def main():
parser = argparse.ArgumentParser(description = "get cell inform from pickle file")
parser.add_argument('dir', metavar="<DIR>", help='directory containing subdirectories for each cell to be summarised')
parser.add_argument('--ignore_inkt', '-i', help='ignore iNKT cells ', action="store_true")
parser.add_argument("--sample", dest="sample_id", type=str, nargs='?', default="SAMPLE", help="sample id")
parser.add_argument("-o", "--output_prefix", metavar="STR", type=str, dest="output_prefix",
default="output_prefix", help="output prefix [default: %(default)s]")
args = parser.parse_args()
root_dir = os.path.abspath(args.dir)
subdirectories = os.walk(root_dir).next()[1]
pkl_dir = "filtered_TCR_seqs"
outdir = "{}/filtered_TCR_summary".format(root_dir)
tracer.makeOutputDir(outdir)
out1 = open("%s.dna_seq.fa" % args.output_prefix,"w")
out2 = open("%s.protein_seq.fa" % args.output_prefix,"w")
for d in subdirectories:
cell_pkl = "{root_dir}/{d}/{pkl_dir}/{d}.pkl".format(pkl_dir=pkl_dir, d=d, root_dir=root_dir)
if os.path.isfile(cell_pkl):
cl = pickle.load(open(cell_pkl))
if not cl.is_empty and not (cl.is_inkt and args.ignore_inkt):
for locus in ['A','B','D', 'G']:
if cl.all_recombinants[locus] is not None:
for recombinant in cl.all_recombinants[locus]:
aaseq = Seq(str(recombinant.dna_seq), generic_dna).translate()
seqAnn = "productive=%s in_frame=%s stop_codon=%s cdr3=%s TPM=%f" % (recombinant.productive, recombinant.in_frame, recombinant.stop_codon, recombinant.cdr3,recombinant.TPM)
if recombinant.productive and ( (locus == "A" and recombinant.TPM > 10) or ( (locus == "B" and recombinant.TPM > 15) ) ) :
print(">%s %s\n%s" % ("|".join([cl.name,locus, recombinant.contig_name, recombinant.identifier]), seqAnn, recombinant.dna_seq), file = out1)
print(">%s %s\n%s" % ("|".join([cl.name,locus, recombinant.contig_name, recombinant.identifier]), seqAnn, str(aaseq)), file = out2)
print("",file = out1)
print("",file = out2)
def main():
parser = argparse.ArgumentParser(description = "get cell inform from pickle file")
parser.add_argument('dir', metavar="<DIR>", help='directory containing subdirectories for each cell to be summarised')
parser.add_argument('--ignore_inkt', '-i', help='ignore iNKT cells ', action="store_true")
parser.add_argument("--sample", dest="sample_id", type=str, nargs='?', default="SAMPLE", help="sample id")
parser.add_argument("-o", "--output_prefix", metavar="STR", type=str, dest="output_prefix",
default="output_prefix", help="output prefix [default: %(default)s]")
args = parser.parse_args()
root_dir = os.path.abspath(args.dir)
subdirectories = os.walk(root_dir).next()[1]
pkl_dir = "filtered_TCR_seqs"
outdir = "{}/filtered_TCR_summary".format(root_dir)
tracer.makeOutputDir(outdir)
out1 = open("%s.dna_seq.fa" % args.output_prefix,"w")
out2 = open("%s.protein_seq.fa" % args.output_prefix,"w")
for d in subdirectories:
cell_pkl = "{root_dir}/{d}/{pkl_dir}/{d}.pkl".format(pkl_dir=pkl_dir, d=d, root_dir=root_dir)
if os.path.isfile(cell_pkl):
cl = pickle.load(open(cell_pkl))
if not cl.is_empty and not (cl.is_inkt and args.ignore_inkt):
for locus in ['A','B','D', 'G']:
if cl.all_recombinants[locus] is not None:
for recombinant in cl.all_recombinants[locus]:
aaseq = Seq(str(recombinant.dna_seq), generic_dna).translate()
seqAnn = "productive=%s in_frame=%s stop_codon=%s cdr3=%s" % (recombinant.productive, recombinant.in_frame, recombinant.stop_codon, recombinant.cdr3)
print(">%s %s\n%s" % ("|".join([cl.name,locus, recombinant.contig_name, recombinant.identifier]), seqAnn, recombinant.dna_seq), file = out1)
print(">%s %s\n%s" % ("|".join([cl.name,locus, recombinant.contig_name, recombinant.identifier]), seqAnn, str(aaseq)), file = out2)
print("",file = out1)
print("",file = out2)
def main():
parser = argparse.ArgumentParser(description = "Summarise set of cells with reconstructed TCR sequences")
parser.add_argument('dir', metavar="<DIR>", help='directory containing subdirectories for each cell to be summarised')
parser.add_argument('--keep_inkt', '-i', help='ignore iNKT cells when constructing networks', action="store_true")
parser.add_argument("--sample", dest="sample_id", type=str, nargs='?', default="SAMPLE", help="sample id")
args = parser.parse_args()
root_dir = os.path.abspath(args.dir)
cells = {}
empty_cells = []
NKT_cells = {}
subdirectories = os.walk(root_dir).next()[1]
pkl_dir = "filtered_TCR_seqs"
outdir = "{}/filtered_TCR_summary".format(root_dir)
tracer.makeOutputDir(outdir)
for d in subdirectories:
cell_pkl = "{root_dir}/{d}/{pkl_dir}/{d}.pkl".format(pkl_dir=pkl_dir, d=d, root_dir=root_dir)
if os.path.isfile(cell_pkl):
cl = pickle.load(open(cell_pkl))
cells[d] = cl
if cl.is_empty:
empty_cells.append(d)
if cl.is_inkt:
NKT_cells[d] = (cl.is_inkt, cl.getMainRecombinantIdentifiersForLocus('B'))
for cell_name in empty_cells:
del cells[cell_name]
if not args.keep_inkt:
for cell_name in NKT_cells.keys():
del cells[cell_name]
#output cells' info
get_cells_component(cells, args.sample_id)
#plot clonotype sizes
plt.figure()
clonotype_sizes = tracer.get_component_groups_sizes(cells)
w = 0.85
x_range = range(1, len(clonotype_sizes) + 1)
plt.bar(x_range, height=clonotype_sizes, width=w, color='black', align='center')
plt.gca().set_xticks(x_range)
plt.savefig("{}/clonotype_sizes_test.pdf".format(outdir))
