def draw_light_heavy_graph(directory):
analysis_path = os.path.join(directory, "analysis")
heavy_path = os.path.join(analysis_path, "heavy.fa")
light_path = os.path.join(analysis_path, "light.fa")
heavy = read_fasta(heavy_path)
light = read_fasta(light_path)
heavy_clusters = load_clusters(analysis_path, "heavy")
light_clusters = load_clusters(analysis_path, "light")
heavy_nodes = get_node_lists(heavy)
light_nodes = get_node_lists(light)
heavy_colors = color_nodes(heavy_nodes, heavy_clusters)
light_colors = color_nodes(light_nodes, light_clusters)
colors = unify_pair(heavy_colors, light_colors)
heavy_labels = get_labels(heavy_nodes)
light_labels = get_labels(light_nodes)
labels = unify_pair(heavy_labels, light_labels)
hl_edges = get_edges_list(heavy_nodes, light_nodes)
edges = unify_edges(len(heavy_nodes), hl_edges, labels)
G = nx.Graph()
G.add_nodes_from(get_vertices(edges))
G.add_edges_from(edges)
nx.draw_networkx(G, color=colors)
plt.savefig(os.path.join(analysis_path, "connections.pdf"))
def process(directory,
similarity_name="similarity",
minlen=100,
shead=0,
prct=1.):
similarity_dir = os.path.join(directory, similarity_name)
if os.path.isdir(similarity_dir):
shutil.rmtree(similarity_dir)
os.mkdir(similarity_dir)
common = get_common_name(directory, "-amino-aligned.fa")
print(common)
light = process_one(
similarity_dir, "light",
read_fasta(
os.path.join(directory, common + "light-chains-amino-aligned.fa"),
False), minlen, shead, prct)
heavy = process_one(
similarity_dir, "heavy",
read_fasta(
os.path.join(directory, common + "heavy-chains-amino-aligned.fa"),
False), minlen, shead, prct)
with open(os.path.join(similarity_dir, "report.htm"), "wt") as fd:
fd.write(generate_report(heavy, light, minlen, shead, prct))
def process(directory, analysis_name="analysis", prct=False):
analysis_dir = os.path.join(directory, analysis_name)
if os.path.isdir(analysis_dir):
shutil.rmtree(analysis_dir)
os.mkdir(analysis_dir)
common = get_common_name(directory, "-amino-aligned.fa")
process_one(analysis_dir, "light", read_fasta(os.path.join(directory, common + "light-chains-amino-aligned.fa"),
False), prct)
process_one(analysis_dir, "heavy", read_fasta(os.path.join(directory, common + "heavy-chains-amino-aligned.fa"),
False), prct)
def load_clusters(directory, strtype):
result = []
clusters_dir = os.path.join(os.path.join(directory, strtype), "clusters")
for cluster in filter(lambda x: x.endswith(".fasta") or x.endswith(".fa"),
os.listdir(clusters_dir)):
result.append(read_fasta(os.path.join(clusters_dir, cluster), False))
return result
def process(directory, similarity_name="similarity", minlen=100, shead=0, prct=1.):
similarity_dir = os.path.join(directory, similarity_name)
if os.path.isdir(similarity_dir):
shutil.rmtree(similarity_dir)
os.mkdir(similarity_dir)
common = get_common_name(directory, "-amino-aligned.fa")
print(common)
light = process_one(similarity_dir, "light", read_fasta(os.path.join(directory,
common + "light-chains-amino-aligned.fa"),
False), minlen, shead, prct)
heavy = process_one(similarity_dir, "heavy", read_fasta(os.path.join(directory,
common + "heavy-chains-amino-aligned.fa"),
False), minlen, shead, prct)
with open(os.path.join(similarity_dir, "report.htm"), "wt") as fd:
fd.write(generate_report(heavy, light, minlen, shead, prct))
def make_alignments(clusterdir):
for cluster in os.listdir(clusterdir):
logging.debug("cluster: {}".format(cluster))
fasta = read_fasta(os.path.join(clusterdir, cluster), False)
pos = cluster.rfind('.')
name = cluster[:pos]
with open(os.path.join(clusterdir, "{}.htm".format(name)), "wt") as fd:
fd.write(make_document(name, fasta))
def process(directory, analysis_name="analysis", prct=False):
analysis_dir = os.path.join(directory, analysis_name)
if os.path.isdir(analysis_dir):
shutil.rmtree(analysis_dir)
os.mkdir(analysis_dir)
common = get_common_name(directory, "-amino-aligned.fa")
process_one(
analysis_dir, "light",
read_fasta(
os.path.join(directory, common + "light-chains-amino-aligned.fa"),
False), prct)
process_one(
analysis_dir, "heavy",
read_fasta(
os.path.join(directory, common + "heavy-chains-amino-aligned.fa"),
False), prct)
def make_alignments(clusterdir):
for cluster in os.listdir(clusterdir):
logging.debug("cluster: {}".format(cluster))
fasta = read_fasta(os.path.join(clusterdir, cluster), False)
pos = cluster.rfind('.')
name = cluster[:pos]
with open(os.path.join(clusterdir, "{}.htm".format(name)), "wt") as fd:
fd.write(make_document(name, fasta))
def run(self):
prct = False if not self.radio_confidence.isChecked() else float(self.spinadd.value()) / 100
out = str(self.linein.text())
if self.radio_sanger.isChecked():
analysis_dir = "analysis"
process(out, analysis_dir, prct)
hl_process(os.path.join(out, analysis_dir), prct)
else:
analysis_dir = os.path.dirname(out)
chain = os.path.basename(out)
chain = chain[:chain.rfind(".")] + "-analysis"
process_one(analysis_dir, chain, read_fasta(out, False), prct)
self.lok.setText("READY")
def multiple_alignment_use_files(file_input, alignment_type=SeqTypeData().TYPE_DEFAULT):
muscle_cmd = SeqTypeData().type2cmd_f[alignment_type]
result = iterpipes.run(iterpipes.linecmd(str(muscle_cmd) + " -in \"" + file_input + "\""))
result_list = []
for i in result:
result_list.append(i.strip())
# iores = StringIO()
# for i in result_list:
# print(i[0])
# iores.write(i)
name = file_input[:file_input.rfind('.')] + "-aligned.fa"
with open(name, "wt") as tmpfd:
tmpfd.write("\n".join(result_list))
return fasta_tools.read_fasta(name, False)
def load_chains(directory, suffix):
filenames = list(filter(lambda f: f.endswith(suffix), os.listdir(directory)))
heavy_chains = FastaDict()
light_chains = FastaDict()
logging.debug("Creating VL and VH arrays")
for filename in filenames:
name = os.path.join(directory, filename)
fd = read_fasta(name)
for seq in fd.keys():
if seq.endswith(CHAIN_MARKER_VL):
light_chains.set(seq, fd.get(seq))
elif seq.endswith(CHAIN_MARKER_VH):
heavy_chains.set(seq, fd.get(seq))
return light_chains, heavy_chains
def run(self):
minlen = 100 if not self.radio_confidence.isChecked() else int(self.spinmin.value())
shead = 0 if not self.radio_confidence.isChecked() else self.spinaddh.value()
prct = 1. if not self.radio_confidence.isChecked() else float(self.spinaddt.value()) / 100
out = str(self.linein.text())
if self.radio_sanger.isChecked():
analysis_dir = "similarity"
process(out, analysis_dir, minlen, shead, prct)
else:
analysis_dir = os.path.dirname(out)
chain = os.path.basename(out)
chain = chain[:chain.rfind(".")] + "-similarity"
process_one(analysis_dir, chain, read_fasta(out, False), minlen, shead, prct)
self.lok.setText("READY")
def multiple_alignment_use_files(file_input,
alignment_type=SeqTypeData().TYPE_DEFAULT):
muscle_cmd = SeqTypeData().type2cmd_f[alignment_type]
result = iterpipes.run(
iterpipes.linecmd(str(muscle_cmd) + " -in \"" + file_input + "\""))
result_list = []
for i in result:
result_list.append(i.strip())
# iores = StringIO()
# for i in result_list:
# print(i[0])
# iores.write(i)
name = file_input[:file_input.rfind('.')] + "-aligned.fa"
with open(name, "wt") as tmpfd:
tmpfd.write("\n".join(result_list))
return fasta_tools.read_fasta(name, False)
def run(self):
minlen = 100 if not self.radio_confidence.isChecked() else int(
self.spinmin.value())
shead = 0 if not self.radio_confidence.isChecked(
) else self.spinaddh.value()
prct = 1. if not self.radio_confidence.isChecked() else float(
self.spinaddt.value()) / 100
out = str(self.linein.text())
if self.radio_sanger.isChecked():
analysis_dir = "similarity"
process(out, analysis_dir, minlen, shead, prct)
else:
analysis_dir = os.path.dirname(out)
chain = os.path.basename(out)
chain = chain[:chain.rfind(".")] + "-similarity"
process_one(analysis_dir, chain, read_fasta(out, False), minlen,
shead, prct)
self.lok.setText("READY")
def load_clusters(directory, strtype):
result = []
clusters_dir = os.path.join(os.path.join(directory, strtype), "clusters")
for cluster in filter(lambda x: x.endswith(".fasta") or x.endswith(".fa"), os.listdir(clusters_dir)):
result.append(read_fasta(os.path.join(clusters_dir, cluster), False))
return result
def load_clusters(directory, strtype):
result = []
clusters_dir = os.path.join(os.path.join(directory, strtype), "clusters")
for cluster in os.listdir(clusters_dir):
result.append(read_fasta(os.path.join(clusters_dir, cluster)))
return result