def Prepare_file(self, request,build=False):
self.Tree = PrepareTree(build)
a=Alignment()
sets = ProteinSet.objects.all()
#### Get additional data ####
crysts=[]
for n in sets:
if n.id==1:
for prot in n.proteins.all():
crysts.append(prot.entry_name)
#############################
# get the user selection from session
if build != False:
################################## FOR BUILDING STATISTICS ONLY##########################
build_proteins=[]
if build == '001':
cons_prots = []
for prot in Protein.objects.filter(sequence_type__slug='consensus', species_id=1):
if prot.family.slug.startswith('001') and len(prot.family.slug.split('_'))==3:
build_proteins.append(prot)
for set in sets:
if set.id==1:
for prot in set.proteins.all():
if prot.family.slug.startswith('001_') and prot.species.latin_name=='H**o sapiens':
build_proteins.append(prot)
else:
for prot in Protein.objects.filter(sequence_type__slug='wt', species_id=1):
if prot.family.slug.startswith(build):
build_proteins.append(prot)
a.load_proteins(build_proteins)
segments = ProteinSegment.objects.all()
a.load_segments(segments)
self.bootstrap,self.UPGMA,self.branches,self.ttype=[0,1,1,0]
##################################################################
else:
simple_selection=request.session.get('selection', False)
a.load_proteins_from_selection(simple_selection)
a.load_segments_from_selection(simple_selection)
self.bootstrap,self.UPGMA,self.branches,self.ttype = map(int,simple_selection.tree_settings)
if self.bootstrap!=0:
self.bootstrap=pow(10,self.bootstrap)
#### Create an alignment object
a.build_alignment()
a.calculate_statistics()
a.calculate_similarity()
self.total = len(a.proteins)
total_length = 0
for chain in a.proteins[0].alignment:
total_length += len(a.proteins[0].alignment[chain])
families = ProteinFamily.objects.all()
self.famdict = {}
for n in families:
self.famdict[self.Tree.trans_0_2_A(n.slug)]=n.name
dirname = unique_filename = uuid.uuid4()
os.mkdir('/tmp/%s' %dirname)
infile = open('/tmp/%s/infile' %dirname,'w')
infile.write(' '+str(self.total)+' '+str(total_length)+'\n')
if len(a.proteins) < 3:
return 'More_prots',None, None, None, None,None,None,None,None
####Get additional protein information
accesions = {}
for n in a.proteins:
fam = self.Tree.trans_0_2_A(n.protein.family.slug)
if n.protein.sequence_type.slug == 'consensus':
fam+='_CON'
entry_name = n.protein.entry_name
name = n.protein.name.replace('<sub>','').replace('</sub>','').replace('<i>','').replace('</i>','')
if '&' in name and ';' in name:
name = name.replace('&','').replace(';',' ')
acc = n.protein.accession
if acc:
acc = acc.replace('-','_')
else:
acc = link.replace('-','_')[:6]
spec = str(n.protein.species)
fam += '_'+n.protein.species.common_name.replace(' ','_').upper()
desc = name
if entry_name in crysts:
if not fam in self.Additional_info['crystal']['proteins']:
self.Additional_info['crystal']['proteins'].append(fam)
if len(name)>25:
name=name[:25]+'...'
self.family[entry_name] = {'name':name,'family':fam,'description':desc,'species':spec,'class':'','accession':acc,'ligand':'','type':'','link': entry_name}
accesions[acc]=entry_name
####Write PHYLIP input
sequence = ''
for chain in n.alignment:
for residue in n.alignment[chain]:
sequence += residue[2].replace('_','-')
infile.write(acc+' '*9+sequence+'\n')
infile.close()
####Run bootstrap
if self.bootstrap:
### Write phylip input options
inp = open('/tmp/%s/temp' %dirname,'w')
inp.write('\n'.join(['r',str(self.bootstrap),'y','77','y'])+'\n')
inp.close()
###
try:
subprocess.check_output(['phylip seqboot<temp'], shell=True, cwd = '/tmp/%s' %dirname, timeout=300)
os.rename('/tmp/%s/outfile' %dirname, '/tmp/%s/infile' %dirname)
except:
kill_phylo() #FIXME, needs better way of handling this!
return "too big","too big","too big","too big","too big","too big","too big","too big","too big"
### Write phylip input options
inp = open('/tmp/%s/temp' %dirname,'w')
if self.bootstrap:
inp.write('\n'.join(['m','d',str(self.bootstrap),'y'])+'\n')
else:
inp.write('y\n')
inp.close()
###
try:
subprocess.check_output(['phylip protdist<temp>>log'], shell=True, cwd = '/tmp/%s' %dirname, timeout=300)
except:
kill_phylo() #FIXME, needs better way of handling this!
return "too big","too big","too big","too big","too big","too big","too big","too big","too big"
os.rename('/tmp/%s/infile' %dirname, '/tmp/%s/dupa' %dirname)
os.rename('/tmp/%s/outfile' %dirname, '/tmp/%s/infile' %dirname)
inp = open('/tmp/%s/temp' %dirname,'w')
if self.bootstrap:
### Write phylip input options
if self.UPGMA:
inp.write('\n'.join(['N','m',str(self.bootstrap),'111','y'])+'\n')
else:
inp.write('\n'.join(['m',str(self.bootstrap),'111','y'])+'\n')
else:
if self.UPGMA:
inp.write('N\ny\n')
else:
inp.write('y\n')
inp.close()
###
try:
subprocess.check_output(['phylip neighbor<temp'], shell=True, cwd = '/tmp/%s' %dirname, timeout=300)
except:
kill_phylo() #FIXME, needs better way of handling this!
return "too big","too big","too big","too big","too big","too big","too big","too big"
if self.bootstrap:
os.rename('/tmp/%s/outfile' %dirname, '/tmp/%s/infile' %dirname)
os.rename('/tmp/%s/outtree' %dirname, '/tmp/%s/intree' %dirname)
### Write phylip input options
inp = open('/tmp/%s/temp' %dirname,'w')
inp.write('y\n')
inp.close()
###
try:
subprocess.check_output(['phylip consense<temp'], shell=True, cwd = '/tmp/%s' %dirname, timeout=300)
except:
kill_phylo() #FIXME, needs better way of handling this!
return "too big","too big","too big","too big","too big","too big","too big","too big"
self.phylip = open('/tmp/%s/outtree' %dirname).read()
for acc in accesions.keys():
self.phylip=self.phylip.replace(acc,accesions[acc])
# self.phylogeny_output = self.phylip
self.outtree = open('/tmp/%s/outfile' %dirname).read().lstrip()
phylogeny_input = self.get_phylogeny('/tmp/%s/' %dirname)
shutil.rmtree('/tmp/%s' %dirname)
if build != False:
open('static/home/images/'+build+'_legend.svg','w').write(str(self.Tree.legend))
open('static/home/images/'+build+'_tree.xml','w').write(phylogeny_input)
else:
return phylogeny_input, self.branches, self.ttype, self.total, str(self.Tree.legend), self.Tree.box, self.Additional_info, self.buttons, a.proteins
class Treeclass:
family = {}
def __init__(self):
self.Additional_info={"crystal": {"include":"False", "order":6, "colours":{"crystal_true":"#6dcde1","crystal_false":"#EEE"}, "color_type":"single", "proteins":[], "parent":None, "child": None, "name":"Crystals"},
"class": {"include":"True", "order":0, "colours":{}, "proteins":[], "color_type":"grayscale", "parent":[], "child": ["family,ligand"], "name":"Class"},
"family": {"include":"True", "order":1, "colours":{}, "proteins":[], "color_type":"spectrum", "parent":[], "child": ["ligand"], "name":"Ligand type"},
"ligand": {"include":"True", "order":2, "colours":{}, "proteins":[], "color_type":"spectrum", "parent":["family","class"], "child": [], "name":"Receptor type"},
"mutant": {"include":"False", "order":3, "colours":{"mutant_true":"#6dcde1","mutant_false":"#EEE"}, "color_type":"single", "proteins":[], "parent":[], "child": ["mutant_plus","mutant_minus"], "name":"Mutated proteins"},
"mutant_plus": {"include":"False", "order":4, "colours":{"mutant_plus_true":"#6dcde1","mutant_plus_false":"#EEE"}, "color_type":"single", "proteins":[], "parent":"mutant", "child": [], "name":"Positive affinity mutants"},
"mutant_minus": {"include":"False", "order":5, "colours":{"mutant_minus_true":"#6dcde1","mutant_minus_false":"#EEE"}, "color_type":"single", "proteins":[], "parent":"mutant", "child": [], "name":"Negative affinity mutants"}
}
self.buttons = [(x[1]['order'],x[1]['name']) for x in sorted(self.Additional_info.items(), key= lambda x: x[1]['order']) if x[1]['include']=='True']
self.family = {}
self.phylip = None
self.outtree = None
self.dir = ''
def Prepare_file(self, request,build=False):
self.Tree = PrepareTree(build)
a=Alignment()
sets = ProteinSet.objects.all()
#### Get additional data ####
crysts=[]
for n in sets:
if n.id==1:
for prot in n.proteins.all():
crysts.append(prot.entry_name)
#############################
# get the user selection from session
if build != False:
################################## FOR BUILDING STATISTICS ONLY##########################
build_proteins=[]
if build == '001':
cons_prots = []
for prot in Protein.objects.filter(sequence_type__slug='consensus', species_id=1):
if prot.family.slug.startswith('001') and len(prot.family.slug.split('_'))==3:
build_proteins.append(prot)
for set in sets:
if set.id==1:
for prot in set.proteins.all():
if prot.family.slug.startswith('001_') and prot.species.latin_name=='H**o sapiens':
build_proteins.append(prot)
else:
for prot in Protein.objects.filter(sequence_type__slug='wt', species_id=1):
if prot.family.slug.startswith(build):
build_proteins.append(prot)
a.load_proteins(build_proteins)
segments = ProteinSegment.objects.all()
a.load_segments(segments)
self.bootstrap,self.UPGMA,self.branches,self.ttype=[0,1,1,0]
##################################################################
else:
simple_selection=request.session.get('selection', False)
a.load_proteins_from_selection(simple_selection)
a.load_segments_from_selection(simple_selection)
self.bootstrap,self.UPGMA,self.branches,self.ttype = map(int,simple_selection.tree_settings)
if self.bootstrap!=0:
self.bootstrap=pow(10,self.bootstrap)
#### Create an alignment object
a.build_alignment()
a.calculate_statistics()
a.calculate_similarity()
self.total = len(a.proteins)
total_length = 0
for chain in a.proteins[0].alignment:
total_length += len(a.proteins[0].alignment[chain])
families = ProteinFamily.objects.all()
self.famdict = {}
for n in families:
self.famdict[self.Tree.trans_0_2_A(n.slug)]=n.name
dirname = unique_filename = uuid.uuid4()
os.mkdir('/tmp/%s' %dirname)
infile = open('/tmp/%s/infile' %dirname,'w')
infile.write(' '+str(self.total)+' '+str(total_length)+'\n')
if len(a.proteins) < 3:
return 'More_prots',None, None, None, None,None,None,None
####Get additional protein information
accesions = {}
for n in a.proteins:
fam = self.Tree.trans_0_2_A(n.protein.family.slug)
if n.protein.sequence_type.slug == 'consensus':
fam+='_CON'
entry_name = n.protein.entry_name
name = n.protein.name.replace('<sub>','').replace('</sub>','').replace('<i>','').replace('</i>','')
if '&' in name and ';' in name:
name = name.replace('&','').replace(';',' ')
acc = n.protein.accession
if acc:
acc = acc.replace('-','_')
else:
acc = link.replace('-','_')[:6]
spec = str(n.protein.species)
fam += '_'+n.protein.species.common_name.replace(' ','_').upper()
desc = name
if entry_name in crysts:
if not fam in self.Additional_info['crystal']['proteins']:
self.Additional_info['crystal']['proteins'].append(fam)
if len(name)>25:
name=name[:25]+'...'
self.family[entry_name] = {'name':name,'family':fam,'description':desc,'species':spec,'class':'','accession':acc,'ligand':'','type':'','link': entry_name}
accesions[acc]=entry_name
####Write PHYLIP input
sequence = ''
for chain in n.alignment:
for residue in n.alignment[chain]:
sequence += residue[2].replace('_','-')
infile.write(acc+' '*9+sequence+'\n')
infile.close()
####Run bootstrap
if self.bootstrap:
### Write phylip input options
inp = open('/tmp/%s/temp' %dirname,'w')
inp.write('\n'.join(['r',str(self.bootstrap),'y','77','y'])+'\n')
inp.close()
###
try:
subprocess.check_output(['phylip seqboot<temp'], shell=True, cwd = '/tmp/%s' %dirname, timeout=60)
os.rename('/tmp/%s/outfile' %dirname, '/tmp/%s/infile' %dirname)
except:
kill_phylo() #FIXME, needs better way of handling this!
return "too big","too big","too big","too big","too big","too big","too big","too big"
### Write phylip input options
inp = open('/tmp/%s/temp' %dirname,'w')
if self.bootstrap:
inp.write('\n'.join(['m','d',str(self.bootstrap),'y'])+'\n')
else:
inp.write('y\n')
inp.close()
###
try:
subprocess.check_output(['phylip protdist<temp>>log'], shell=True, cwd = '/tmp/%s' %dirname, timeout=60)
except:
kill_phylo() #FIXME, needs better way of handling this!
return "too big","too big","too big","too big","too big","too big","too big","too big"
os.rename('/tmp/%s/infile' %dirname, '/tmp/%s/dupa' %dirname)
os.rename('/tmp/%s/outfile' %dirname, '/tmp/%s/infile' %dirname)
inp = open('/tmp/%s/temp' %dirname,'w')
if self.bootstrap:
### Write phylip input options
if self.UPGMA:
inp.write('\n'.join(['N','m',str(self.bootstrap),'111','y'])+'\n')
else:
inp.write('\n'.join(['m',str(self.bootstrap),'111','y'])+'\n')
else:
if self.UPGMA:
inp.write('N\ny\n')
else:
inp.write('y\n')
inp.close()
###
try:
subprocess.check_output(['phylip neighbor<temp'], shell=True, cwd = '/tmp/%s' %dirname, timeout=60)
except:
kill_phylo() #FIXME, needs better way of handling this!
return "too big","too big","too big","too big","too big","too big","too big","too big"
if self.bootstrap:
os.rename('/tmp/%s/outfile' %dirname, '/tmp/%s/infile' %dirname)
os.rename('/tmp/%s/outtree' %dirname, '/tmp/%s/intree' %dirname)
### Write phylip input options
inp = open('/tmp/%s/temp' %dirname,'w')
inp.write('y\n')
inp.close()
###
try:
subprocess.check_output(['phylip consense<temp'], shell=True, cwd = '/tmp/%s' %dirname, timeout=60)
except:
kill_phylo() #FIXME, needs better way of handling this!
return "too big","too big","too big","too big","too big","too big","too big","too big"
self.phylip = open('/tmp/%s/outtree' %dirname).read()
for acc in accesions.keys():
self.phylip=self.phylip.replace(acc,accesions[acc])
# self.phylogeny_output = self.phylip
self.outtree = open('/tmp/%s/outfile' %dirname).read().lstrip()
phylogeny_input = self.get_phylogeny('/tmp/%s/' %dirname)
shutil.rmtree('/tmp/%s' %dirname)
if build != False:
open('static/home/images/'+build+'_legend.svg','w').write(str(self.Tree.legend))
open('static/home/images/'+build+'_tree.xml','w').write(phylogeny_input)
else:
return phylogeny_input, self.branches, self.ttype, self.total, str(self.Tree.legend), self.Tree.box, self.Additional_info, self.buttons
def get_phylogeny(self, dirname):
self.Tree.treeDo(dirname, self.phylip,self.branches,self.family,self.Additional_info, self.famdict)
phylogeny_input = open('%s/out.xml' %dirname,'r').read().replace('\n','')
return phylogeny_input
def get_data(self):
return self.branches, self.ttype, self.total, str(self.Tree.legend), self.Tree.box, self.Additional_info, self.buttons
class Treeclass:
family = {}
def __init__(self):
self.Additional_info = {
"crystal": {
"include": "False",
"order": 6,
"colours": {
"crystal_true": "#6dcde1",
"crystal_false": "#EEE"
},
"color_type": "single",
"proteins": [],
"parent": None,
"child": None,
"name": "Crystals"
},
"class": {
"include": "True",
"order": 0,
"colours": {},
"proteins": [],
"color_type": "grayscale",
"parent": [],
"child": ["family,ligand"],
"name": "Class"
},
"family": {
"include": "True",
"order": 1,
"colours": {},
"proteins": [],
"color_type": "spectrum",
"parent": [],
"child": ["ligand"],
"name": "Ligand type"
},
"ligand": {
"include": "True",
"order": 2,
"colours": {},
"proteins": [],
"color_type": "spectrum",
"parent": ["family", "class"],
"child": [],
"name": "Receptor type"
},
"mutant": {
"include": "False",
"order": 3,
"colours": {
"mutant_true": "#6dcde1",
"mutant_false": "#EEE"
},
"color_type": "single",
"proteins": [],
"parent": [],
"child": ["mutant_plus", "mutant_minus"],
"name": "Mutated proteins"
},
"mutant_plus": {
"include": "False",
"order": 4,
"colours": {
"mutant_plus_true": "#6dcde1",
"mutant_plus_false": "#EEE"
},
"color_type": "single",
"proteins": [],
"parent": "mutant",
"child": [],
"name": "Positive affinity mutants"
},
"mutant_minus": {
"include": "False",
"order": 5,
"colours": {
"mutant_minus_true": "#6dcde1",
"mutant_minus_false": "#EEE"
},
"color_type": "single",
"proteins": [],
"parent": "mutant",
"child": [],
"name": "Negative affinity mutants"
}
}
self.buttons = [(x[1]['order'], x[1]['name'])
for x in sorted(self.Additional_info.items(),
key=lambda x: x[1]['order'])
if x[1]['include'] == 'True']
self.family = {}
self.phylip = None
self.outtree = None
self.dir = ''
def Prepare_file(self, request, build=False):
self.Tree = PrepareTree(build)
a = Alignment()
sets = ProteinSet.objects.all()
#### Get additional data ####
crysts = []
for n in sets:
if n.id == 1:
for prot in n.proteins.all():
crysts.append(prot.accession)
#############################
# get the user selection from session
if build != False:
################################## FOR BUILDING STATISTICS ONLY##########################
build_proteins = []
if build == '001':
cons_prots = []
for prot in Protein.objects.filter(
sequence_type__slug='consensus', species_id=1):
if prot.family.slug.startswith('001') and len(
prot.family.slug.split('_')) == 3:
build_proteins.append(prot)
for set in sets:
if set.id == 1:
for prot in set.proteins.all():
if prot.family.slug.startswith(
'001_'
) and prot.species.latin_name == 'H**o sapiens':
build_proteins.append(prot)
else:
for prot in Protein.objects.filter(sequence_type__slug='wt',
species_id=1):
if prot.family.slug.startswith(build):
build_proteins.append(prot)
a.load_proteins(build_proteins)
segments = ProteinSegment.objects.all()
a.load_segments(segments)
self.bootstrap, self.UPGMA, self.branches, self.ttype = [
0, 1, 1, 0
]
##################################################################
else:
simple_selection = request.session.get('selection', False)
a.load_proteins_from_selection(simple_selection)
a.load_segments_from_selection(simple_selection)
self.bootstrap, self.UPGMA, self.branches, self.ttype = map(
int, simple_selection.tree_settings)
if self.bootstrap != 0:
self.bootstrap = pow(10, self.bootstrap)
#### Create an alignment object
a.build_alignment()
a.calculate_statistics()
a.calculate_similarity()
self.total = len(a.proteins)
total_length = 0
for chain in a.proteins[0].alignment:
total_length += len(a.proteins[0].alignment[chain])
families = ProteinFamily.objects.all()
self.famdict = {}
for n in families:
self.famdict[self.Tree.trans_0_2_A(n.slug)] = n.name
dirname = unique_filename = uuid.uuid4()
os.mkdir('/tmp/%s' % dirname)
infile = open('/tmp/%s/infile' % dirname, 'w')
infile.write(' ' + str(self.total) + ' ' + str(total_length) +
'\n')
if len(a.proteins) < 3:
return 'More_prots', None, None, None, None, None, None, None
####Get additional protein information
for n in a.proteins:
fam = self.Tree.trans_0_2_A(n.protein.family.slug)
if n.protein.sequence_type.slug == 'consensus':
fam += '_CON'
link = n.protein.entry_name
name = n.protein.name.replace('<sub>',
'').replace('</sub>', '').replace(
'<i>', '').replace('</i>', '')
if '&' in name and ';' in name:
name = name.replace('&', '').replace(';', ' ')
acc = n.protein.accession
if acc:
acc = acc.replace('-', '_')
else:
acc = link.replace('-', '_')[:6]
spec = str(n.protein.species)
fam += '_' + n.protein.species.common_name.replace(' ',
'_').upper()
desc = name
if acc in crysts:
if not fam in self.Additional_info['crystal']['proteins']:
self.Additional_info['crystal']['proteins'].append(fam)
if len(name) > 25:
name = name[:25] + '...'
self.family[acc] = {
'name': name,
'family': fam,
'description': desc,
'species': spec,
'class': '',
'ligand': '',
'type': '',
'link': link
}
####Write PHYLIP input
sequence = ''
for chain in n.alignment:
for residue in n.alignment[chain]:
sequence += residue[2].replace('_', '-')
infile.write(acc + ' ' * 9 + sequence + '\n')
infile.close()
####Run bootstrap
if self.bootstrap:
### Write phylip input options
inp = open('/tmp/%s/temp' % dirname, 'w')
inp.write('\n'.join(['r', str(self.bootstrap), 'y', '77', 'y']) +
'\n')
inp.close()
###
subprocess.check_output(['phylip seqboot<temp'],
shell=True,
cwd='/tmp/%s' % dirname)
os.rename('/tmp/%s/outfile' % dirname, '/tmp/%s/infile' % dirname)
### Write phylip input options
inp = open('/tmp/%s/temp' % dirname, 'w')
if self.bootstrap:
inp.write('\n'.join(['m', 'd', str(self.bootstrap), 'y']) + '\n')
else:
inp.write('y\n')
inp.close()
###
subprocess.check_output(['phylip protdist<temp>>log'],
shell=True,
cwd='/tmp/%s' % dirname)
os.rename('/tmp/%s/infile' % dirname, '/tmp/%s/dupa' % dirname)
os.rename('/tmp/%s/outfile' % dirname, '/tmp/%s/infile' % dirname)
inp = open('/tmp/%s/temp' % dirname, 'w')
if self.bootstrap:
### Write phylip input options
if self.UPGMA:
inp.write('\n'.join(
['N', 'm', str(self.bootstrap), '111', 'y']) + '\n')
else:
inp.write('\n'.join(['m', str(self.bootstrap), '111', 'y']) +
'\n')
else:
if self.UPGMA:
inp.write('N\ny\n')
else:
inp.write('y\n')
inp.close()
###
subprocess.check_output(['phylip neighbor<temp'],
shell=True,
cwd='/tmp/%s' % dirname)
if self.bootstrap:
os.rename('/tmp/%s/outfile' % dirname, '/tmp/%s/infile' % dirname)
os.rename('/tmp/%s/outtree' % dirname, '/tmp/%s/intree' % dirname)
### Write phylip input options
inp = open('/tmp/%s/temp' % dirname, 'w')
inp.write('y\n')
inp.close()
###
subprocess.check_output(['phylip consense<temp'],
shell=True,
cwd='/tmp/%s' % dirname)
self.phylip = open('/tmp/%s/outtree' % dirname).read()
self.outtree = open('/tmp/%s/outfile' % dirname).read().lstrip()
phylogeny_input = self.get_phylogeny('/tmp/%s/' % dirname)
shutil.rmtree('/tmp/%s' % dirname)
if build != False:
open('static/home/images/' + build + '_legend.svg',
'w').write(str(self.Tree.legend))
open('static/home/images/' + build + '_tree.xml',
'w').write(phylogeny_input)
else:
return phylogeny_input, self.branches, self.ttype, self.total, str(
self.Tree.legend
), self.Tree.box, self.Additional_info, self.buttons
def get_phylogeny(self, dirname):
self.Tree.treeDo(dirname, self.phylip, self.branches, self.family,
self.Additional_info, self.famdict)
phylogeny_input = open('%s/out.xml' % dirname,
'r').read().replace('\n', '')
return phylogeny_input
def get_data(self):
return self.branches, self.ttype, self.total, str(
self.Tree.legend
), self.Tree.box, self.Additional_info, self.buttons
def Prepare_file(self, request,build=False):
self.Tree = PrepareTree(build)
a=Alignment()
sets = ProteinSet.objects.all()
#### Get additional data ####
crysts=[]
for n in sets:
if n.id==1:
for prot in n.proteins.all():
crysts.append(prot.entry_name)
#############################
# get the user selection from session
if build != False:
################################## FOR BUILDING STATISTICS ONLY##########################
build_proteins=[]
if build == '001':
cons_prots = []
for prot in Protein.objects.filter(sequence_type__slug='consensus', species_id=1):
if prot.family.slug.startswith('001') and len(prot.family.slug.split('_'))==3:
build_proteins.append(prot)
for set in sets:
if set.id==1:
for prot in set.proteins.all():
if prot.family.slug.startswith('001_') and prot.species.latin_name=='H**o sapiens':
build_proteins.append(prot)
else:
for prot in Protein.objects.filter(sequence_type__slug='wt', species_id=1):
if prot.family.slug.startswith(build):
build_proteins.append(prot)
a.load_proteins(build_proteins)
segments = ProteinSegment.objects.all()
a.load_segments(segments)
self.bootstrap,self.UPGMA,self.branches,self.ttype=[0,1,1,0]
##################################################################
else:
simple_selection=request.session.get('selection', False)
a.load_proteins_from_selection(simple_selection)
a.load_segments_from_selection(simple_selection)
self.bootstrap,self.UPGMA,self.branches,self.ttype = map(int,simple_selection.tree_settings)
if self.bootstrap!=0:
self.bootstrap=pow(10,self.bootstrap)
#### Create an alignment object
a.build_alignment()
a.calculate_statistics()
a.calculate_similarity()
self.total = len(a.proteins)
total_length = 0
for chain in a.proteins[0].alignment:
total_length += len(a.proteins[0].alignment[chain])
families = ProteinFamily.objects.all()
self.famdict = {}
for n in families:
self.famdict[self.Tree.trans_0_2_A(n.slug)]=n.name
dirname = unique_filename = uuid.uuid4()
os.mkdir('/tmp/%s' %dirname)
infile = open('/tmp/%s/infile' %dirname,'w')
infile.write(' '+str(self.total)+' '+str(total_length)+'\n')
if len(a.proteins) < 3:
return 'More_prots',None, None, None, None,None,None,None
####Get additional protein information
accesions = {}
for n in a.proteins:
fam = self.Tree.trans_0_2_A(n.protein.family.slug)
if n.protein.sequence_type.slug == 'consensus':
fam+='_CON'
entry_name = n.protein.entry_name
name = n.protein.name.replace('<sub>','').replace('</sub>','').replace('<i>','').replace('</i>','')
if '&' in name and ';' in name:
name = name.replace('&','').replace(';',' ')
acc = n.protein.accession
if acc:
acc = acc.replace('-','_')
else:
acc = link.replace('-','_')[:6]
spec = str(n.protein.species)
fam += '_'+n.protein.species.common_name.replace(' ','_').upper()
desc = name
if entry_name in crysts:
if not fam in self.Additional_info['crystal']['proteins']:
self.Additional_info['crystal']['proteins'].append(fam)
if len(name)>25:
name=name[:25]+'...'
self.family[entry_name] = {'name':name,'family':fam,'description':desc,'species':spec,'class':'','accession':acc,'ligand':'','type':'','link': entry_name}
accesions[acc]=entry_name
####Write PHYLIP input
sequence = ''
for chain in n.alignment:
for residue in n.alignment[chain]:
sequence += residue[2].replace('_','-')
infile.write(acc+' '*9+sequence+'\n')
infile.close()
####Run bootstrap
if self.bootstrap:
### Write phylip input options
inp = open('/tmp/%s/temp' %dirname,'w')
inp.write('\n'.join(['r',str(self.bootstrap),'y','77','y'])+'\n')
inp.close()
###
subprocess.check_output(['phylip seqboot<temp'], shell=True, cwd = '/tmp/%s' %dirname)
os.rename('/tmp/%s/outfile' %dirname, '/tmp/%s/infile' %dirname)
### Write phylip input options
inp = open('/tmp/%s/temp' %dirname,'w')
if self.bootstrap:
inp.write('\n'.join(['m','d',str(self.bootstrap),'y'])+'\n')
else:
inp.write('y\n')
inp.close()
###
subprocess.check_output(['phylip protdist<temp>>log'], shell=True, cwd = '/tmp/%s' %dirname)
os.rename('/tmp/%s/infile' %dirname, '/tmp/%s/dupa' %dirname)
os.rename('/tmp/%s/outfile' %dirname, '/tmp/%s/infile' %dirname)
inp = open('/tmp/%s/temp' %dirname,'w')
if self.bootstrap:
### Write phylip input options
if self.UPGMA:
inp.write('\n'.join(['N','m',str(self.bootstrap),'111','y'])+'\n')
else:
inp.write('\n'.join(['m',str(self.bootstrap),'111','y'])+'\n')
else:
if self.UPGMA:
inp.write('N\ny\n')
else:
inp.write('y\n')
inp.close()
###
subprocess.check_output(['phylip neighbor<temp'], shell=True, cwd = '/tmp/%s' %dirname)
if self.bootstrap:
os.rename('/tmp/%s/outfile' %dirname, '/tmp/%s/infile' %dirname)
os.rename('/tmp/%s/outtree' %dirname, '/tmp/%s/intree' %dirname)
### Write phylip input options
inp = open('/tmp/%s/temp' %dirname,'w')
inp.write('y\n')
inp.close()
###
subprocess.check_output(['phylip consense<temp'], shell=True, cwd = '/tmp/%s' %dirname)
self.phylip = open('/tmp/%s/outtree' %dirname).read()
for acc in accesions.keys():
self.phylip=self.phylip.replace(acc,accesions[acc])
# self.phylogeny_output = self.phylip
self.outtree = open('/tmp/%s/outfile' %dirname).read().lstrip()
phylogeny_input = self.get_phylogeny('/tmp/%s/' %dirname)
shutil.rmtree('/tmp/%s' %dirname)
if build != False:
open('static/home/images/'+build+'_legend.svg','w').write(str(self.Tree.legend))
open('static/home/images/'+build+'_tree.xml','w').write(phylogeny_input)
else:
return phylogeny_input, self.branches, self.ttype, self.total, str(self.Tree.legend), self.Tree.box, self.Additional_info, self.buttons
