def run(self):
df = dd.read_csv(mods.get_file("daskin_" + g.PARAMS['tab']),
delimiter=";",
header=None,
usecols=[0, 1, g.PARAMS['col']],
names=["source", "target", "score"],
dtype={'score': 'float64'})
df = df[df['source'] != df['target']]
a = df.loc[:, ['source', 'target']].min(axis=1)
b = df.loc[:, ['source', 'target']].max(axis=1)
df['source'] = a
df['target'] = b
if (g.PARAMS['threshold'] == "ev"):
df = df.groupby(['source', 'target']).min()['score'].reset_index()
elif (g.PARAMS['threshold'] == "bs"):
df = df.groupby(['source', 'target']).max()['score'].reset_index()
df.to_csv(mods.get_file("tmp"),
compute=True,
index=False,
sep=';',
header=None)
queue.pop_queue()
return
def run(self):
folders = mods.get_folders(g.PARAMS['tab'])
for folder in folders:
create_dir(folders[folder])
if (g.PARAMS['tab'] == "metanodes"):
new_networks_path = folders['networks'] + "/" + os.path.basename(
g.FILES['input'])
try:
copyfile(g.FILES['input'], new_networks_path)
except Exception:
pass
sys.stderr = mods.RedirectError()
g.PARAMS['write_to_logfile'] = 1
print(ssn.text['dsp'] + str(g.PARAMS['dsp']))
print(ssn.text['input'] + g.FILES['input'])
print(ssn.text['output'] + g.FOLDERS['output'])
queue.pop_queue()
return
def run(self):
if (g.PARAMS['tab'] == "ssn"):
g.PARAMS['num_seqs'] = mods.HandleFasta.count_fasta(
g.FILES['input'])
if (g.PARAMS['threshold'] == "ev"):
g.PARAMS['parameter'] = "E-value"
g.PARAMS['col'] = 2
elif (g.PARAMS['threshold'] == "bs"):
g.PARAMS['parameter'] = "Bitscore"
if ((g.PARAMS['tab'] == "ssn") or (g.PARAMS['tab'] == "refine")):
g.PARAMS['col'] = 3
elif (g.PARAMS['tab'] == "metanodes"):
g.PARAMS['col'] = 2
if ((g.PARAMS['tab'] == "ssn") or (g.PARAMS['tab'] == "refine")):
if (g.PARAMS['min'] != g.PARAMS['max']):
even_threshold_max = ceil(
(g.PARAMS['max'] - g.PARAMS['min']) /
g.PARAMS['step']) * g.PARAMS['step'] + g.PARAMS['min']
if (even_threshold_max != g.PARAMS['max']):
g.PARAMS['max'] = even_threshold_max
g.PARAMS['num_ssn'] = (g.PARAMS['max'] -
g.PARAMS['min']) / g.PARAMS['step'] + 1
else:
g.PARAMS['num_ssn'] = 1
outfmt = []
if (g.PARAMS['outfmt_cs'] == 1):
outfmt.append("Cytoscape")
if (g.PARAMS['outfmt_pj'] == 1):
outfmt.append("Pajek")
if (g.PARAMS['outfmt_gp'] == 1):
outfmt.append("Gephi")
g.PARAMS['format'] = ', '.join(outfmt)
print(ssn.text['outfmt'] + g.PARAMS['format'])
print(ssn.text['score'] + g.PARAMS['parameter'])
if ((g.PARAMS['tab'] == "ssn") or (g.PARAMS['tab'] == "refine")):
print(ssn.text['min'] + str(g.PARAMS['min']))
print(ssn.text['step'] + str(g.PARAMS['step']))
print(ssn.text['max'] + str(g.PARAMS['max']))
print(ssn.text['num_ssn'] + str(g.PARAMS['num_ssn']))
elif (g.PARAMS['tab'] == "metanodes"):
print(ssn.text['meta_min'] + str(g.PARAMS['min']))
print(ssn.text['meta_max'] + str(g.PARAMS['max']))
elif (g.PARAMS['tab'] == "analysis"):
print(ssn.text['meta_min'] + str(g.PARAMS['min']))
queue.pop_queue()
return
def run(self):
filenames = glob(mods.get_file("master_in"))
with open(mods.get_file("masterfile"), 'w') as out:
for fastaheader in SeqIO.parse(mods.get_file("newhead"), "fasta"):
out.write(fastaheader.description + ";0;0;0;0;0;0\n")
for fn in filenames:
if (g.SSN_QUEUE[0] == "stop"):
break
with open(fn) as f:
for line in f:
if (len(line.rstrip()) != 0):
out.write(line)
queue.pop_queue()
return
def run(self):
warning = 0
infile = glob(mods.get_file("tmp"))
if ((g.PARAMS['outfmt_pj'] == 1) or (g.PARAMS['outfmt_cs'] == 1)):
if ((g.PARAMS['tab'] == "ssn") or (g.PARAMS['tab'] == "refine")):
outfile = mods.get_file("networks_cs") % int(g.PARAMS['min'])
elif (g.PARAMS['tab'] == "metanodes"):
outfile = mods.get_file("metanodes_cs") % int(g.PARAMS['max'])
mods.FormatOutput.cytoscape(infile, outfile, g.PARAMS['min'], ";",
"\t", 1)
warning = warning + mods.analysis_network(outfile, g.PARAMS['min'],
"\t")
if (g.PARAMS['outfmt_gp'] == 1):
if ((g.PARAMS['tab'] == "ssn") or (g.PARAMS['tab'] == "refine")):
outfile = mods.get_file("networks_gp") % int(g.PARAMS['min'])
elif (g.PARAMS['tab'] == "metanodes"):
outfile = mods.get_file("metanodes_gp") % int(g.PARAMS['max'])
mods.FormatOutput.gephi(infile, outfile, g.PARAMS['min'], ";", ";",
1)
warning = warning + mods.analysis_network(outfile, g.PARAMS['min'],
";")
if (g.PARAMS['outfmt_pj'] == 1):
outfile = mods.get_file("networks_pj") % int(g.PARAMS['min'])
infile = mods.get_file("networks_cs") % int(g.PARAMS['min'])
mods.FormatOutput.pajek(infile, outfile)
if (g.PARAMS['step'] != 0):
for i in range(g.PARAMS['min'] + g.PARAMS['step'],
g.PARAMS['max'] + g.PARAMS['step'],
g.PARAMS['step']):
if (g.SSN_QUEUE[0] == "stop"):
break
if ((g.PARAMS['outfmt_pj'] == 1)
or (g.PARAMS['outfmt_cs'] == 1)):
infile = [
mods.get_file("networks_cs") %
int(i - g.PARAMS['step'])
]
outfile = mods.get_file("networks_cs") % int(i)
mods.FormatOutput.cytoscape(infile, outfile, i, "\t", "\t",
0)
warning = warning + mods.analysis_network(outfile, i, "\t")
if (g.PARAMS['outfmt_gp'] == 1):
infile = [
mods.get_file("networks_gp") %
int(i - g.PARAMS['step'])
]
outfile = mods.get_file("networks_gp") % int(i)
mods.FormatOutput.gephi(infile, outfile, i, ";", ";", 0)
warning = warning + mods.analysis_network(outfile, i, ";")
if (g.PARAMS['outfmt_pj'] == 1):
infile = mods.get_file("networks_cs") % int(i)
outfile = mods.get_file("networks_pj") % int(i)
mods.FormatOutput.pajek(infile, outfile)
if (warning >= 1):
print(ssn.text['edge_warning'] % ssn.settings['edge_limit'] +
ssn.text['edge_solution'])
queue.pop_queue()
return
def kill(self, proc_pid):
process = psutil.Process(proc_pid)
for proc in process.children(recursive=True):
proc.kill()
process.kill()
queue.pop_queue()
def check_sub(self):
if self.proc.poll() is not None:
queue.pop_queue()
def write(self, message):
self.log.write(message)
queue.pop_queue("error")
def run(self):
rmtree(mods.get_folders(g.PARAMS['tab'])['tmp'])
queue.pop_queue()
return
def run(self):
if (g.PARAMS['outfmt_cs'] == 1):
delimiter = "\t"
elif (g.PARAMS['outfmt_gp'] == 1):
delimiter = ";"
outfile = mods.get_file("metanodes")
if ((g.PARAMS['tab'] == "metanodes") and (g.PARAMS['outfmt_cs'] == 1)):
parentfile = mods.get_file("metanodes_info_cs")
elif ((g.PARAMS['tab'] == "metanodes")
and (g.PARAMS['outfmt_gp'] == 1)):
parentfile = mods.get_file("metanodes_info_gp")
elif ((g.PARAMS['tab'] == "analysis") and
((g.PARAMS['outfmt_cs'] == 1) or (g.PARAMS['outfmt_gp'] == 1))):
parentfile = mods.get_file("analysis")
if (g.PARAMS['outfmt_cs'] == 1):
infile = mods.get_file("networks_cs") % int(g.PARAMS['min'])
mods.FormatOutput.cytoscape([infile], outfile, g.PARAMS['max'],
delimiter, delimiter, 2)
if (g.PARAMS['outfmt_gp'] == 1):
infile = mods.get_file("networks_gp") % int(g.PARAMS['min'])
mods.FormatOutput.gephi([infile], outfile, g.PARAMS['max'],
delimiter, delimiter, 2)
meta = dict()
H = nx.read_weighted_edgelist(outfile, delimiter=delimiter)
connected_components = sorted(nx.connected_components(H),
key=len,
reverse=True)
with open(parentfile, "w") as parent:
if (g.PARAMS['tab'] == "metanodes"):
parent.write("Metanode" + delimiter + "Size" + delimiter +
"Subnodes\n")
elif (g.PARAMS['tab'] == "analysis"):
parent.write("Node\tSubgroup\n")
i = 1
for group in connected_components:
group = list(group)
if (g.PARAMS['tab'] == "metanodes"):
parent.write(
str(group[0]) + delimiter + str(len(group)) +
delimiter + str(", ".join(group)) + "\n")
for entry in group:
meta.update({entry: group[0]})
elif (g.PARAMS['tab'] == "analysis"):
for entry in group:
meta.update({entry: group[0]})
parent.write(str(entry) + "\t" + str(i) + "\n")
i = i + 1
if (g.PARAMS['tab'] == "metanodes"):
with open(mods.get_file("daskin_" + g.PARAMS['tab']),
"w") as out, open(parentfile, "a") as parent:
with open(infile) as f:
next(f)
for line in f:
field = line.rstrip().split(delimiter)
if (len(field) >= 2):
if meta.get(field[0]) is None:
f_zero = field[0]
else:
f_zero = meta.get(field[0])
if meta.get(field[1]) is None:
f_one = field[1]
else:
f_one = meta.get(field[1])
out.write(
str(f_zero) + ";" + str(f_one) + ";" +
str(field[2]) + "\n")
elif ((len(field) == 1)
and (meta.get(field[0]) is None)):
parent.write(field[0] + delimiter + "1" +
delimiter + field[0] + "\n")
if (g.PARAMS['tab'] == "analysis"):
with open(parentfile, "a") as parent:
with open(infile) as f:
next(f)
for line in f:
field = line.rstrip().split(delimiter)
if ((len(field) == 1)
and (meta.get(field[0]) is None)):
parent.write(field[0] + delimiter + "0\n")
queue.pop_queue()