def render_signature(request):
# grab the selections from session data
# targets set #1
ss_pos = request.session.get('targets_pos', False)
# targets set #2
ss_neg = request.session.get('selection', False)
# setup signature
signature = SequenceSignature()
signature.setup_alignments_from_selection(ss_pos, ss_neg)
# calculate the signature
signature.calculate_signature()
# calculate the Z-scores signatures
signature.calculate_zscales_signature()
# save for later
# signature_map = feats_delta.argmax(axis=0)
request.session['signature'] = signature.prepare_session_data()
request.session.modified = True
return_html = render(request, 'sequence_signature.html',
signature.prepare_display_data())
return return_html
def render_signature(request):
# grab the selections from session data
# targets set #1
ss_pos = request.session.get('targets_pos', False)
# targets set #2
ss_neg = request.session.get('selection', False)
# setup signature
signature = SequenceSignature()
signature.setup_alignments_from_selection(ss_pos, ss_neg)
# calculate the signature
signature.calculate_signature()
# calculate the Z-scores signatures
signature.calculate_zscales_signature()
# save for later
# signature_map = feats_delta.argmax(axis=0)
request.session['signature'] = signature.prepare_session_data()
request.session.modified = True
return_html = render(
request,
'sequence_signature.html',
signature.prepare_display_data()
)
return return_html
def IMSequenceSignature(request):
'''Accept set of proteins + generic numbers and calculate the signature for those'''
t1 = time.time()
pos_set_in = get_entry_names(request)
ignore_in_alignment = get_ignore_info(request)
segments = get_protein_segments(request)
# get pos objects
pos_set = Protein.objects.filter(entry_name__in=pos_set_in).select_related(
'residue_numbering_scheme', 'species')
# Calculate Sequence Signature
signature = SequenceSignature()
signature.setup_alignments_signprot(
segments, pos_set, ignore_in_alignment=ignore_in_alignment)
signature.calculate_signature_onesided()
# preprocess data for return
signature_data = signature.prepare_display_data_onesided()
# FEATURES AND REGIONS
feats = [feature for feature in signature_data['a_pos'].features_combo]
# GET GENERIC NUMBERS
generic_numbers = get_generic_numbers(signature_data)
# FEATURE FREQUENCIES
signature_features = get_signature_features(signature_data,
generic_numbers, feats)
grouped_features = group_signature_features(signature_features)
# # FEATURE CONSENSUS
# generic_numbers_flat = list(chain.from_iterable(generic_numbers))
# sigcons = get_signature_consensus(signature_data, generic_numbers_flat)
# rec_class = pos_set[0].get_protein_class()
# dump = {
# 'rec_class': rec_class,
# 'signature': signature,
# 'consensus': signature_data,
# }
# with open('signprot/notebooks/interface_pickles/{}.p'.format(rec_class), 'wb+') as out_file:
# pickle.dump(dump, out_file)
# pass back to front
res = {
# 'cons': sigcons,
'feat_ungrouped': signature_features,
'feat': grouped_features,
}
request.session['signature'] = signature.prepare_session_data()
request.session.modified = True
t2 = time.time()
print('Runtime: {}'.format((t2 - t1) * 1000.0))
return JsonResponse(res, safe=False)
def render_alignment_excel(request):
# Grab all targets
targets = request.session.get('selection', False)
# create placeholder seq signature
signature = SequenceSignature()
signature.setup_alignments_from_selection(targets, targets)
# calculate the signture
signature.calculate_signature()
signature.calculate_zscales_signature()
outstream = BytesIO()
wb = xlsxwriter.Workbook(outstream, {'in_memory': True})
# Sequence alignment of targets
signature.prepare_excel_worksheet(wb, 'Alignment', 'positive', 'alignment')
# Residue properties stats
signature.prepare_excel_worksheet(wb, 'Property_conservation', 'positive',
'features')
# Z-scales
signature.zscales_excel(wb, "Z-scales", 'positive')
wb.close()
outstream.seek(0)
response = HttpResponse(
outstream.read(),
content_type=
"application/vnd.openxmlformats-officedocument.spreadsheetml.sheet")
response[
'Content-Disposition'] = "attachment; filename=gpcrdb_alignment.xlsx"
return response
def get_context_data (self, **kwargs):
# setup caches
cache_name = "RFB"
rfb_panel = cache.get(cache_name)
# rfb_panel = None
if rfb_panel == None:
rfb_panel = {}
# Signatures
rfb_panel["signatures"] = {}
# Grab relevant segments
segments = list(ProteinSegment.objects.filter(proteinfamily='GPCR'))
# Grab High/Low CA GPCRs (class A)
high_ca = ["5ht2c_human", "acm4_human", "drd1_human", "fpr1_human", "ghsr_human", "cnr1_human", "aa1r_human", "gpr6_human", "gpr17_human", "gpr87_human"]
low_ca = ["agtr1_human", "ednrb_human", "gnrhr_human", "acthr_human", "v2r_human", "gp141_human", "gp182_human"]
# Signature High vs Low CA
high_ca_gpcrs = Protein.objects.filter(entry_name__in=high_ca).select_related('residue_numbering_scheme', 'species')
low_ca_gpcrs = Protein.objects.filter(entry_name__in=low_ca).select_related('residue_numbering_scheme', 'species')
signature = SequenceSignature()
signature.setup_alignments(segments, high_ca_gpcrs, low_ca_gpcrs)
signature.calculate_signature()
rfb_panel["signatures"]["cah"] = signature.signature
rfb_panel["signatures"]["cah_positions"] = signature.common_gn
signature = SequenceSignature()
signature.setup_alignments(segments, low_ca_gpcrs, high_ca_gpcrs)
signature.calculate_signature()
rfb_panel["signatures"]["cal"] = signature.signature
rfb_panel["signatures"]["cal_positions"] = signature.common_gn
# Grab Gi/Gs/Gq/GI12 GPCR sets (class A)
human_class_a_gpcrs = Protein.objects.filter(species_id=1, sequence_type_id=1, family__slug__startswith='001').distinct().prefetch_related('proteingprotein_set', 'residue_numbering_scheme')
gs = list(human_class_a_gpcrs.filter(proteingprotein__slug="100_001_001"))
gio = list(human_class_a_gpcrs.filter(proteingprotein__slug="100_001_002"))
gq = list(human_class_a_gpcrs.filter(proteingprotein__slug="100_001_003"))
g12 = list(human_class_a_gpcrs.filter(proteingprotein__slug="100_001_004"))
all = set(gs + gio + gq + g12)
# Create sequence signatures for the G-protein sets
for gprotein in ["gs", "gio", "gq", "g12"]:
# print("Processing " + gprotein)
# Signature receptors specific for a G-protein vs all others
signature = SequenceSignature()
signature.setup_alignments(segments, locals()[gprotein], all.difference(locals()[gprotein]))
signature.calculate_signature()
rfb_panel["signatures"][gprotein] = signature.signature
rfb_panel["signatures"][gprotein + "_positions"] = signature.common_gn
# Add class A alignment features
signature = SequenceSignature()
signature.setup_alignments(segments, human_class_a_gpcrs, [list(human_class_a_gpcrs)[0]])
signature.calculate_signature()
rfb_panel["class_a_positions"] = signature.common_gn
rfb_panel["class_a_aa"] = signature.aln_pos.consensus
rfb_panel["class_a_prop"] = signature.features_consensus_pos
# Add X-ray ligand contacts
# Optionally include the curation with the following filter: structure_ligand_pair__annotated=True
class_a_interactions = ResidueFragmentInteraction.objects.filter(
structure_ligand_pair__structure__protein_conformation__protein__family__slug__startswith="001").exclude(interaction_type__type='hidden')\
.values("rotamer__residue__generic_number__label").annotate(unique_receptors=Count("rotamer__residue__protein_conformation__protein__family_id", distinct=True))
rfb_panel["ligand_binding"] = {entry["rotamer__residue__generic_number__label"] : entry["unique_receptors"] for entry in list(class_a_interactions)}
# Add genetic variations
all_nat_muts = NaturalMutations.objects.filter(protein__family__slug__startswith="001").values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["natural_mutations"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_nat_muts)}
# Add PTMs
all_ptms = PTMs.objects.filter(protein__family__slug__startswith="001").values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["ptms"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_ptms)}
all_phos = PTMs.objects.filter(protein__family__slug__startswith="001").filter(modification="Phosphorylation").values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["phos"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_phos)}
all_palm = PTMs.objects.filter(protein__family__slug__startswith="001").filter(modification="Palmitoylation").values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["palm"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_palm)}
all_glyc = PTMs.objects.filter(protein__family__slug__startswith="001").filter(modification__endswith="Glycosylation").values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["glyc"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_glyc)}
all_ubiq = PTMs.objects.filter(protein__family__slug__startswith="001").filter(modification="Ubiquitylation").values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["ubiq"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_ubiq)}
# Thermostabilizing
all_thermo = ConstructMutation.objects.filter(construct__protein__family__slug__startswith="001", effects__slug='thermostabilising')\
.values("residue__generic_number__label").annotate(unique_receptors=Count("construct__protein__family_id", distinct=True))
rfb_panel["thermo_mutations"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_thermo)}
# Class A ligand mutations >5 fold effect - count unique receptors
all_ligand_mutations = MutationExperiment.objects.filter(Q(foldchange__gte = 5) | Q(foldchange__lte = -5), protein__family__slug__startswith="001")\
.values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["ligand_mutations"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_ligand_mutations)}
# Class A mutations with >30% increase/decrease basal activity
all_basal_mutations = MutationExperiment.objects.filter(Q(opt_basal_activity__gte = 130) | Q(opt_basal_activity__lte = 70), protein__family__slug__startswith="001")\
.values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["basal_mutations"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_basal_mutations)}
# Intrasegment contacts
all_contacts = InteractingResiduePair.objects.filter(~Q(res1__protein_segment_id = F('res2__protein_segment_id')), referenced_structure__protein_conformation__protein__family__slug__startswith="001")\
.values("res1__generic_number__label").annotate(unique_receptors=Count("referenced_structure__protein_conformation__protein__family_id", distinct=True))
rfb_panel["intrasegment_contacts"] = {entry["res1__generic_number__label"] : entry["unique_receptors"] for entry in list(all_contacts)}
# Active/Inactive contacts
all_active_contacts = InteractingResiduePair.objects.filter(~Q(res2__generic_number__label = None), ~Q(res1__generic_number__label = None),\
referenced_structure__state__slug = "active", referenced_structure__protein_conformation__protein__family__slug__startswith="001")\
.values("res1__generic_number__label", "res2__generic_number__label")
# OPTIMIZE
active_contacts = {}
for entry in list(all_active_contacts):
if entry["res1__generic_number__label"] not in active_contacts:
active_contacts[entry["res1__generic_number__label"]] = set()
active_contacts[entry["res1__generic_number__label"]].update([entry["res2__generic_number__label"]])
rfb_panel["active_contacts"] = active_contacts
all_inactive_contacts = InteractingResiduePair.objects.filter(~Q(res2__generic_number__label = None), ~Q(res1__generic_number__label = None),\
referenced_structure__state__slug = "inactive", referenced_structure__protein_conformation__protein__family__slug__startswith="001")\
.values("res1__generic_number__label", "res2__generic_number__label")
# OPTIMIZE
inactive_contacts = {}
for entry in list(all_inactive_contacts):
if entry["res1__generic_number__label"] not in inactive_contacts:
inactive_contacts[entry["res1__generic_number__label"]] = set()
inactive_contacts[entry["res1__generic_number__label"]].update([entry["res2__generic_number__label"]])
rfb_panel["inactive_contacts"] = inactive_contacts
cache.set(cache_name, rfb_panel, 3600*24*7) # cache a week
# Other rules
# structural_rule_tree = create_structural_rule_trees(STRUCTURAL_RULES)
######## CREATE REFERENCE sets (or use structural rules)
## MICROSWITCHES
ms_labels = [residue.label for residue in ResiduePositionSet.objects.get(name="State (micro-)switches").residue_position.all()]
## SODIUM POCKET
sp_labels = [residue.label for residue in ResiduePositionSet.objects.get(name="Sodium ion pocket").residue_position.all()]
## ROTAMER SWITCHES
rotamer_labels = []
for entry in STRUCTURAL_SWITCHES["A"]:
if entry["Rotamer Switch"] != "-":
rotamer_labels.append(entry["AA1 Pos"])
rotamer_labels.append(entry["AA2 Pos"])
## G PROTEIN INTERACTION POSITIONS
# gprotein_labels = [residue.label for residue in ResiduePositionSet.objects.get(name="Signalling protein pocket").residue_position.all()]
# Class A G-protein X-ray contacts
# TODO: replace with automatically generated sets from X-rays stored in database
gprotein_labels = {"1x60": {"001_006_001_001", " 001_006_001_002"},
"12x48": {"001_001_003_008", " 001_006_001_001", " 001_006_001_002", " 001_009_001_001"},
"12x49": {"001_001_003_008", " 001_006_001_002"},
"12x51": {"001_006_001_002"},
"2x37": {"001_006_001_001"},
"2x39": {"001_002_022_003"},
"2x40": {"001_006_001_001"},
"3x49": {"001_001_003_008", " 001_002_022_003"},
"3x50": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002", " 001_009_001_001"},
"3x53": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002"},
"3x54": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002", " 001_009_001_001"},
"3x55": {"001_001_003_008", " 001_006_001_002"},
"3x56": {"001_006_001_002", " 001_009_001_001"},
"34x50": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002"},
"34x51": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002"},
"34x52": {"001_001_003_008", " 001_002_022_003", " 001_006_001_002"},
"34x53": {"001_001_003_008", " 001_006_001_002"},
"34x54": {"001_001_003_008", " 001_002_022_003", " 001_006_001_002"},
"34x55": {"001_001_003_008", " 001_002_022_003", " 001_006_001_002", " 001_009_001_001"},
"34x57": {"001_001_001_002", " 001_002_022_003"},
"4x40": {"001_002_022_003"},
"5x61": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002"},
"5x64": {"001_001_003_008", " 001_002_022_003", " 001_006_001_002"},
"5x65": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002"},
"5x67": {"001_001_003_008"},
"5x68": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002"},
"5x69": {"001_001_001_002", " 001_001_003_008", " 001_006_001_001", " 001_006_001_002"},
"5x71": {"001_001_003_008", " 001_006_001_001", " 001_006_001_002"},
"5x72": {"001_001_003_008", " 001_006_001_002", " 001_009_001_001"},
"5x74": {"001_001_003_008"},
"6x23": {"001_002_022_003"},
"6x24": {"001_009_001_001"},
"6x25": {"001_002_022_003", " 001_006_001_001", " 001_009_001_001"},
"6x26": {"001_002_022_003", " 001_009_001_001"},
"6x28": {"001_009_001_001"},
"6x29": {"001_001_001_002", " 001_006_001_001", " 001_006_001_002", " 001_009_001_001"},
"6x32": {"001_001_001_002", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002", " 001_009_001_001"},
"6x33": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002", " 001_009_001_001"},
"6x36": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_002", " 001_009_001_001"},
"6x37": {"001_001_001_002", " 001_001_003_008", " 001_006_001_001", " 001_006_001_002"},
"7x56": {"001_001_001_002", " 001_006_001_001", " 001_006_001_002", " 001_009_001_001"},
"8x47": {"001_001_001_002", " 001_002_022_003", " 001_006_001_001", " 001_009_001_001"},
"8x48": {"001_002_022_003", " 001_006_001_002", " 001_009_001_001"},
"8x49": {"001_006_001_001", " 001_006_001_002"},
"8x51": {"001_006_001_002"},
"8x56": {"001_006_001_001"}}
# TODO: replace with automatically generated sets from X-rays stored in database
# Class A Arrestin X-ray contacts
arrestin_labels = {"12x49": {"001_009_001_001"},
"2x37": {"001_009_001_001"},
"2x38": {"001_009_001_001"},
"2x39": {"001_009_001_001"},
"2x40": {"001_009_001_001"},
"2x43": {"001_009_001_001"},
"3x50": {"001_009_001_001"},
"3x54": {"001_009_001_001"},
"3x55": {"001_009_001_001"},
"3x56": {"001_009_001_001"},
"34x50": {"001_009_001_001"},
"34x51": {"001_009_001_001"},
"34x53": {"001_009_001_001"},
"34x54": {"001_009_001_001"},
"34x55": {"001_009_001_001"},
"34x56": {"001_009_001_001"},
"4x38": {"001_009_001_001"},
"5x61": {"001_009_001_001"},
"5x64": {"001_009_001_001"},
"5x68": {"001_009_001_001"},
"5x69": {"001_009_001_001"},
"5x71": {"001_009_001_001"},
"5x72": {"001_009_001_001"},
"6x24": {"001_009_001_001"},
"6x25": {"001_009_001_001"},
"6x26": {"001_009_001_001"},
"6x28": {"001_009_001_001"},
"6x29": {"001_009_001_001"},
"6x32": {"001_009_001_001"},
"6x33": {"001_009_001_001"},
"6x36": {"001_009_001_001"},
"6x37": {"001_009_001_001"},
"6x40": {"001_009_001_001"},
"8x47": {"001_009_001_001"},
"8x48": {"001_009_001_001"},
"8x49": {"001_009_001_001"},
"8x50": {"001_009_001_001"}}
# Positions in center of membrane selected using 4BVN (ADRB1) together with OPM membrane positioning
# Reference: ['1x44', '2x52', '3x36', '4x54', '5x46', '6x48', '7x43']
mid_membrane_classA = {'TM1': 44,'TM2': 52,'TM3': 36,'TM4': 54,'TM5': 46, 'TM6': 48, 'TM7': 43}
# NOTE: We might need to split this into B1 and B2 when adhesion X-rays are published
# Positions in center of membrane selected using 5XEZ (GCGR) together with OPM membrane positioning
# Reference: ['1x51', '2x58', '3x41', '4x54', '5x45', '6x49', '7x50']
mid_membrane_classB = {'TM1': 51,'TM2': 58,'TM3': 41,'TM4': 54,'TM5': 45, 'TM6': 49, 'TM7': 50}
# Positions in center of membrane selected using 4OR2 (mGLUR1) together with OPM membrane positioning
# Reference: ['1x49', '2x48', '3x40', '4x41', '5x48', '6x48', '7.39x40']
mid_membrane_classC = {'TM1': 49,'TM2': 48,'TM3': 40,'TM4': 41,'TM5': 48, 'TM6': 48, 'TM7': 40}
# Positions in center of membrane selected using 6BD4 (FZD4) together with OPM membrane positioning
# Reference: ['1x43', '2x53', '3x38', '4x53', '5x53', '6x43', '7x47']
mid_membrane_classF = {'TM1': 43,'TM2': 53,'TM3': 38,'TM4': 53,'TM5': 53, 'TM6': 43, 'TM7': 47}
# Positions within membrane layer selected using 4BVN together with OPM membrane positioning
core_membrane_classA = {'TM1': [33, 55],'TM2': [42,65],'TM3': [23,47],'TM4': [43,64],'TM5': [36,59], 'TM6': [37,60], 'TM7': [32,54]}
# TODO: other classes
core_membrane_classB = {'TM1': [33, 55],'TM2': [42,65],'TM3': [23,47],'TM4': [43,64],'TM5': [36,59], 'TM6': [37,60], 'TM7': [32,54]}
core_membrane_classC = {'TM1': [33, 55],'TM2': [42,65],'TM3': [23,47],'TM4': [43,64],'TM5': [36,59], 'TM6': [37,60], 'TM7': [32,54]}
core_membrane_classF = {'TM1': [33, 55],'TM2': [42,65],'TM3': [23,47],'TM4': [43,64],'TM5': [36,59], 'TM6': [37,60], 'TM7': [32,54]}
# Residue oriented outward of bundle (based on inactive 4BVN and active 3SN6)
outward_orientation = {
'TM1' : [29, 30, 33, 34, 36, 37, 38, 40, 41, 44, 45, 48, 51, 52, 54, 55, 58],
'TM2' : [38, 41, 45, 48, 52, 55, 56, 58, 59, 60, 62, 63, 66],
'TM3' : [23, 24, 27, 31, 48, 51, 52, 55],
'TM4' : [40, 41, 43, 44, 47, 48, 50, 51, 52, 54, 55, 58, 59, 62, 63, 81],
'TM5' : [36, 37, 38, 40, 41, 42, 44, 45, 46, 48, 49, 52, 53, 55, 56, 57, 59, 60, 62, 63, 64, 66, 67, 68, 70, 71, 73, 74],
'TM6' : [25, 28, 29, 31, 32, 34, 35, 38, 39, 42, 43, 45, 46, 49, 50, 53, 54, 56, 57, 60],
'TM7' : [33, 34, 35, 37, 40, 41, 43, 44, 48, 51, 52, 54, 55]
}
########
# prepare context for output
context = {"signatures" : []}
index = 0
for h, segment in enumerate(rfb_panel["signatures"]["gs_positions"]["gpcrdba"]):
segment_first = True
for i, position in enumerate(rfb_panel["signatures"]["gs_positions"]["gpcrdba"][segment]):
if len(position) <= 5:
# To filter segment headers with non-GN numbering
if segment_first:
context["signatures"].append({"position" : segment})
index += 1
segment_first = False
# Add data
context["signatures"].append({})
context["signatures"][index]["segment"] = segment
context["signatures"][index]["sort"] = index
context["signatures"][index]["position"] = position
# Normalized position in TM
partial_position = int(position.split('x')[1][:2])
# RESIDUE PLACEMENT
context["signatures"][index]["membane_placement"] = "-"
context["signatures"][index]["membane_segment"] = "Extracellular"
context["signatures"][index]["residue_orientation"] = "-"
if segment in mid_membrane_classA: # TM helix
# parse position
context["signatures"][index]["membane_placement"] = partial_position - mid_membrane_classA[segment]
# negative is toward cytoplasm
if segment in ['TM1', 'TM3', 'TM5', 'TM7']: # downwards
context["signatures"][index]["membane_placement"] = -1 * context["signatures"][index]["membane_placement"]
# Segment selection
if partial_position >= core_membrane_classA[segment][0] and partial_position <= core_membrane_classA[segment][1]:
context["signatures"][index]["membane_segment"] = "Membrane"
elif segment in ['TM1', 'TM3', 'TM5', 'TM7']:
if partial_position > core_membrane_classA[segment][1]:
context["signatures"][index]["membane_segment"] = "Intracellular"
else:
if partial_position < core_membrane_classA[segment][0]:
context["signatures"][index]["membane_segment"] = "Intracellular"
# Orientation
if partial_position in outward_orientation[segment]:
context["signatures"][index]["residue_orientation"] = "Outward"
else:
if partial_position > min(outward_orientation[segment]) and partial_position < max(outward_orientation[segment]):
context["signatures"][index]["residue_orientation"] = "Inward"
# Intracellular segments
elif segment in ['ICL1', 'ICL2', 'ICL3', 'TM8', 'C-term']:
context["signatures"][index]["membane_segment"] = "Intracellular"
# COUNTS: all db results in a singe loop
for key in ["ligand_binding", "natural_mutations", "thermo_mutations", "ligand_mutations", "basal_mutations", "intrasegment_contacts", "phos", "palm", "glyc", "ubiq" ]: # Add in future "gprotein_interface", "arrestin_interface"
context["signatures"][index][key] = 0
if position in rfb_panel[key]:
context["signatures"][index][key] = rfb_panel[key][position]
# G-protein interface
context["signatures"][index]["gprotein_interface"] = 0
if position in gprotein_labels:
context["signatures"][index]["gprotein_interface"] = len(gprotein_labels[position])
# Arrestin interface
context["signatures"][index]["arrestin_interface"] = 0
if position in arrestin_labels:
context["signatures"][index]["arrestin_interface"] = len(arrestin_labels[position])
# BINARY
# Microswitch
context["signatures"][index]["microswitch"] = position in ms_labels
# Sodium pocket
context["signatures"][index]["sodium"] = position in sp_labels
# Rotamer switch
context["signatures"][index]["rotamer_switch"] = position in rotamer_labels
# contacts
context["signatures"][index]["active_contacts"] = 0
if position in rfb_panel["active_contacts"]:
if position in rfb_panel["inactive_contacts"]:
context["signatures"][index]["active_contacts"] = len(rfb_panel["active_contacts"][position].difference(rfb_panel["inactive_contacts"][position]))
else:
context["signatures"][index]["active_contacts"] = len(rfb_panel["active_contacts"][position])
context["signatures"][index]["inactive_contacts"] = 0
if position in rfb_panel["inactive_contacts"]:
if position in rfb_panel["active_contacts"]:
context["signatures"][index]["inactive_contacts"] = len(rfb_panel["inactive_contacts"][position].difference(rfb_panel["active_contacts"][position]))
else:
context["signatures"][index]["inactive_contacts"] = len(rfb_panel["inactive_contacts"][position])
# CLASS A sequence + property consensus
if position in rfb_panel["class_a_positions"]["gpcrdba"][segment]:
ca_index = list(rfb_panel["class_a_positions"]["gpcrdba"][segment]).index(position)
# Sequence consensus
context["signatures"][index]["class_a_aa"] = rfb_panel["class_a_aa"][segment][position][0]
context["signatures"][index]["class_a_aa_name"] = FULL_AMINO_ACIDS[rfb_panel["class_a_aa"][segment][position][0]]
if context["signatures"][index]["class_a_aa"] == '+':
context["signatures"][index]["class_a_aa_name"] += ": "+rfb_panel["class_a_aa"][segment][position][3]
context["signatures"][index]["class_a_aa_cons"] = rfb_panel["class_a_aa"][segment][position][2]
# Property consensus
context["signatures"][index]["class_a_symb"] = rfb_panel["class_a_prop"][segment][i][0]
context["signatures"][index]["class_a_prop"] = rfb_panel["class_a_prop"][segment][i][1]
context["signatures"][index]["class_a_prop_cons"] = rfb_panel["class_a_prop"][segment][i][2]
# SEQUENCE SIGNATURES
for signature_type in ["cah", "cal", "gs", "gio", "gq", "g12"]:
if position in rfb_panel["signatures"][signature_type + "_positions"]["gpcrdba"][segment]:
ca_index = list(rfb_panel["signatures"][signature_type + "_positions"]["gpcrdba"][segment]).index(position)
context["signatures"][index][signature_type + "_score"] = rfb_panel["signatures"][signature_type][segment][ca_index][2]
context["signatures"][index][signature_type + "_prop"] = rfb_panel["signatures"][signature_type][segment][ca_index][1]
context["signatures"][index][signature_type + "_symb"] = rfb_panel["signatures"][signature_type][segment][ca_index][0]
index += 1
# Human Class A alignment - consensus/conservation
return context
def render_signature_excel(request):
# version #2 - 5 sheets with separate pieces of signature outline
# step 1 - repeat the data preparation for a sequence signature
# targets set #1
ss_pos = request.session.get('targets_pos', False)
# targets set #2
ss_neg = request.session.get('selection', False)
signature = SequenceSignature()
signature.setup_alignments_from_selection(ss_pos, ss_neg)
# calculate the signture
signature.calculate_signature()
outstream = BytesIO()
# wb = xlsxwriter.Workbook('excel_test.xlsx', {'in_memory': False})
wb = xlsxwriter.Workbook(outstream, {'in_memory': True})
# Feature stats for signature
signature.prepare_excel_worksheet(wb, 'signature_properties', 'signature',
'features')
# Feature stats for positive group alignment
signature.prepare_excel_worksheet(wb, 'protein_set1_properties',
'positive', 'features')
# Positive group alignment
signature.prepare_excel_worksheet(wb, 'protein_set1_aln', 'positive',
'alignment')
# Feature stats for negative group alignment
signature.prepare_excel_worksheet(wb, 'protein_set2_properties',
'negative', 'features')
# Negative group alignment
signature.prepare_excel_worksheet(wb, 'protein_set2_aln', 'negative',
'alignment')
signature.per_gn_signature_excel(wb)
wb.close()
outstream.seek(0)
response = HttpResponse(
outstream.read(),
content_type=
"application/vnd.openxmlformats-officedocument.spreadsheetml.sheet")
response[
'Content-Disposition'] = "attachment; filename=sequence_signature.xlsx"
return response
def get_context_data (self, **kwargs):
# setup caches
cache_name = "RFB"
rfb_panel = cache.get(cache_name)
# rfb_panel = None
if rfb_panel == None:
rfb_panel = {}
# Signatures
rfb_panel["signatures"] = {}
# Grab relevant segments
segments = list(ProteinSegment.objects.filter(proteinfamily='GPCR'))
# Grab High/Low CA GPCRs (class A)
high_ca = ["5ht2c_human", "acm4_human", "drd1_human", "fpr1_human", "ghsr_human", "cnr1_human", "aa1r_human", "gpr6_human", "gpr17_human", "gpr87_human"]
low_ca = ["agtr1_human", "ednrb_human", "gnrhr_human", "acthr_human", "v2r_human", "gp141_human", "gp182_human"]
# Signature High vs Low CA
high_ca_gpcrs = Protein.objects.filter(entry_name__in=high_ca).select_related('residue_numbering_scheme', 'species')
low_ca_gpcrs = Protein.objects.filter(entry_name__in=low_ca).select_related('residue_numbering_scheme', 'species')
signature = SequenceSignature()
signature.setup_alignments(segments, high_ca_gpcrs, low_ca_gpcrs)
signature.calculate_signature()
rfb_panel["signatures"]["cah"] = signature.signature
rfb_panel["signatures"]["cah_positions"] = signature.common_gn
signature = SequenceSignature()
signature.setup_alignments(segments, low_ca_gpcrs, high_ca_gpcrs)
signature.calculate_signature()
rfb_panel["signatures"]["cal"] = signature.signature
rfb_panel["signatures"]["cal_positions"] = signature.common_gn
# Grab Gi/Gs/Gq/GI12 GPCR sets (class A)
human_class_a_gpcrs = Protein.objects.filter(species_id=1, sequence_type_id=1, family__slug__startswith='001').distinct().prefetch_related('proteingprotein_set', 'residue_numbering_scheme')
gs = list(human_class_a_gpcrs.filter(proteingprotein__slug="100_001_001"))
gio = list(human_class_a_gpcrs.filter(proteingprotein__slug="100_001_002"))
gq = list(human_class_a_gpcrs.filter(proteingprotein__slug="100_001_003"))
g12 = list(human_class_a_gpcrs.filter(proteingprotein__slug="100_001_004"))
all = set(gs + gio + gq + g12)
# Create sequence signatures for the G-protein sets
for gprotein in ["gs", "gio", "gq", "g12"]:
# print("Processing " + gprotein)
# Signature receptors specific for a G-protein vs all others
signature = SequenceSignature()
signature.setup_alignments(segments, locals()[gprotein], all.difference(locals()[gprotein]))
signature.calculate_signature()
rfb_panel["signatures"][gprotein] = signature.signature
rfb_panel["signatures"][gprotein + "_positions"] = signature.common_gn
# Add class A alignment features
signature = SequenceSignature()
signature.setup_alignments(segments, human_class_a_gpcrs, [list(human_class_a_gpcrs)[0]])
signature.calculate_signature()
rfb_panel["class_a_positions"] = signature.common_gn
rfb_panel["class_a_aa"] = signature.aln_pos.consensus
rfb_panel["class_a_prop"] = signature.features_consensus_pos
# Add X-ray ligand contacts
# Optionally include the curation with the following filter: structure_ligand_pair__annotated=True
class_a_interactions = ResidueFragmentInteraction.objects.filter(
structure_ligand_pair__structure__protein_conformation__protein__family__slug__startswith="001").exclude(interaction_type__type='hidden')\
.values("rotamer__residue__generic_number__label").annotate(unique_receptors=Count("rotamer__residue__protein_conformation__protein__family_id", distinct=True))
rfb_panel["ligand_binding"] = {entry["rotamer__residue__generic_number__label"] : entry["unique_receptors"] for entry in list(class_a_interactions)}
# Add genetic variations
all_nat_muts = NaturalMutations.objects.filter(protein__family__slug__startswith="001").values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["natural_mutations"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_nat_muts)}
# Add PTMs
all_ptms = PTMs.objects.filter(protein__family__slug__startswith="001").values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["ptms"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_ptms)}
all_phos = PTMs.objects.filter(protein__family__slug__startswith="001").filter(modification="Phosphorylation").values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["phos"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_phos)}
all_palm = PTMs.objects.filter(protein__family__slug__startswith="001").filter(modification="Palmitoylation").values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["palm"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_palm)}
all_glyc = PTMs.objects.filter(protein__family__slug__startswith="001").filter(modification__endswith="Glycosylation").values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["glyc"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_glyc)}
all_ubiq = PTMs.objects.filter(protein__family__slug__startswith="001").filter(modification="Ubiquitylation").values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["ubiq"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_ubiq)}
# Thermostabilizing
all_thermo = ConstructMutation.objects.filter(construct__protein__family__slug__startswith="001", effects__slug='thermostabilising')\
.values("residue__generic_number__label").annotate(unique_receptors=Count("construct__protein__family_id", distinct=True))
rfb_panel["thermo_mutations"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_thermo)}
# Class A ligand mutations >5 fold effect - count unique receptors
all_ligand_mutations = MutationExperiment.objects.filter(Q(foldchange__gte = 5) | Q(foldchange__lte = -5), protein__family__slug__startswith="001")\
.values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["ligand_mutations"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_ligand_mutations)}
# Class A mutations with >30% increase/decrease basal activity
all_basal_mutations = MutationExperiment.objects.filter(Q(opt_basal_activity__gte = 130) | Q(opt_basal_activity__lte = 70), protein__family__slug__startswith="001")\
.values("residue__generic_number__label").annotate(unique_receptors=Count("protein__family_id", distinct=True))
rfb_panel["basal_mutations"] = {entry["residue__generic_number__label"] : entry["unique_receptors"] for entry in list(all_basal_mutations)}
# Intrasegment contacts
all_contacts = InteractingResiduePair.objects.filter(~Q(res1__protein_segment_id = F('res2__protein_segment_id')), referenced_structure__protein_conformation__protein__family__slug__startswith="001")\
.values("res1__generic_number__label").annotate(unique_receptors=Count("referenced_structure__protein_conformation__protein__family_id", distinct=True))
rfb_panel["intrasegment_contacts"] = {entry["res1__generic_number__label"] : entry["unique_receptors"] for entry in list(all_contacts)}
# Active/Inactive contacts
all_active_contacts = InteractingResiduePair.objects.filter(~Q(res2__generic_number__label = None), ~Q(res1__generic_number__label = None),\
referenced_structure__state__slug = "active", referenced_structure__protein_conformation__protein__family__slug__startswith="001")\
.values("res1__generic_number__label", "res2__generic_number__label")
# OPTIMIZE
active_contacts = {}
for entry in list(all_active_contacts):
if entry["res1__generic_number__label"] not in active_contacts:
active_contacts[entry["res1__generic_number__label"]] = set()
active_contacts[entry["res1__generic_number__label"]].update([entry["res2__generic_number__label"]])
rfb_panel["active_contacts"] = active_contacts
all_inactive_contacts = InteractingResiduePair.objects.filter(~Q(res2__generic_number__label = None), ~Q(res1__generic_number__label = None),\
referenced_structure__state__slug = "inactive", referenced_structure__protein_conformation__protein__family__slug__startswith="001")\
.values("res1__generic_number__label", "res2__generic_number__label")
# OPTIMIZE
inactive_contacts = {}
for entry in list(all_inactive_contacts):
if entry["res1__generic_number__label"] not in inactive_contacts:
inactive_contacts[entry["res1__generic_number__label"]] = set()
inactive_contacts[entry["res1__generic_number__label"]].update([entry["res2__generic_number__label"]])
rfb_panel["inactive_contacts"] = inactive_contacts
cache.set(cache_name, rfb_panel, 3600*24*7) # cache a week
# Other rules
# structural_rule_tree = create_structural_rule_trees(STRUCTURAL_RULES)
######## CREATE REFERENCE sets (or use structural rules)
## MICROSWITCHES
ms_labels = [residue.label for residue in ResiduePositionSet.objects.get(name="State (micro-)switches").residue_position.all()]
## SODIUM POCKET
sp_labels = [residue.label for residue in ResiduePositionSet.objects.get(name="Sodium ion pocket").residue_position.all()]
## ROTAMER SWITCHES
rotamer_labels = []
for entry in STRUCTURAL_SWITCHES["A"]:
if entry["Rotamer Switch"] != "-":
rotamer_labels.append(entry["AA1 Pos"])
rotamer_labels.append(entry["AA2 Pos"])
## G PROTEIN INTERACTION POSITIONS
# gprotein_labels = [residue.label for residue in ResiduePositionSet.objects.get(name="Signalling protein pocket").residue_position.all()]
# Class A G-protein X-ray contacts
# TODO: replace with automatically generated sets from X-rays stored in database
gprotein_labels = {"1x60": {"001_006_001_001", " 001_006_001_002"},
"12x48": {"001_001_003_008", " 001_006_001_001", " 001_006_001_002", " 001_009_001_001"},
"12x49": {"001_001_003_008", " 001_006_001_002"},
"12x51": {"001_006_001_002"},
"2x37": {"001_006_001_001"},
"2x39": {"001_002_022_003"},
"2x40": {"001_006_001_001"},
"3x49": {"001_001_003_008", " 001_002_022_003"},
"3x50": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002", " 001_009_001_001"},
"3x53": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002"},
"3x54": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002", " 001_009_001_001"},
"3x55": {"001_001_003_008", " 001_006_001_002"},
"3x56": {"001_006_001_002", " 001_009_001_001"},
"34x50": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002"},
"34x51": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002"},
"34x52": {"001_001_003_008", " 001_002_022_003", " 001_006_001_002"},
"34x53": {"001_001_003_008", " 001_006_001_002"},
"34x54": {"001_001_003_008", " 001_002_022_003", " 001_006_001_002"},
"34x55": {"001_001_003_008", " 001_002_022_003", " 001_006_001_002", " 001_009_001_001"},
"34x57": {"001_001_001_002", " 001_002_022_003"},
"4x40": {"001_002_022_003"},
"5x61": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002"},
"5x64": {"001_001_003_008", " 001_002_022_003", " 001_006_001_002"},
"5x65": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002"},
"5x67": {"001_001_003_008"},
"5x68": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002"},
"5x69": {"001_001_001_002", " 001_001_003_008", " 001_006_001_001", " 001_006_001_002"},
"5x71": {"001_001_003_008", " 001_006_001_001", " 001_006_001_002"},
"5x72": {"001_001_003_008", " 001_006_001_002", " 001_009_001_001"},
"5x74": {"001_001_003_008"},
"6x23": {"001_002_022_003"},
"6x24": {"001_009_001_001"},
"6x25": {"001_002_022_003", " 001_006_001_001", " 001_009_001_001"},
"6x26": {"001_002_022_003", " 001_009_001_001"},
"6x28": {"001_009_001_001"},
"6x29": {"001_001_001_002", " 001_006_001_001", " 001_006_001_002", " 001_009_001_001"},
"6x32": {"001_001_001_002", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002", " 001_009_001_001"},
"6x33": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_001", " 001_006_001_002", " 001_009_001_001"},
"6x36": {"001_001_001_002", " 001_001_003_008", " 001_002_022_003", " 001_006_001_002", " 001_009_001_001"},
"6x37": {"001_001_001_002", " 001_001_003_008", " 001_006_001_001", " 001_006_001_002"},
"7x56": {"001_001_001_002", " 001_006_001_001", " 001_006_001_002", " 001_009_001_001"},
"8x47": {"001_001_001_002", " 001_002_022_003", " 001_006_001_001", " 001_009_001_001"},
"8x48": {"001_002_022_003", " 001_006_001_002", " 001_009_001_001"},
"8x49": {"001_006_001_001", " 001_006_001_002"},
"8x51": {"001_006_001_002"},
"8x56": {"001_006_001_001"}}
# TODO: replace with automatically generated sets from X-rays stored in database
# Class A Arrestin X-ray contacts
arrestin_labels = {"12x49": {"001_009_001_001"},
"2x37": {"001_009_001_001"},
"2x38": {"001_009_001_001"},
"2x39": {"001_009_001_001"},
"2x40": {"001_009_001_001"},
"2x43": {"001_009_001_001"},
"3x50": {"001_009_001_001"},
"3x54": {"001_009_001_001"},
"3x55": {"001_009_001_001"},
"3x56": {"001_009_001_001"},
"34x50": {"001_009_001_001"},
"34x51": {"001_009_001_001"},
"34x53": {"001_009_001_001"},
"34x54": {"001_009_001_001"},
"34x55": {"001_009_001_001"},
"34x56": {"001_009_001_001"},
"4x38": {"001_009_001_001"},
"5x61": {"001_009_001_001"},
"5x64": {"001_009_001_001"},
"5x68": {"001_009_001_001"},
"5x69": {"001_009_001_001"},
"5x71": {"001_009_001_001"},
"5x72": {"001_009_001_001"},
"6x24": {"001_009_001_001"},
"6x25": {"001_009_001_001"},
"6x26": {"001_009_001_001"},
"6x28": {"001_009_001_001"},
"6x29": {"001_009_001_001"},
"6x32": {"001_009_001_001"},
"6x33": {"001_009_001_001"},
"6x36": {"001_009_001_001"},
"6x37": {"001_009_001_001"},
"6x40": {"001_009_001_001"},
"8x47": {"001_009_001_001"},
"8x48": {"001_009_001_001"},
"8x49": {"001_009_001_001"},
"8x50": {"001_009_001_001"}}
# Positions in center of membrane selected using 4BVN (ADRB1) together with OPM membrane positioning
# Reference: ['1x44', '2x52', '3x36', '4x54', '5x46', '6x48', '7x43']
mid_membrane_classA = {'TM1': 44,'TM2': 52,'TM3': 36,'TM4': 54,'TM5': 46, 'TM6': 48, 'TM7': 43}
# NOTE: We might need to split this into B1 and B2 when adhesion X-rays are published
# Positions in center of membrane selected using 5XEZ (GCGR) together with OPM membrane positioning
# Reference: ['1x51', '2x58', '3x41', '4x54', '5x45', '6x49', '7x50']
mid_membrane_classB = {'TM1': 51,'TM2': 58,'TM3': 41,'TM4': 54,'TM5': 45, 'TM6': 49, 'TM7': 50}
# Positions in center of membrane selected using 4OR2 (mGLUR1) together with OPM membrane positioning
# Reference: ['1x49', '2x48', '3x40', '4x41', '5x48', '6x48', '7.39x40']
mid_membrane_classC = {'TM1': 49,'TM2': 48,'TM3': 40,'TM4': 41,'TM5': 48, 'TM6': 48, 'TM7': 40}
# Positions in center of membrane selected using 6BD4 (FZD4) together with OPM membrane positioning
# Reference: ['1x43', '2x53', '3x38', '4x53', '5x53', '6x43', '7x47']
mid_membrane_classF = {'TM1': 43,'TM2': 53,'TM3': 38,'TM4': 53,'TM5': 53, 'TM6': 43, 'TM7': 47}
# Positions within membrane layer selected using 4BVN together with OPM membrane positioning
core_membrane_classA = {'TM1': [33, 55],'TM2': [42,65],'TM3': [23,47],'TM4': [43,64],'TM5': [36,59], 'TM6': [37,60], 'TM7': [32,54]}
# TODO: other classes
core_membrane_classB = {'TM1': [33, 55],'TM2': [42,65],'TM3': [23,47],'TM4': [43,64],'TM5': [36,59], 'TM6': [37,60], 'TM7': [32,54]}
core_membrane_classC = {'TM1': [33, 55],'TM2': [42,65],'TM3': [23,47],'TM4': [43,64],'TM5': [36,59], 'TM6': [37,60], 'TM7': [32,54]}
core_membrane_classF = {'TM1': [33, 55],'TM2': [42,65],'TM3': [23,47],'TM4': [43,64],'TM5': [36,59], 'TM6': [37,60], 'TM7': [32,54]}
# Residue oriented outward of bundle (based on inactive 4BVN and active 3SN6)
outward_orientation = {
'TM1' : [29, 30, 33, 34, 36, 37, 38, 40, 41, 44, 45, 48, 51, 52, 54, 55, 58],
'TM2' : [38, 41, 45, 48, 52, 55, 56, 58, 59, 60, 62, 63, 66],
'TM3' : [23, 24, 27, 31, 48, 51, 52, 55],
'TM4' : [40, 41, 43, 44, 47, 48, 50, 51, 52, 54, 55, 58, 59, 62, 63, 81],
'TM5' : [36, 37, 38, 40, 41, 42, 44, 45, 46, 48, 49, 52, 53, 55, 56, 57, 59, 60, 62, 63, 64, 66, 67, 68, 70, 71, 73, 74],
'TM6' : [25, 28, 29, 31, 32, 34, 35, 38, 39, 42, 43, 45, 46, 49, 50, 53, 54, 56, 57, 60],
'TM7' : [33, 34, 35, 37, 40, 41, 43, 44, 48, 51, 52, 54, 55]
}
########
# prepare context for output
context = {"signatures" : []}
index = 0
for h, segment in enumerate(rfb_panel["signatures"]["gs_positions"]["gpcrdba"]):
segment_first = True
for i, position in enumerate(rfb_panel["signatures"]["gs_positions"]["gpcrdba"][segment]):
if len(position) <= 5:
# To filter segment headers with non-GN numbering
if segment_first:
context["signatures"].append({"position" : segment})
index += 1
segment_first = False
# Add data
context["signatures"].append({})
context["signatures"][index]["segment"] = segment
context["signatures"][index]["sort"] = index
context["signatures"][index]["position"] = position
# Normalized position in TM
partial_position = int(position.split('x')[1][:2])
# RESIDUE PLACEMENT
context["signatures"][index]["membane_placement"] = "-"
context["signatures"][index]["membane_segment"] = "Extracellular"
context["signatures"][index]["residue_orientation"] = "-"
if segment in mid_membrane_classA: # TM helix
# parse position
context["signatures"][index]["membane_placement"] = partial_position - mid_membrane_classA[segment]
# negative is toward cytoplasm
if segment in ['TM1', 'TM3', 'TM5', 'TM7']: # downwards
context["signatures"][index]["membane_placement"] = -1 * context["signatures"][index]["membane_placement"]
# Segment selection
if partial_position >= core_membrane_classA[segment][0] and partial_position <= core_membrane_classA[segment][1]:
context["signatures"][index]["membane_segment"] = "Membrane"
elif segment in ['TM1', 'TM3', 'TM5', 'TM7']:
if partial_position > core_membrane_classA[segment][1]:
context["signatures"][index]["membane_segment"] = "Intracellular"
else:
if partial_position < core_membrane_classA[segment][0]:
context["signatures"][index]["membane_segment"] = "Intracellular"
# Orientation
if partial_position in outward_orientation[segment]:
context["signatures"][index]["residue_orientation"] = "Outward"
else:
if partial_position > min(outward_orientation[segment]) and partial_position < max(outward_orientation[segment]):
context["signatures"][index]["residue_orientation"] = "Inward"
# Intracellular segments
elif segment in ['ICL1', 'ICL2', 'ICL3', 'TM8', 'C-term']:
context["signatures"][index]["membane_segment"] = "Intracellular"
# COUNTS: all db results in a singe loop
for key in ["ligand_binding", "natural_mutations", "thermo_mutations", "ligand_mutations", "basal_mutations", "intrasegment_contacts", "phos", "palm", "glyc", "ubiq" ]: # Add in future "gprotein_interface", "arrestin_interface"
context["signatures"][index][key] = 0
if position in rfb_panel[key]:
context["signatures"][index][key] = rfb_panel[key][position]
# G-protein interface
context["signatures"][index]["gprotein_interface"] = 0
if position in gprotein_labels:
context["signatures"][index]["gprotein_interface"] = len(gprotein_labels[position])
# Arrestin interface
context["signatures"][index]["arrestin_interface"] = 0
if position in arrestin_labels:
context["signatures"][index]["arrestin_interface"] = len(arrestin_labels[position])
# BINARY
# Microswitch
context["signatures"][index]["microswitch"] = position in ms_labels
# Sodium pocket
context["signatures"][index]["sodium"] = position in sp_labels
# Rotamer switch
context["signatures"][index]["rotamer_switch"] = position in rotamer_labels
# contacts
context["signatures"][index]["active_contacts"] = 0
if position in rfb_panel["active_contacts"]:
if position in rfb_panel["inactive_contacts"]:
context["signatures"][index]["active_contacts"] = len(rfb_panel["active_contacts"][position].difference(rfb_panel["inactive_contacts"][position]))
else:
context["signatures"][index]["active_contacts"] = len(rfb_panel["active_contacts"][position])
context["signatures"][index]["inactive_contacts"] = 0
if position in rfb_panel["inactive_contacts"]:
if position in rfb_panel["active_contacts"]:
context["signatures"][index]["inactive_contacts"] = len(rfb_panel["inactive_contacts"][position].difference(rfb_panel["active_contacts"][position]))
else:
context["signatures"][index]["inactive_contacts"] = len(rfb_panel["inactive_contacts"][position])
# CLASS A sequence + property consensus
if position in rfb_panel["class_a_positions"]["gpcrdba"][segment]:
ca_index = list(rfb_panel["class_a_positions"]["gpcrdba"][segment]).index(position)
# Sequence consensus
context["signatures"][index]["class_a_aa"] = rfb_panel["class_a_aa"][segment][position][0]
context["signatures"][index]["class_a_aa_name"] = FULL_AMINO_ACIDS[rfb_panel["class_a_aa"][segment][position][0]]
if context["signatures"][index]["class_a_aa"] == '+':
context["signatures"][index]["class_a_aa_name"] += ": "+rfb_panel["class_a_aa"][segment][position][3]
context["signatures"][index]["class_a_aa_cons"] = rfb_panel["class_a_aa"][segment][position][2]
# Property consensus
context["signatures"][index]["class_a_symb"] = rfb_panel["class_a_prop"][segment][i][0]
context["signatures"][index]["class_a_prop"] = rfb_panel["class_a_prop"][segment][i][1]
context["signatures"][index]["class_a_prop_cons"] = rfb_panel["class_a_prop"][segment][i][2]
# SEQUENCE SIGNATURES
for signature_type in ["cah", "cal", "gs", "gio", "gq", "g12"]:
if position in rfb_panel["signatures"][signature_type + "_positions"]["gpcrdba"][segment]:
ca_index = list(rfb_panel["signatures"][signature_type + "_positions"]["gpcrdba"][segment]).index(position)
context["signatures"][index][signature_type + "_score"] = rfb_panel["signatures"][signature_type][segment][ca_index][2]
context["signatures"][index][signature_type + "_prop"] = rfb_panel["signatures"][signature_type][segment][ca_index][1]
context["signatures"][index][signature_type + "_symb"] = rfb_panel["signatures"][signature_type][segment][ca_index][0]
index += 1
# Human Class A alignment - consensus/conservation
return context
def render_signature_excel(request):
# version #2 - 5 sheets with separate pieces of signature outline
# step 1 - repeat the data preparation for a sequence signature
# targets set #1
ss_pos = request.session.get('targets_pos', False)
# targets set #2
ss_neg = request.session.get('selection', False)
signature = SequenceSignature()
signature.setup_alignments_from_selection(ss_pos, ss_neg)
# calculate the signture
signature.calculate_signature()
outstream = BytesIO()
# wb = xlsxwriter.Workbook('excel_test.xlsx', {'in_memory': False})
wb = xlsxwriter.Workbook(outstream, {'in_memory': True})
# Feature stats for signature
signature.prepare_excel_worksheet(
wb,
'signature_properties',
'signature',
'features'
)
# Feature stats for positive group alignment
signature.prepare_excel_worksheet(
wb,
'protein_set1_properties',
'positive',
'features'
)
# Positive group alignment
signature.prepare_excel_worksheet(
wb,
'protein_set1_aln',
'positive',
'alignment'
)
# Feature stats for negative group alignment
signature.prepare_excel_worksheet(
wb,
'protein_set2_properties',
'negative',
'features'
)
# Negative group alignment
signature.prepare_excel_worksheet(
wb,
'protein_set2_aln',
'negative',
'alignment'
)
signature.per_gn_signature_excel(wb)
wb.close()
outstream.seek(0)
response = HttpResponse(
outstream.read(),
content_type="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
)
response['Content-Disposition'] = "attachment; filename=sequence_signature.xlsx"
return response