try:
- c = Compound.objects.get(canonical_smile=row[1])
- b = Bibliography.objects.get(id_source=row[2])
- r = RefCompoundBiblio()
- r.compound_id = c.id
- r.bibliography_id = b.id
- r.compound_name = '-'.join(re.split('[-.]',row[0])[1:])
- # mysql format for this field is [PMID/WIPOID]-name: we remove the first part
- # sometimes there is a . instead of the dash
- # sometimes we have more than one -
- r.save()
- except Exception as e:
- if options['stoponfail']:
- import traceback
- self.stderr.write(traceback.format_exc())
- raise CommandError(
- 'Failed inserting {} {}'.format(row[1], row[2]))
- else:
- self.stdout.write(
- self.style.ERROR('Failed inserting {} {}'.format(row[1], row[2])))
- else:
- self.stdout.write(
- self.style.SUCCESS('Successfully inserted {} {}'.format(row[1], row[2])))
+ if options.get('wscache'):
+ requests_cache.install_cache('ws_cache')
+ task_option = options.get('task')
+ # map task names to task classes
+ option_to_task = {task_class.option: task_class for task_class in self.task_classes}
+ # map task names to the list of task names that depend upon them
+ dependency_to_task = {task_class.option: [] for task_class in self.task_classes}
+ for task_class in self.task_classes:
+ for task_depency_class in task_class.depends_on:
+ dependency_to_task[task_depency_class.option].append(task_class.option)
+ dependency_to_task['all'] = option_to_task.keys()
+ def append_dependencies(option):
+ dependencies = set([option])
+ for depending_option in dependency_to_task.get(option,[]):
+ dependencies.update(append_dependencies(depending_option))
+ return dependencies
+ if task_option:
+ # list of tasks to run because they depend on initial option,
+ # directly or not
+ dependencies = append_dependencies(task_option)
+ print(dependencies)
+ dependencies_classes = {task_class for opt, task_class in option_to_task.items() if opt in dependencies}
+ conn = mysql.connector.connect(
+ converter_class=MyConverter, host="localhost", user="root", password="ippidb", database="ippidb")
+ while len(dependencies_classes)>0:
+ for task_class in copy.copy(dependencies_classes):
+ if not(set(task_class.depends_on) & dependencies_classes):
+ task = task_class(self, options['errortb'], options['stoponfail'], options['progress_bar'])
+ if hasattr(task, 'set_mysql_conn'):
+ task.set_mysql_conn(conn)
+ task.run()
+ dependencies_classes.remove(task_class)
diff --git a/ippisite/ippidb/models.py b/ippisite/ippidb/models.py
index 7e6f9f2a401d03c0d8f7739eb00cdc698bce895b..5fa8acdd5362703f5fc156f3919ec608f5b424db 100644
--- a/ippisite/ippidb/models.py
+++ b/ippisite/ippidb/models.py
@@ -1205,6 +1205,9 @@ class RefCompoundBiblio(models.Model):
class Meta:
unique_together = (('compound', 'bibliography'),)
+ def __str__(self):
+ return 'Ref. {} on {}'.format(self.bibliography, self.compound)
+
class DrugBankCompound(models.Model):
"""