def retro(request, smiles=None, chiral=True, mincount=0, max_n=200):
'''
Retrosynthesis homepage
'''
context = {}
# Set default inputs
context['form'] = {}
context['form']['template_prioritization'] = request.session.get(
'template_prioritization', 'Relevance')
context['form']['precursor_prioritization'] = request.session.get(
'precursor_prioritization', 'RelevanceHeuristic')
context['form']['template_count'] = request.session.get(
'template_count', '100')
context['form']['max_cum_prob'] = request.session.get(
'max_cum_prob', '0.995')
context['form']['filter_threshold'] = request.session.get(
'filter_threshold', '0.75')
print(request)
if request.method == 'POST':
print(request)
smiles = str(request.POST['smiles']) # not great...
context['form']['template_prioritization'] = str(
request.POST['template_prioritization'])
request.session['template_prioritization'] = context['form'][
'template_prioritization']
context['form']['precursor_prioritization'] = str(
request.POST['precursor_prioritization'])
request.session['precursor_prioritization'] = context['form'][
'precursor_prioritization']
if 'template_count' in request.POST:
context['form']['template_count'] = str(
request.POST['template_count'])
request.session['template_count'] = context['form'][
'template_count']
if 'max_cum_prob' in request.POST:
context['form']['max_cum_prob'] = str(request.POST['max_cum_prob'])
request.session['max_cum_prob'] = context['form']['max_cum_prob']
if 'filter_threshold' in request.POST:
context['form']['filter_threshold'] = str(
request.POST['filter_threshold'])
request.session['filter_threshold'] = context['form'][
'filter_threshold']
smiles = resolve_smiles(smiles)
if smiles is None:
context['err'] = 'Could not parse!'
return render(request, 'retro.html', context)
if smiles is not None and not is_banned(request, smiles):
# OLD: ALWAYS CHIRAL NOW
# if 'retro_lit' in request.POST: return redirect('retro_lit_target', smiles=smiles)
# if 'retro' in request.POST:
# return retro_target(request, smiles, chiral=False)
# if 'retro_chiral' in request.POST:
# return retro_target(request, smiles, chiral=True)
# Look up target
smiles_img = reverse('draw_smiles', kwargs={'smiles': smiles})
context['target'] = {'smiles': smiles, 'img': smiles_img}
# Perform retrosynthesis
context['form']['smiles'] = smiles
template_prioritization = context['form']['template_prioritization']
precursor_prioritization = context['form']['precursor_prioritization']
filter_threshold = context['form']['filter_threshold']
try:
template_count = int(context['form']['template_count'])
if (template_count < 1):
raise ValueError
except ValueError:
context['err'] = 'Invalid template count specified!'
return render(request, 'retro.html', context)
try:
max_cum_prob = float(context['form']['max_cum_prob'])
if (max_cum_prob <= 0):
raise ValueError
except ValueError:
context[
'err'] = 'Invalid maximum cumulative probability specified!'
return render(request, 'retro.html', context)
try:
filter_threshold = float(context['form']['filter_threshold'])
filter_threshold = min(1, filter_threshold)
filter_threshold = max(0, filter_threshold)
apply_fast_filter = filter_threshold > 0
except ValueError:
context['err'] = 'Invalid filter threshold specified!'
return render(request, 'retro.html', context)
if template_prioritization == 'Popularity':
template_count = 1e9
print('Retro expansion conditions:')
print(smiles)
print(template_prioritization)
print(precursor_prioritization)
print(template_count)
print(max_cum_prob)
print(filter_threshold)
startTime = time.time()
if chiral:
res = get_top_precursors_c.delay(
smiles,
template_prioritization,
precursor_prioritization,
mincount=0,
max_branching=max_n,
template_count=template_count,
max_cum_prob=max_cum_prob,
apply_fast_filter=apply_fast_filter,
filter_threshold=filter_threshold)
(smiles, precursors) = res.get(300)
# allow up to 5 minutes...can be pretty slow
context['precursors'] = precursors
context['footnote'] = RETRO_CHIRAL_FOOTNOTE
else:
# Use apply_async so we can force high priority
res = get_top_precursors.delay(smiles,
template_prioritization,
precursor_prioritization,
mincount=0,
max_branching=max_n,
template_count=template_count,
max_cum_prob=max_cum_prob,
apply_fast_filter=apply_fast_filter,
filter_threshold=filter_threshold)
context['precursors'] = res.get(120)
context['footnote'] = RETRO_FOOTNOTE
context['time'] = '%0.3f' % (time.time() - startTime)
# Change 'tform' field to be reaction SMARTS, not ObjectID from Mongo
# Also add up total number of examples
for (i, precursor) in enumerate(context['precursors']):
context['precursors'][i]['tforms'] = \
[dict(RetroTransformer.lookup_id(ObjectId(_id)), **
{'id': str(_id)}) for _id in precursor['tforms']]
context['precursors'][i]['mols'] = []
# Overwrite num examples
context['precursors'][i]['num_examples'] = sum(
tform['count'] for tform in precursor['tforms'])
for smiles in precursor['smiles_split']:
ppg = price_smiles_func(smiles)
context['precursors'][i]['mols'].append({
'smiles':
smiles,
'ppg':
'${}/g'.format(ppg) if ppg else 'cannot buy',
})
elif smiles is not None:
context[
'err'] = 'ASKCOS does not provide results for compounds on restricted lists such as the CWC and DEA schedules'
else:
# Define suggestions
context['suggestions'] = [
{
'name': 'Diphenhydramine',
'smiles': 'CN(C)CCOC(c1ccccc1)c2ccccc2'
},
{
'name': 'Fluconazole',
'smiles': 'OC(Cn1cncn1)(Cn2cncn2)c3ccc(F)cc3F'
},
{
'name': 'Nevirapine',
'smiles': 'Cc1ccnc2N(C3CC3)c4ncccc4C(=O)Nc12'
},
{
'name': 'Atropine',
'smiles': 'CN1C2CCC1CC(C2)OC(=O)C(CO)c3ccccc3'
},
{
'name': 'Diazepam',
'smiles': 'CN1C(=O)CN=C(c2ccccc2)c3cc(Cl)ccc13'
},
]
context['footnote'] = RETRO_CHIRAL_FOOTNOTE
return render(request, 'retro.html', context)
def retro(request, smiles=None, chiral=True, mincount=0, max_n=200):
'''
Retrosynthesis homepage
'''
context = {}
# Set default inputs
context['form'] = {}
context['form']['template_prioritization'] = request.session.get(
'template_prioritization', 'Relevance')
context['form']['precursor_prioritization'] = request.session.get(
'precursor_prioritization', 'RelevanceHeuristic')
context['form']['template_count'] = request.session.get(
'template_count', '100')
context['form']['max_cum_prob'] = request.session.get(
'max_cum_prob', '0.995')
context['form']['filter_threshold'] = request.session.get(
'filter_threshold', '0.75')
print(request)
if request.method == 'POST':
print(request)
smiles = str(request.POST['smiles']) # not great...
context['form']['template_prioritization'] = str(
request.POST['template_prioritization'])
request.session['template_prioritization'] = context['form'][
'template_prioritization']
context['form']['precursor_prioritization'] = str(
request.POST['precursor_prioritization'])
request.session['precursor_prioritization'] = context['form'][
'precursor_prioritization']
if 'template_count' in request.POST:
context['form']['template_count'] = str(
request.POST['template_count'])
request.session['template_count'] = context['form'][
'template_count']
if 'max_cum_prob' in request.POST:
context['form']['max_cum_prob'] = str(request.POST['max_cum_prob'])
request.session['max_cum_prob'] = context['form']['max_cum_prob']
if 'filter_threshold' in request.POST:
context['form']['filter_threshold'] = str(
request.POST['filter_threshold'])
request.session['filter_threshold'] = context['form'][
'filter_threshold']
smiles = resolve_smiles(smiles)
if smiles is None:
context['err'] = 'Could not parse!'
return render(request, 'retro.html', context)
if smiles is not None and not is_banned(request, smiles):
# OLD: ALWAYS CHIRAL NOW
# if 'retro_lit' in request.POST: return redirect('retro_lit_target', smiles=smiles)
# if 'retro' in request.POST:
# return retro_target(request, smiles, chiral=False)
# if 'retro_chiral' in request.POST:
# return retro_target(request, smiles, chiral=True)
# Look up target
smiles_img = reverse('draw_smiles', kwargs={'smiles': smiles})
context['target'] = {'smiles': smiles, 'img': smiles_img}
# Perform retrosynthesis
context['form']['smiles'] = smiles
template_prioritization = context['form']['template_prioritization']
precursor_prioritization = context['form']['precursor_prioritization']
filter_threshold = context['form']['filter_threshold']
try:
template_count = int(context['form']['template_count'])
if (template_count < 1):
raise ValueError
except ValueError:
context['err'] = 'Invalid template count specified!'
return render(request, 'retro.html', context)
try:
max_cum_prob = float(context['form']['max_cum_prob'])
if (max_cum_prob <= 0):
raise ValueError
except ValueError:
context[
'err'] = 'Invalid maximum cumulative probability specified!'
return render(request, 'retro.html', context)
try:
filter_threshold = float(context['form']['filter_threshold'])
filter_threshold = min(1, filter_threshold)
filter_threshold = max(0, filter_threshold)
apply_fast_filter = filter_threshold > 0
except ValueError:
context['err'] = 'Invalid filter threshold specified!'
return render(request, 'retro.html', context)
if template_prioritization == 'Popularity':
template_count = 1e9
print('Retro expansion conditions:')
print(smiles)
print(template_prioritization)
print(precursor_prioritization)
print(template_count)
print(max_cum_prob)
print(filter_threshold)
startTime = time.time()
if chiral:
if max_cum_prob > 0.999 and template_count > 1000:
res = get_top_precursors_p.delay(
smiles,
max_num_templates=template_count,
max_cum_prob=max_cum_prob,
fast_filter_threshold=filter_threshold)
else:
res = get_top_precursors_c.delay(
smiles,
max_num_templates=template_count,
max_cum_prob=max_cum_prob,
fast_filter_threshold=filter_threshold)
(smiles, precursors) = res.get(300)
# allow up to 5 minutes...can be pretty slow
context['precursors'] = precursors
context['footnote'] = RETRO_CHIRAL_FOOTNOTE
else:
# Use apply_async so we can force high priority
res = get_top_precursors.delay(smiles,
template_prioritization,
precursor_prioritization,
mincount=0,
max_branching=max_n,
template_count=template_count,
max_cum_prob=max_cum_prob,
apply_fast_filter=apply_fast_filter,
filter_threshold=filter_threshold)
context['precursors'] = res.get(120)
context['footnote'] = ''
context['time'] = '%0.3f' % (time.time() - startTime)
# Change 'tform' field to be reaction SMARTS, not ObjectID from Mongo
# Also add up total number of examples
for (i, precursor) in enumerate(context['precursors']):
context['precursors'][i]['tforms'] = \
[dict(RetroTransformer.lookup_id(_id), **{'id': str(_id)}) for _id in precursor['tforms']]
context['precursors'][i]['mols'] = []
# Overwrite num examples
context['precursors'][i]['num_examples'] = sum(
tform['count'] for tform in precursor['tforms'])
for smiles in precursor['smiles_split']:
ppg = Pricer.lookup_smiles(smiles)
context['precursors'][i]['mols'].append({
'smiles':
smiles,
'ppg':
'${}/g'.format(ppg) if ppg else 'cannot buy',
})
elif smiles is not None:
context[
'err'] = 'ASKCOS does not provide results for compounds on restricted lists such as the CWC and DEA schedules'
else:
# Define suggestions
context['suggestions'] = [
{
'name': 'Diphenhydramine',
'smiles': 'CN(C)CCOC(c1ccccc1)c2ccccc2'
},
{
'name': 'Fluconazole',
'smiles': 'OC(Cn1cncn1)(Cn2cncn2)c3ccc(F)cc3F'
},
{
'name': 'Nevirapine',
'smiles': 'Cc1ccnc2N(C3CC3)c4ncccc4C(=O)Nc12'
},
{
'name': 'Atropine',
'smiles': 'CN1C2CCC1CC(C2)OC(=O)C(CO)c3ccccc3'
},
{
'name': 'Diazepam',
'smiles': 'CN1C(=O)CN=C(c2ccccc2)c3cc(Cl)ccc13'
},
]
hidden = [
{
'name': 'Hydroxychloroquine',
'smiles': 'CCN(CCO)CCCC(C)Nc1ccnc2cc(Cl)ccc12'
},
{
'name': 'Ibuprofen',
'smiles': 'CC(C)Cc1ccc(cc1)C(C)C(O)=O'
},
{
'name': 'Tramadol',
'smiles': 'CN(C)C[C@H]1CCCC[C@@]1(C2=CC(=CC=C2)OC)O'
},
{
'name': 'Lamivudine',
'smiles': 'NC1=NC(=O)N(C=C1)[C@@H]2CS[C@H](CO)O2'
},
{
'name': 'Pregabalin',
'smiles': 'CC(C)C[C@H](CN)CC(O)=O'
},
{
'name': 'Naproxen',
'smiles': 'COc1ccc2cc([C@H](C)C(=O)O)ccc2c1'
},
{
'name':
'Imatinib',
'smiles':
'CN1CCN(CC1)Cc2ccc(cc2)C(=O)Nc3ccc(C)c(Nc4nccc(n4)c5cccnc5)c3'
},
{
'name':
'Quinapril',
'smiles':
'CCOC(=O)[C@H](CCc1ccccc1)N[C@@H](C)C(=O)N2Cc3ccccc3C[C@H]2C(O)=O'
},
{
'name':
'Atorvastatin',
'smiles':
'CC(C)c1n(CC[C@@H](O)C[C@@H](O)CC(O)=O)c(c2ccc(F)cc2)c(c3ccccc3)c1C(=O)Nc4ccccc4'
},
{
'name':
'Bortezomib',
'smiles':
'CC(C)C[C@@H](NC(=O)[C@@H](Cc1ccccc1)NC(=O)c2cnccn2)B(O)O'
},
{
'name':
'Itraconazole',
'smiles':
'CCC(C)N1N=CN(C1=O)c2ccc(cc2)N3CCN(CC3)c4ccc(OC[C@H]5CO[C@@](Cn6cncn6)(O5)c7ccc(Cl)cc7Cl)cc4'
},
{
'name':
'6-Carboxytetramethylrhodamine',
'smiles':
'CN(C)C1=CC2=C(C=C1)C(=C3C=CC(=[N+](C)C)C=C3O2)C4=C(C=CC(=C4)C(=O)[O-])C(=O)O'
},
{
'name':
'6-Carboxytetramethylrhodamine isomer',
'smiles':
'CN(C)c1ccc2c(c1)Oc1cc(N(C)C)ccc1C21OC(=O)c2ccc(C(=O)O)c1c2'
},
{
'name': '(S)-Warfarin',
'smiles': 'CC(=O)C[C@@H](C1=CC=CC=C1)C2=C(C3=CC=CC=C3OC2=O)O'
},
{
'name': 'Tranexamic Acid',
'smiles': 'NC[C@@H]1CC[C@H](CC1)C(O)=O'
},
]
context['footnote'] = RETRO_CHIRAL_FOOTNOTE
return render(request, 'retro.html', context)