Mercurial > repos > davidvanzessen > imgt_locus_split
comparison imgt_locus_split.py @ 1:418b7dbc8947 draft
Uploaded
author | davidvanzessen |
---|---|
date | Mon, 17 Jul 2017 08:54:02 -0400 |
parents | b00c257f0a67 |
children | 4bb8f6523130 |
comparison
equal
deleted
inserted
replaced
0:b00c257f0a67 | 1:418b7dbc8947 |
---|---|
42 file = os.path.join(check, file) | 42 file = os.path.join(check, file) |
43 shutil.move(file, new_file) | 43 shutil.move(file, new_file) |
44 shutil.rmtree(check) | 44 shutil.rmtree(check) |
45 | 45 |
46 | 46 |
47 def filter_tabular_file(old_file, new_file, column, regex): | 47 def filter_imgt_file(old_file, new_file, column, fltr): |
48 logging.debug("Filtering {0} with {1}".format(old_file, regex.pattern)) | 48 logging.debug("Filtering {0} with {1}".format(old_file, fltr)) |
49 first = True | 49 first = True |
50 total = 0 | 50 total = 0 |
51 remain = 0 | 51 remain = 0 |
52 with open(old_file, 'r') as of, open(new_file, 'w') as nf: | 52 with open(old_file, 'r') as of, open(new_file, 'w') as nf: |
53 column_index = -1 | 53 column_index = -1 |
57 column_index = splt.index(column) | 57 column_index = splt.index(column) |
58 first = False | 58 first = False |
59 nf.write(line) | 59 nf.write(line) |
60 continue | 60 continue |
61 total += 1 | 61 total += 1 |
62 if len(splt) >= column_index and regex.search(splt[column_index]): | 62 if len(splt) > column_index and splt[column_index].find(fltr) != -1: |
63 remain += 1 | 63 remain += 1 |
64 nf.write(line) | 64 nf.write(line) |
65 return total, remain | 65 return total, remain |
66 | 66 |
67 | 67 |
68 def all_same_in_list(l): | 68 def all_same_in_list(l): |
69 return all(l[0] == x for x in l[1:]) | 69 return all(l[0] == x for x in l[1:]) |
70 | 70 |
71 | 71 |
72 def filter_imgt_dir(imgt_dir, loci): | 72 def filter_imgt_dir(imgt_dir, locus): |
73 logging.info("Filtering {0} with {1}".format(imgt_dir, loci)) | 73 logging.info("Working on {0}".format(locus)) |
74 imgt_files = [f for f in os.listdir(imgt_dir) if imgt_file_regex.match(f)] | 74 imgt_files = [f for f in os.listdir(imgt_dir) if imgt_file_regex.match(f)] |
75 tmp_file = os.path.join(imgt_dir, "tmp.txt") | 75 tmp_file = os.path.join(imgt_dir, "tmp.txt") |
76 totals = [] | 76 totals = [] |
77 remains = [] | 77 remains = [] |
78 loci_regex = re.compile("|".join(loci)) | |
79 for imgt_file in imgt_files: | 78 for imgt_file in imgt_files: |
80 imgt_file = os.path.join(imgt_dir, imgt_file) | 79 imgt_file = os.path.join(imgt_dir, imgt_file) |
81 total, remain = filter_tabular_file(imgt_file, tmp_file, "V-GENE and allele", loci_regex) | 80 total, remain = filter_imgt_file(imgt_file, tmp_file, "V-GENE and allele", locus) |
82 totals.append(total) | 81 totals.append(total) |
83 remains.append(remain) | 82 remains.append(remain) |
84 logging.debug("{0} rows, {1} after filtering".format(total, remain)) | 83 logging.debug("{0} rows, {1} after filtering".format(total, remain)) |
85 shutil.move(tmp_file, imgt_file) | 84 shutil.move(tmp_file, imgt_file) |
86 if not (all_same_in_list(totals) and all_same_in_list(remains)): | 85 if not (all_same_in_list(totals) and all_same_in_list(remains)): |
87 logging.warning("Not all files had the same number of sequences remaining for {0}: {1}".format(imgt_dir, remains)) | 86 logging.warning("Not all files had the same number of sequences remaining for {0}".format(imgt_dir)) |
88 return totals[0], remains[0] | 87 return totals[0], remains[0] |
89 | 88 |
90 | 89 |
91 def make_new_xz_file(input_dir, output_file): | 90 def make_new_xz_file(input_dir, output_file): |
92 logging.info("Creating new IMGT zip for {0} at {1}".format(input_dir, output_file)) | 91 logging.info("Creating new IMGT zip for {0} at {1}".format(input_dir, output_file)) |
98 out.add(imgt_file, arcname=os.path.basename(imgt_file)) | 97 out.add(imgt_file, arcname=os.path.basename(imgt_file)) |
99 | 98 |
100 | 99 |
101 def main(): | 100 def main(): |
102 parser = argparse.ArgumentParser() | 101 parser = argparse.ArgumentParser() |
103 parser.add_argument("-i", "--input", help="The input IMGT file", required=True) | 102 parser.add_argument("--input", help="The input IMGT file", required=True) |
104 parser.add_argument("-l", "--loci", help="The Loci to filter on", required=True) | 103 parser.add_argument("--output-ig", help="The output file for new IMGT ZIP with just IG sequences", default="None") |
105 parser.add_argument("-o", "--output", help="The output file for the new IMGT zip with just the filtered sequences", required=True) | 104 parser.add_argument("--output-igh", help="The output file for new IMGT ZIP with just IGH sequences", default="None") |
105 parser.add_argument("--output-igk", help="The output file for new IMGT ZIP with just IGK sequences", default="None") | |
106 parser.add_argument("--output-igl", help="The output file for new IMGT ZIP with just IGL sequences", default="None") | |
107 parser.add_argument("--output-tr", help="The output file for new IMGT ZIP with just TR sequences", default="None") | |
108 parser.add_argument("--output-tra", help="The output file for new IMGT ZIP with just TRA sequences", default="None") | |
109 parser.add_argument("--output-trb", help="The output file for new IMGT ZIP with just TRB sequences", default="None") | |
110 parser.add_argument("--output-trd", help="The output file for new IMGT ZIP with just TRD sequences", default="None") | |
111 parser.add_argument("--output-trg", help="The output file for new IMGT ZIP with just TRG sequences", default="None") | |
106 | 112 |
107 logging.basicConfig(filename="./log.html", level=logging.DEBUG, format="%(asctime)s: %(message)s <br />", | 113 logging.basicConfig(filename="./log.html", level=logging.DEBUG, format="%(asctime)s: %(message)s <br />", |
108 datefmt='%Y/%m/%d %H:%M:%S') | 114 datefmt='%Y/%m/%d %H:%M:%S') |
109 logging.getLogger().addHandler(logging.StreamHandler(sys.stdout)) | 115 logging.getLogger().addHandler(logging.StreamHandler(sys.stdout)) |
110 logging.info("Started IMGT locus split") | 116 logging.info("Started IMGT locus split") |
111 | 117 |
112 args = parser.parse_args() | 118 args = parser.parse_args() |
113 input_file = args.input | 119 input_file = args.input |
114 loci = args.loci.split(",") | 120 output_ig = args.output_ig |
115 output_file = args.output | 121 output_igh = args.output_igh |
122 output_igk = args.output_igk | |
123 output_igl = args.output_igl | |
124 output_tr = args.output_tr | |
125 output_tra = args.output_tra | |
126 output_trb = args.output_trb | |
127 output_trd = args.output_trd | |
128 output_trg = args.output_trg | |
129 | |
130 loci = { | |
131 "IG": output_ig, | |
132 "IGH": output_igh, | |
133 "IGK": output_igk, | |
134 "IGL": output_igl, | |
135 "TR": output_tr, | |
136 "TRA": output_tra, | |
137 "TRB": output_trb, | |
138 "TRD": output_trd, | |
139 "TRG": output_trg | |
140 } | |
141 | |
142 loci_to_filter = {} | |
116 | 143 |
117 logging.debug("All Parameters:") | 144 logging.debug("All Parameters:") |
118 logging.debug("Input: {0}".format(input_file)) | 145 logging.debug("Input: {0}".format(input_file)) |
119 logging.debug("Loci: {0}".format(loci)) | 146 for locus, path in loci.items(): |
120 logging.debug("Output: {0}".format(output_file)) | 147 logging.debug("{0}: {1}".format(locus, path)) |
148 if path != "None" and os.path.isdir(os.path.split(path)[0]): | |
149 loci_to_filter[locus] = path | |
121 | 150 |
122 if len(loci) == 0: | 151 if len(loci_to_filter) == 0: |
123 raise Exception("No locus selected, nothing to do") | 152 raise Exception("No locus selected, nothing to do") |
153 | |
154 logging.info("Parameters:") | |
155 for locus, path in loci_to_filter.items(): | |
156 logging.info("{0}: {1}".format(locus, path)) | |
124 | 157 |
125 work_dir = tempfile.mkdtemp() | 158 work_dir = tempfile.mkdtemp() |
126 original_files_dir = os.path.join(work_dir, "original") | 159 original_files_dir = os.path.join(work_dir, "original") |
127 os.mkdir(original_files_dir) | 160 os.mkdir(original_files_dir) |
128 | 161 |
129 unpack_imgt_zip(input_file, original_files_dir) | 162 unpack_imgt_zip(input_file, original_files_dir) |
130 | 163 |
131 total, remain = filter_imgt_dir(original_files_dir, loci) | 164 for locus, path in loci_to_filter.items(): |
132 logging.info("{0}\t{1}".format(total, remain)) | 165 locus_dir = os.path.join(work_dir, locus) |
166 shutil.copytree(original_files_dir, locus_dir) | |
167 total, remain = filter_imgt_dir(locus_dir, locus) | |
168 logging.info("{0}\t{1}\t{2}\t{3}".format(locus, path, total, remain)) | |
133 | 169 |
134 make_new_xz_file(original_files_dir, output_file) | 170 make_new_xz_file(locus_dir, loci_to_filter[locus]) |
135 | 171 |
136 | 172 |
137 if __name__ == "__main__": | 173 if __name__ == "__main__": |
138 main() | 174 main() |