def get_snake_plot(self):
residuelist = Residue.objects.filter(
protein_conformation__protein__entry_name=str(
self)).prefetch_related('protein_segment',
'display_generic_number',
'generic_number')
return DrawSnakePlot(residuelist, self.get_protein_class(), str(self))
def detail(request, slug):
# get constructs
c = Construct.objects.filter(name=slug).prefetch_related(
'modifications', 'mutations', 'deletions','insertions','insertions__insert_type',
'expression','solubilization','purification','crystallization','crystal').all()[0]
# get residues
residues = Residue.objects.filter(protein_conformation__protein=c.protein).order_by('sequence_number').prefetch_related(
'protein_segment', 'generic_number', 'display_generic_number')
residues_lookup = {}
for r in residues:
residues_lookup[r.sequence_number] = r
schematics = c.schematic()
snake = cache.get(c.name+'_snake')
if snake==None:
print(c.name+'_snake no cache')
snake = cache.set(c.name+'_snake', DrawSnakePlot(residues,c.protein.get_protein_class(),str(c.protein),nobuttons = True), 60*60*24*2) #two days
snake = cache.get(c.name+'_snake')
else:
print(c.name+'_snake used cache')
chunk_size = 10
context = {'c':c, 'chunk_size': chunk_size, 'snake': snake, 'annotations': json.dumps(schematics['annotations']), 'schematics': schematics, 'residues_lookup': residues_lookup}
return render(request,'construct/construct_detail.html',context)
def snake(self):
## Use cache if possible
temp = self.snakecache
if temp==None:
# print(self.name+'_snake no cache')
residues = Residue.objects.filter(protein_conformation__protein=self.protein).order_by('sequence_number').prefetch_related(
'protein_segment', 'generic_number', 'display_generic_number')
temp = DrawSnakePlot(residues,self.protein.get_protein_class(),str(self.protein),nobuttons = True)
self.snakecache = pickle.dumps(temp)
self.save()
else:
temp = pickle.loads(temp)
return temp
def detail(request, slug):
# get family
pf = ProteinFamily.objects.get(slug=slug)
# get family list
ppf = pf
families = [ppf.name]
while ppf.parent.parent:
families.append(ppf.parent.name)
ppf = ppf.parent
families.reverse()
# number of proteins
proteins = Protein.objects.filter(family__slug__startswith=pf.slug,
sequence_type__slug='wt')
no_of_proteins = proteins.count()
no_of_human_proteins = Protein.objects.filter(
family__slug__startswith=pf.slug,
species__id=1,
sequence_type__slug='wt').count()
# get structures of this family
structures = Structure.objects.filter(
protein_conformation__protein__parent__family__slug__startswith=slug
).order_by('-representative',
'resolution').prefetch_related('pdb_code__web_resource')
mutations = MutationExperiment.objects.filter(protein__in=proteins)
# fetch proteins and segments
proteins = Protein.objects.filter(family__slug__startswith=slug,
sequence_type__slug='wt',
species__id=1)
segments = ProteinSegment.objects.filter(partial=False)
# create an alignment object
a = Alignment()
# load data into the alignment
a.load_proteins(proteins)
a.load_segments(segments)
# build the alignment data matrix
a.build_alignment()
# calculate consensus sequence + amino acid and feature frequency
a.calculate_statistics()
HelixBox = DrawHelixBox(a.full_consensus, 'Class A', str('test'))
SnakePlot = DrawSnakePlot(a.full_consensus, 'Class A', str('test'))
try:
pc = ProteinConformation.objects.get(
protein__family__slug=slug,
protein__sequence_type__slug='consensus')
except ProteinConformation.DoesNotExist:
pc = ProteinConformation.objects.get(protein__family__slug=slug,
protein__species_id=1,
protein__sequence_type__slug='wt')
residues = Residue.objects.filter(
protein_conformation=pc).order_by('sequence_number').prefetch_related(
'protein_segment', 'generic_number', 'display_generic_number')
# process residues and return them in chunks of 10
# this is done for easier scaling on smaller screens
chunk_size = 10
r_chunks = []
r_buffer = []
last_segment = False
border = False
title_cell_skip = 0
for i, r in enumerate(residues):
# title of segment to be written out for the first residue in each segment
segment_title = False
# keep track of last residues segment (for marking borders)
if r.protein_segment.slug != last_segment:
last_segment = r.protein_segment.slug
border = True
# if on a border, is there room to write out the title? If not, write title in next chunk
if i == 0 or (border and len(last_segment) <=
(chunk_size - i % chunk_size)):
segment_title = True
border = False
title_cell_skip += len(
last_segment
) # skip cells following title (which has colspan > 1)
if i and i % chunk_size == 0:
r_chunks.append(r_buffer)
r_buffer = []
r_buffer.append((r, segment_title, title_cell_skip))
# update cell skip counter
if title_cell_skip > 0:
title_cell_skip -= 1
if r_buffer:
r_chunks.append(r_buffer)
context = {
'pf': pf,
'families': families,
'structures': structures,
'no_of_proteins': no_of_proteins,
'no_of_human_proteins': no_of_human_proteins,
'a': a,
'HelixBox': HelixBox,
'SnakePlot': SnakePlot,
'mutations': mutations,
'r_chunks': r_chunks,
'chunk_size': chunk_size
}
return render(request, 'family/family_detail.html', context)
def render_variants(request,
protein=None,
family=None,
download=None,
receptor_class=None,
gn=None,
aa=None,
**response_kwargs):
simple_selection = request.session.get('selection', False)
proteins = []
if protein: # if protein static page
proteins.append(Protein.objects.get(entry_name=protein.lower()))
target_type = 'protein'
# flatten the selection into individual proteins
if simple_selection:
for target in simple_selection.targets:
if target.type == 'protein':
proteins.append(target.item)
elif target.type == 'family':
target_type = 'family'
familyname = target.item
# species filter
species_list = []
for species in simple_selection.species:
species_list.append(species.item)
# annotation filter
protein_source_list = []
for protein_source in simple_selection.annotation:
protein_source_list.append(protein_source.item)
if species_list:
family_proteins = Protein.objects.filter(
family__slug__startswith=target.item.slug,
species__in=(species_list),
source__in=(protein_source_list)).select_related(
'residue_numbering_scheme', 'species')
else:
family_proteins = Protein.objects.filter(
family__slug__startswith=target.item.slug,
source__in=(protein_source_list)).select_related(
'residue_numbering_scheme', 'species')
for fp in family_proteins:
proteins.append(fp)
NMs = NaturalMutations.objects.filter(
Q(protein__in=proteins)).prefetch_related(
'residue__generic_number', 'residue__display_generic_number',
'residue__protein_segment', 'protein')
ptms = PTMs.objects.filter(
Q(protein__in=proteins)).prefetch_related('residue')
ptms_dict = {}
## MICROSWITCHES
micro_switches_rset = ResiduePositionSet.objects.get(name="Microswitches")
ms_label = []
for residue in micro_switches_rset.residue_position.all():
ms_label.append(residue.label)
ms_object = Residue.objects.filter(
protein_conformation__protein=proteins[0],
generic_number__label__in=ms_label)
ms_sequence_numbers = []
for ms in ms_object:
ms_sequence_numbers.append(ms.sequence_number)
## SODIUM POCKET
sodium_pocket_rset = ResiduePositionSet.objects.get(name="Sodium pocket")
sp_label = []
for residue in sodium_pocket_rset.residue_position.all():
sp_label.append(residue.label)
sp_object = Residue.objects.filter(
protein_conformation__protein=proteins[0],
generic_number__label__in=ms_label)
sp_sequence_numbers = []
for sp in sp_object:
sp_sequence_numbers.append(sp.sequence_number)
for ptm in ptms:
ptms_dict[ptm.residue.sequence_number] = ptm.modification
## G PROTEIN INTERACTION POSITIONS
# THIS SHOULD BE CLASS SPECIFIC (different set)
rset = ResiduePositionSet.objects.get(name='Signalling protein pocket')
gprotein_generic_set = []
for residue in rset.residue_position.all():
gprotein_generic_set.append(residue.label)
### GET LB INTERACTION DATA
# get also ortholog proteins, which might have been crystallised to extract
# interaction data also from those
if protein:
orthologs = Protein.objects.filter(
family__slug=proteins[0].family.slug, sequence_type__slug='wt')
else:
orthologs = Protein.objects.filter(
family__slug__startswith=proteins[0].family.slug,
sequence_type__slug='wt')
interactions = ResidueFragmentInteraction.objects.filter(
structure_ligand_pair__structure__protein_conformation__protein__parent__in
=orthologs,
structure_ligand_pair__annotated=True).exclude(
interaction_type__type='hidden').all()
interaction_data = {}
for interaction in interactions:
if interaction.rotamer.residue.generic_number:
sequence_number = interaction.rotamer.residue.sequence_number
# sequence_number = lookup[interaction.rotamer.residue.generic_number.label]
label = interaction.rotamer.residue.generic_number.label
aa = interaction.rotamer.residue.amino_acid
interactiontype = interaction.interaction_type.name
if sequence_number not in interaction_data:
interaction_data[sequence_number] = []
if interactiontype not in interaction_data[sequence_number]:
interaction_data[sequence_number].append(interactiontype)
if target_type == 'family':
pc = ProteinConformation.objects.get(
protein__family__name=familyname,
protein__sequence_type__slug='consensus')
residuelist = Residue.objects.filter(protein_conformation=pc).order_by(
'sequence_number').prefetch_related('protein_segment',
'generic_number',
'display_generic_number')
else:
residuelist = Residue.objects.filter(
protein_conformation__protein=proteins[0]).prefetch_related(
'protein_segment', 'display_generic_number', 'generic_number')
jsondata = {}
for NM in NMs:
functional_annotation = ''
SN = NM.residue.sequence_number
if NM.residue.generic_number:
GN = NM.residue.generic_number.label
else:
GN = ''
if SN in sp_sequence_numbers:
functional_annotation += 'SodiumPocket '
if SN in ms_sequence_numbers:
functional_annotation += 'MicroSwitch '
if SN in ptms_dict:
functional_annotation += 'PTM (' + ptms_dict[SN] + ') '
if SN in interaction_data:
functional_annotation += 'LB (' + ', '.join(
interaction_data[SN]) + ') '
if GN in gprotein_generic_set:
functional_annotation += 'GP (contact) '
ms_type = NM.type
if ms_type == 'missense':
effect = 'deleterious' if NM.sift_score <= 0.05 or NM.polyphen_score >= 0.1 else 'tolerated'
color = '#e30e0e' if NM.sift_score <= 0.05 or NM.polyphen_score >= 0.1 else '#70c070'
else:
effect = 'deleterious'
color = '#575c9d'
# account for multiple mutations at this position!
NM.functional_annotation = functional_annotation
# print(NM.functional_annotation)
jsondata[SN] = [
NM.amino_acid, NM.allele_frequency, NM.allele_count,
NM.allele_number, NM.number_homozygotes, NM.type, effect, color,
functional_annotation
]
natural_mutation_list = {}
max_snp_pos = 1
for NM in NMs:
if NM.residue.generic_number:
if NM.residue.generic_number.label in natural_mutation_list:
natural_mutation_list[
NM.residue.generic_number.label]['val'] += 1
if not str(NM.amino_acid) in natural_mutation_list[
NM.residue.generic_number.label]['AA']:
natural_mutation_list[NM.residue.generic_number.label][
'AA'] = natural_mutation_list[
NM.residue.generic_number.label]['AA'] + str(
NM.amino_acid) + ' '
if natural_mutation_list[
NM.residue.generic_number.label]['val'] > max_snp_pos:
max_snp_pos = natural_mutation_list[
NM.residue.generic_number.label]['val']
else:
natural_mutation_list[NM.residue.generic_number.label] = {
'val': 1,
'AA': NM.amino_acid + ' '
}
jsondata_natural_mutations = {}
for r in residuelist:
if r.generic_number:
if r.generic_number.label in natural_mutation_list:
jsondata_natural_mutations[
r.sequence_number] = natural_mutation_list[
r.generic_number.label]
jsondata_natural_mutations['color'] = linear_gradient(start_hex="#c79494",
finish_hex="#c40100",
n=max_snp_pos)
# jsondata_cancer_mutations['color'] = linear_gradient(start_hex="#d8baff", finish_hex="#422d65", n=max_cancer_pos)
# jsondata_disease_mutations['color'] = linear_gradient(start_hex="#ffa1b1", finish_hex="#6e000b", n=max_disease_pos)
#
SnakePlot = DrawSnakePlot(residuelist, "Class A", protein, nobuttons=1)
HelixBox = DrawHelixBox(residuelist, 'Class A', protein, nobuttons=1)
# EXCEL TABLE EXPORT
if download:
data = []
for r in NMs:
values = r.__dict__
data.append(values)
headers = [
'type', 'amino_acid', 'allele_count', 'allele_number',
'allele_frequency', 'polyphen_score', 'sift_score',
'number_homozygotes', 'functional_annotation'
]
# EXCEL SOLUTION
output = BytesIO()
workbook = xlsxwriter.Workbook(output)
worksheet = workbook.add_worksheet()
col = 0
for h in headers:
worksheet.write(0, col, h)
col += 1
row = 1
for d in data:
col = 0
for h in headers:
worksheet.write(row, col, str(d[h]))
col += 1
row += 1
workbook.close()
output.seek(0)
xlsx_data = output.read()
response = HttpResponse(
xlsx_data,
content_type=
'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet'
)
response[
'Content-Disposition'] = 'attachment; filename=GPCRdb_' + proteins[
0].entry_name + '_variant_data.xlsx' # % 'mutations'
return response
return render(
request, 'browser.html', {
'mutations': NMs,
'type': target_type,
'HelixBox': HelixBox,
'SnakePlot': SnakePlot,
'receptor': str(proteins[0].entry_name),
'mutations_pos_list': json.dumps(jsondata),
'natural_mutations_pos_list':
json.dumps(jsondata_natural_mutations)
})
def Ginterface(request, protein=None):
residuelist = Residue.objects.filter(
protein_conformation__protein__entry_name=protein).prefetch_related(
'protein_segment', 'display_generic_number', 'generic_number')
SnakePlot = DrawSnakePlot(residuelist,
"Class A (Rhodopsin)",
protein,
nobuttons=1)
# TEST
gprotein_residues = Residue.objects.filter(
protein_conformation__protein__entry_name='gnaz_human'
).prefetch_related('protein_segment', 'display_generic_number',
'generic_number')
gproteinplot = DrawGproteinPlot(gprotein_residues, "Gprotein", protein)
crystal = Structure.objects.get(pdb_code__index="3SN6")
aa_names = definitions.AMINO_ACID_GROUP_NAMES
names_aa = dict(zip(aa_names.values(), aa_names.keys()))
names_aa['Polar (S/T)'] = 'pol_short'
names_aa['Polar (N/Q/H)'] = 'pol_long'
residues_browser = [{
'pos': 135,
'aa': 'I',
'gprotseg': "H5",
'segment': 'TM3',
'ligand': 'Gs',
'type': aa_names['hp'],
'gpcrdb': '3.54x54',
'gpnum': 'G.H5.16',
'gpaa': 'Q384',
'availability': 'interacting'
}, {
'pos': 136,
'aa': 'T',
'gprotseg': "H5",
'segment': 'TM3',
'ligand': 'Gs',
'type': 'Polar (S/T)',
'gpcrdb': '3.55x55',
'gpnum': 'G.H5.12',
'gpaa': 'R380',
'availability': 'interacting'
}, {
'pos': 139,
'aa': 'F',
'gprotseg': "H5",
'segment': 'ICL2',
'ligand': 'Gs',
'type': 'Aromatic',
'gpcrdb': '34.51x51',
'gpnum': 'G.H5.8',
'gpaa': 'F376',
'availability': 'interacting'
}, {
'pos': 139,
'aa': 'F',
'gprotseg': "S1",
'segment': 'ICL2',
'ligand': 'Gs',
'type': 'Aromatic',
'gpcrdb': '34.51x51',
'gpnum': 'G.S1.2',
'gpaa': 'H41',
'availability': 'interacting'
}, {
'pos': 141,
'aa': 'Y',
'gprotseg': "H5",
'segment': 'ICL2',
'ligand': 'Gs',
'type': 'Aromatic',
'gpcrdb': '34.53x53',
'gpnum': 'G.H5.19',
'gpaa': 'H387',
'availability': 'interacting'
}, {
'pos': 225,
'aa': 'E',
'gprotseg': "H5",
'segment': 'TM5',
'ligand': 'Gs',
'type': 'Negative charge',
'gpcrdb': '5.64x64',
'gpnum': 'G.H5.12',
'gpaa': 'R380',
'availability': 'interacting'
}, {
'pos': 225,
'aa': 'E',
'gprotseg': "H5",
'segment': 'TM5',
'ligand': 'Gs',
'type': 'Negative charge',
'gpcrdb': '5.64x64',
'gpnum': 'G.H5.16',
'gpaa': 'Q384',
'availability': 'interacting'
}, {
'pos': 229,
'aa': 'Q',
'gprotseg': "H5",
'segment': 'TM5',
'ligand': 'Gs',
'type': 'Polar (N/Q/H)',
'gpcrdb': '5.68x68',
'gpnum': 'G.H5.13',
'gpaa': 'D381',
'availability': 'interacting'
}, {
'pos': 229,
'aa': 'Q',
'gprotseg': "H5",
'segment': 'TM5',
'ligand': 'Gs',
'type': 'Polar (N/Q/H)',
'gpcrdb': '5.68x68',
'gpnum': 'G.H5.16',
'gpaa': 'Q384',
'availability': 'interacting'
}, {
'pos': 229,
'aa': 'Q',
'gprotseg': "H5",
'segment': 'TM5',
'ligand': 'Gs',
'type': 'Polar (N/Q/H)',
'gpcrdb': '5.68x68',
'gpnum': 'G.H5.17',
'gpaa': 'R385',
'availability': 'interacting'
}, {
'pos': 274,
'aa': 'T',
'gprotseg': "H5",
'segment': 'TM6',
'ligand': 'Gs',
'type': 'Polar (S/T)',
'gpcrdb': '6.36x36',
'gpnum': 'G.H5.24',
'gpaa': 'E392',
'availability': 'interacting'
}, {
'pos': 328,
'aa': 'R',
'gprotseg': "H5",
'segment': 'TM7',
'ligand': 'Gs',
'type': 'Positive charge',
'gpcrdb': '7.55x55',
'gpnum': 'G.H5.24',
'gpaa': 'E392',
'availability': 'interacting'
}, {
'pos': 232,
'aa': 'K',
'segment': 'TM5',
'ligand': 'Gs',
'type': 'Positive charge',
'gpcrdb': '5.71x71',
'gprotseg': "H5",
'gpnum': 'G.H5.13',
'gpaa': 'D381',
'availability': 'interacting'
}]
# accessible_gn = ['3.50x50', '3.53x53', '3.54x54', '3.55x55', '34.50x50', '34.51x51', '34.53x53', '34.54x54', '5.61x61', '5.64x64', '5.65x65', '5.67x67', '5.68x68', '5.71x71', '5.72x72', '5.74x74', '5.75x75', '6.29x29', '6.32x32', '6.33x33', '6.36x36', '6.37x37', '7.55x55', '8.48x48', '8.49x49']
accessible_gn = [
'3.50x50', '3.53x53', '3.54x54', '3.55x55', '3.56x56', '34.50x50',
'34.51x51', '34.52x52', '34.53x53', '34.54x54', '34.55x55', '34.56x56',
'34.57x57', '5.61x61', '5.64x64', '5.65x65', '5.66x66', '5.67x67',
'5.68x68', '5.69x69', '5.71x71', '5.72x72', '5.74x74', '5.75x75',
'6.25x25', '6.26x26', '6.28x28', '6.29x29', '6.32x32', '6.33x33',
'6.36x36', '6.37x37', '6.40x40', '7.55x55', '7.56x56', '8.47x47',
'8.48x48', '8.49x49', '8.51x51'
]
exchange_table = OrderedDict([
('hp', ('V', 'I', 'L', 'M')),
('ar', ('F', 'H', 'W', 'Y')),
('pol_short', ('S', 'T')), # Short/hydroxy
('pol_long', ('N', 'Q', 'H')), # Amino-like (both donor and acceptor
('neg', ('D', 'E')),
('pos', ('K', 'R'))
])
interacting_gn = []
accessible_pos = list(
residuelist.filter(
display_generic_number__label__in=accessible_gn).values_list(
'sequence_number', flat=True))
# Which of the Gs interacting_pos are conserved?
GS_none_equivalent_interacting_pos = []
GS_none_equivalent_interacting_gn = []
for interaction in residues_browser:
interacting_gn.append(interaction['gpcrdb'])
gs_b2_interaction_type_long = (next(
(item['type'] for item in residues_browser
if item['gpcrdb'] == interaction['gpcrdb']), None))
interacting_aa = residuelist.filter(
display_generic_number__label__in=[interaction['gpcrdb']
]).values_list('amino_acid',
flat=True)
if interacting_aa:
interaction['aa'] = interacting_aa[0]
pos = residuelist.filter(
display_generic_number__label__in=[interaction['gpcrdb']
]).values_list(
'sequence_number',
flat=True)[0]
interaction['pos'] = pos
feature = names_aa[gs_b2_interaction_type_long]
if interacting_aa[0] not in exchange_table[feature]:
GS_none_equivalent_interacting_pos.append(pos)
GS_none_equivalent_interacting_gn.append(interaction['gpcrdb'])
GS_equivalent_interacting_pos = list(
residuelist.filter(
display_generic_number__label__in=interacting_gn).values_list(
'sequence_number', flat=True))
gProteinData = ProteinGProteinPair.objects.filter(
protein__entry_name=protein)
primary = []
secondary = []
for entry in gProteinData:
if entry.transduction == 'primary':
primary.append((entry.g_protein.name.replace(
"Gs", "G<sub>s</sub>").replace("Gi", "G<sub>i</sub>").replace(
"Go",
"G<sub>o</sub>").replace("G11", "G<sub>11</sub>").replace(
"G12", "G<sub>12</sub>").replace(
"G13", "G<sub>13</sub>").replace(
"Gq",
"G<sub>q</sub>").replace("G", "Gα"),
entry.g_protein.slug))
elif entry.transduction == 'secondary':
secondary.append((entry.g_protein.name.replace(
"Gs", "G<sub>s</sub>").replace("Gi", "G<sub>i</sub>").replace(
"Go",
"G<sub>o</sub>").replace("G11", "G<sub>11</sub>").replace(
"G12", "G<sub>12</sub>").replace(
"G13", "G<sub>13</sub>").replace(
"Gq",
"G<sub>q</sub>").replace("G", "Gα"),
entry.g_protein.slug))
return render(
request, 'signprot/ginterface.html', {
'pdbname': '3SN6',
'snakeplot': SnakePlot,
'gproteinplot': gproteinplot,
'crystal': crystal,
'interacting_equivalent': GS_equivalent_interacting_pos,
'interacting_none_equivalent': GS_none_equivalent_interacting_pos,
'accessible': accessible_pos,
'residues': residues_browser,
'mapped_protein': protein,
'interacting_gn': GS_none_equivalent_interacting_gn,
'primary_Gprotein': set(primary),
'secondary_Gprotein': set(secondary)
})
def detail(request, slug):
# get family
pf = ProteinFamily.objects.get(slug=slug)
# get family list
ppf = pf
families = [ppf.name]
while ppf.parent.parent:
families.append(ppf.parent.name)
ppf = ppf.parent
families.reverse()
# number of proteins
proteins = Protein.objects.filter(family__slug__startswith=pf.slug,
sequence_type__slug='wt')
no_of_proteins = proteins.count()
no_of_human_proteins = Protein.objects.filter(
family__slug__startswith=pf.slug,
species__id=1,
sequence_type__slug='wt').count()
list_proteins = list(proteins.values_list('pk', flat=True))
# get structures of this family
structures = Structure.objects.filter(
protein_conformation__protein__parent__family__slug__startswith=slug
).order_by('-representative',
'resolution').prefetch_related('pdb_code__web_resource')
mutations = MutationExperiment.objects.filter(
protein__in=proteins).prefetch_related('residue__generic_number',
'exp_qual', 'ligand')
interactions = ResidueFragmentInteraction.objects.filter(
structure_ligand_pair__structure__protein_conformation__protein__parent__in
=proteins,
structure_ligand_pair__annotated=True).exclude(
interaction_type__type='hidden').order_by(
'rotamer__residue__sequence_number')
interaction_list = {}
for interaction in interactions:
if interaction.rotamer.residue.generic_number:
gn = interaction.rotamer.residue.generic_number.label
aa = interaction.rotamer.residue.amino_acid
interactiontype = interaction.interaction_type.name
if gn not in interaction_list:
interaction_list[gn] = []
interaction_list[gn].append([aa, interactiontype])
## Variants
NMs = NaturalMutations.objects.filter(
protein__in=proteins).prefetch_related('residue__generic_number')
natural_mutation_list = {}
max_snp_pos = 1
for NM in NMs:
if NM.residue.generic_number:
if NM.residue.generic_number.label in natural_mutation_list:
natural_mutation_list[
NM.residue.generic_number.label]['val'] += 1
if not str(NM.amino_acid) in natural_mutation_list[
NM.residue.generic_number.label]['AA']:
natural_mutation_list[NM.residue.generic_number.label][
'AA'] = natural_mutation_list[
NM.residue.generic_number.label]['AA'] + str(
NM.amino_acid) + ' '
if natural_mutation_list[
NM.residue.generic_number.label]['val'] > max_snp_pos:
max_snp_pos = natural_mutation_list[
NM.residue.generic_number.label]['val']
else:
natural_mutation_list[NM.residue.generic_number.label] = {
'val': 1,
'AA': NM.amino_acid + ' '
}
## PTMs
ptms = PTMs.objects.filter(
protein__in=proteins).prefetch_related('residue__generic_number')
ptm_list = {}
for ptm in ptms:
if ptm.residue.generic_number:
if ptm.residue.generic_number.label in ptm_list:
ptm_list[ptm.residue.generic_number.label]['val'] += 1
if not str(ptm.modification) in ptm_list[
ptm.residue.generic_number.label]['mod']:
ptm_list[ptm.residue.generic_number.label][
'mod'] = ptm_list[ptm.residue.generic_number.
label]['mod'] + ', ' + str(
ptm.modification)
else:
ptm_list[ptm.residue.generic_number.label] = {
'val': 1,
'mod': ptm.modification
}
# CMs = CancerMutations.objects.filter(
# protein__in=proteins).prefetch_related('residue__generic_number')
#
# cancer_mutation_list = {}
# max_cancer_pos = 1
# for CM in CMs:
# if CM.residue.generic_number:
# if CM.residue.generic_number.label in cancer_mutation_list:
# cancer_mutation_list[CM.residue.generic_number.label]['val'] += 1
# if not str(CM.amino_acid) in cancer_mutation_list[CM.residue.generic_number.label]['AA']:
# cancer_mutation_list[CM.residue.generic_number.label]['AA'] = cancer_mutation_list[CM.residue.generic_number.label]['AA'] + str(CM.amino_acid)
#
# if cancer_mutation_list[CM.residue.generic_number.label]['val'] > max_cancer_pos:
# max_cancer_pos = cancer_mutation_list[CM.residue.generic_number.label]['val']
# else:
# cancer_mutation_list[CM.residue.generic_number.label] = {'val':1, 'AA': CM.amino_acid}
#
# DMs = DiseaseMutations.objects.filter(
# protein__in=proteins).prefetch_related('residue__generic_number')
# disease_mutation_list = {}
# max_disease_pos = 1
# for DM in DMs:
# if DM.residue.generic_number:
# if DM.residue.generic_number.label in disease_mutation_list:
# disease_mutation_list[DM.residue.generic_number.label]['val'] += 1
# if not str(DM.amino_acid) in disease_mutation_list[DM.residue.generic_number.label]['AA']:
# disease_mutation_list[DM.residue.generic_number.label]['AA'] = disease_mutation_list[DM.residue.generic_number.label]['AA'] + str(DM.amino_acid)
#
# if disease_mutation_list[DM.residue.generic_number.label]['val'] > max_cancer_pos:
# max_cancer_pos = disease_mutation_list[DM.residue.generic_number.label]['val']
# else:
# disease_mutation_list[DM.residue.generic_number.label] = {'val':1, 'AA': DM.amino_acid}
mutations_list = {}
for mutation in mutations:
if not mutation.residue.generic_number:
continue #cant map those without display numbers
if mutation.residue.generic_number.label not in mutations_list:
mutations_list[mutation.residue.generic_number.label] = []
if mutation.ligand:
ligand = mutation.ligand.name
else:
ligand = ''
if mutation.exp_qual:
qual = mutation.exp_qual.qual
else:
qual = ''
mutations_list[mutation.residue.generic_number.label].append(
[mutation.foldchange,
ligand.replace("'", "\\'"), qual])
try:
pc = ProteinConformation.objects.get(
protein__family__slug=slug,
protein__sequence_type__slug='consensus')
except ProteinConformation.DoesNotExist:
pc = ProteinConformation.objects.get(protein__family__slug=slug,
protein__species_id=1,
protein__sequence_type__slug='wt')
residues = Residue.objects.filter(
protein_conformation=pc).order_by('sequence_number').prefetch_related(
'protein_segment', 'generic_number', 'display_generic_number')
jsondata = {}
jsondata_interaction = {}
jsondata_natural_mutations = {}
jsondata_ptms = {}
# jsondata_cancer_mutations = {}
# jsondata_disease_mutations = {}
for r in residues:
if r.generic_number:
if r.generic_number.label in mutations_list:
jsondata[r.sequence_number] = [
mutations_list[r.generic_number.label]
]
if r.generic_number.label in interaction_list:
jsondata_interaction[r.sequence_number] = interaction_list[
r.generic_number.label]
if r.generic_number.label in natural_mutation_list:
jsondata_natural_mutations[
r.sequence_number] = natural_mutation_list[
r.generic_number.label]
if r.generic_number.label in ptm_list:
jsondata_ptms[r.sequence_number] = ptm_list[
r.generic_number.label]
# if r.generic_number.label in cancer_mutation_list:
# jsondata_cancer_mutations[r.sequence_number] = cancer_mutation_list[r.generic_number.label]
# if r.generic_number.label in disease_mutation_list:
# jsondata_disease_mutations[r.sequence_number] = disease_mutation_list[r.generic_number.label]
jsondata_natural_mutations['color'] = linear_gradient(start_hex="#c79494",
finish_hex="#c40100",
n=max_snp_pos)
# jsondata_cancer_mutations['color'] = linear_gradient(start_hex="#d8baff", finish_hex="#422d65", n=max_cancer_pos)
# jsondata_disease_mutations['color'] = linear_gradient(start_hex="#ffa1b1", finish_hex="#6e000b", n=max_disease_pos)
HelixBox = DrawHelixBox(residues, 'Class A',
'family_diagram_preloaded_data')
SnakePlot = DrawSnakePlot(residues, 'Class A',
'family_diagram_preloaded_data')
# process residues and return them in chunks of 10
# this is done for easier scaling on smaller screens
chunk_size = 10
r_chunks = []
r_buffer = []
last_segment = False
border = False
title_cell_skip = 0
for i, r in enumerate(residues):
# title of segment to be written out for the first residue in each segment
segment_title = False
# keep track of last residues segment (for marking borders)
if r.protein_segment.slug != last_segment:
last_segment = r.protein_segment.slug
border = True
# if on a border, is there room to write out the title? If not, write title in next chunk
if i == 0 or (border and len(last_segment) <=
(chunk_size - i % chunk_size)):
segment_title = True
border = False
title_cell_skip += len(
last_segment
) # skip cells following title (which has colspan > 1)
if i and i % chunk_size == 0:
r_chunks.append(r_buffer)
r_buffer = []
r_buffer.append((r, segment_title, title_cell_skip))
# update cell skip counter
if title_cell_skip > 0:
title_cell_skip -= 1
if r_buffer:
r_chunks.append(r_buffer)
context = {
'pf': pf,
'families': families,
'structures': structures,
'no_of_proteins': no_of_proteins,
'no_of_human_proteins': no_of_human_proteins,
'HelixBox': HelixBox,
'SnakePlot': SnakePlot,
'mutations': mutations,
'r_chunks': r_chunks,
'chunk_size': chunk_size,
'mutations_pos_list': json.dumps(jsondata),
'interaction_pos_list': json.dumps(jsondata_interaction),
'natural_mutations_pos_list': json.dumps(jsondata_natural_mutations),
'ptms_pos_list': json.dumps(jsondata_ptms)
} # ,'cancer_mutations_pos_list': json.dumps(jsondata_cancer_mutations), 'disease_mutations_pos_list': json.dumps(jsondata_disease_mutations)
return render(request, 'family/family_detail.html', context)
def detail(request, slug):
# get family
pf = ProteinFamily.objects.get(slug=slug)
# get family list
ppf = pf
families = [ppf.name]
while ppf.parent.parent:
families.append(ppf.parent.name)
ppf = ppf.parent
families.reverse()
# number of proteins
proteins = Protein.objects.filter(family__slug__startswith=pf.slug,
sequence_type__slug='wt')
no_of_proteins = proteins.count()
no_of_human_proteins = Protein.objects.filter(
family__slug__startswith=pf.slug,
species__id=1,
sequence_type__slug='wt').count()
list_proteins = list(proteins.values_list('pk', flat=True))
# get structures of this family
structures = Structure.objects.filter(
protein_conformation__protein__parent__family__slug__startswith=slug
).order_by('-representative',
'resolution').prefetch_related('pdb_code__web_resource')
mutations = MutationExperiment.objects.filter(
protein__in=proteins).prefetch_related('residue__generic_number',
'exp_qual', 'ligand')
interactions = ResidueFragmentInteraction.objects.filter(
structure_ligand_pair__structure__protein_conformation__protein__parent__in
=proteins,
structure_ligand_pair__annotated=True).exclude(
interaction_type__type='hidden').order_by(
'rotamer__residue__sequence_number')
interaction_list = {}
for interaction in interactions:
if interaction.rotamer.residue.generic_number:
gn = interaction.rotamer.residue.generic_number.label
aa = interaction.rotamer.residue.amino_acid
interactiontype = interaction.interaction_type.name
if gn not in interaction_list:
interaction_list[gn] = []
interaction_list[gn].append([aa, interactiontype])
mutations_list = {}
for mutation in mutations:
if not mutation.residue.generic_number:
continue #cant map those without display numbers
if mutation.residue.generic_number.label not in mutations_list:
mutations_list[mutation.residue.generic_number.label] = []
if mutation.ligand:
ligand = mutation.ligand.name
else:
ligand = ''
if mutation.exp_qual:
qual = mutation.exp_qual.qual
else:
qual = ''
mutations_list[mutation.residue.generic_number.label].append(
[mutation.foldchange,
ligand.replace("'", "\\'"), qual])
try:
pc = ProteinConformation.objects.get(
protein__family__slug=slug,
protein__sequence_type__slug='consensus')
except ProteinConformation.DoesNotExist:
pc = ProteinConformation.objects.get(protein__family__slug=slug,
protein__species_id=1,
protein__sequence_type__slug='wt')
residues = Residue.objects.filter(
protein_conformation=pc).order_by('sequence_number').prefetch_related(
'protein_segment', 'generic_number', 'display_generic_number')
jsondata = {}
jsondata_interaction = {}
for r in residues:
if r.generic_number:
if r.generic_number.label in mutations_list:
jsondata[r.sequence_number] = [
mutations_list[r.generic_number.label]
]
if r.generic_number.label in interaction_list:
jsondata_interaction[r.sequence_number] = interaction_list[
r.generic_number.label]
HelixBox = DrawHelixBox(residues, 'Class A',
'family_diagram_preloaded_data')
SnakePlot = DrawSnakePlot(
residues, 'Class A',
'family_diagram_preloaded_data') ## was str(list_proteins)
# process residues and return them in chunks of 10
# this is done for easier scaling on smaller screens
chunk_size = 10
r_chunks = []
r_buffer = []
last_segment = False
border = False
title_cell_skip = 0
for i, r in enumerate(residues):
# title of segment to be written out for the first residue in each segment
segment_title = False
# keep track of last residues segment (for marking borders)
if r.protein_segment.slug != last_segment:
last_segment = r.protein_segment.slug
border = True
# if on a border, is there room to write out the title? If not, write title in next chunk
if i == 0 or (border and len(last_segment) <=
(chunk_size - i % chunk_size)):
segment_title = True
border = False
title_cell_skip += len(
last_segment
) # skip cells following title (which has colspan > 1)
if i and i % chunk_size == 0:
r_chunks.append(r_buffer)
r_buffer = []
r_buffer.append((r, segment_title, title_cell_skip))
# update cell skip counter
if title_cell_skip > 0:
title_cell_skip -= 1
if r_buffer:
r_chunks.append(r_buffer)
context = {
'pf': pf,
'families': families,
'structures': structures,
'no_of_proteins': no_of_proteins,
'no_of_human_proteins': no_of_human_proteins,
'HelixBox': HelixBox,
'SnakePlot': SnakePlot,
'mutations': mutations,
'r_chunks': r_chunks,
'chunk_size': chunk_size,
'mutations_pos_list': json.dumps(jsondata),
'interaction_pos_list': json.dumps(jsondata_interaction)
}
return render(request, 'family/family_detail.html', context)