Mercurial > repos > kkonganti > cfsan_bettercallsal

comparison 0.7.0/bin/dl_pdg_metadata.py @ 17:0e7a0053e4a6

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload
author kkonganti
date Mon, 15 Jul 2024 10:42:02 -0400
parents
children
comparison
equal deleted inserted replaced
16:b90e5a7a3d4f 17:0e7a0053e4a6
1 #!/usr/bin/env python3
2
3 # Kranti Konganti
4
5 import argparse
6 import inspect
7 import logging
8 import os
9 import re
10 import shutil
11 import ssl
12 import tempfile
13 from html.parser import HTMLParser
14 from urllib.request import urlopen
15
16 # Set logging.f
17 logging.basicConfig(
18 format="\n" + "=" * 55 + "\n%(asctime)s - %(levelname)s\n" + "=" * 55 + "\n%(message)s\n",
19 level=logging.DEBUG,
20 )
21
22 # Multiple inheritence for pretty printing of help text.
23 class MultiArgFormatClasses(argparse.RawTextHelpFormatter, argparse.ArgumentDefaultsHelpFormatter):
24 pass
25
26
27 # HTMLParser override class to get PDG release and latest Cluster .tsv file
28 class NCBIPathogensHTMLParser(HTMLParser):
29 def __init__(self, *, convert_charrefs: bool = ...) -> None:
30 super().__init__(convert_charrefs=convert_charrefs)
31 self.reset()
32 self.href_data = list()
33
34 def handle_data(self, data):
35 self.href_data.append(data)
36
37
38 def dl_pdg(**kwargs) -> None:
39 """
40 Method to save the PDG metadata file and
41 return the latest PDG release.
42 """
43 db_path, url, regex, suffix, overwrite, release = [kwargs[k] for k in kwargs.keys()]
44
45 contxt = ssl.create_default_context()
46 contxt.check_hostname = False
47 contxt.verify_mode = ssl.CERT_NONE
48
49 if (db_path or url) == None:
50 logging.error("Please provide absolute UNIX path\n" + "to store the result DB flat files.")
51 exit(1)
52
53 if re.match(r"^PDG\d+\.\d+$", release):
54 url = re.sub("latest_snps", release.strip(), url)
55
56 html_parser = NCBIPathogensHTMLParser()
57 logging.info(f"Finding latest NCBI PDG release at:\n{url}")
58
59 with urlopen(url, context=contxt) as response:
60 with tempfile.NamedTemporaryFile(delete=False) as tmp_html_file:
61 shutil.copyfileobj(response, tmp_html_file)
62
63 with open(tmp_html_file.name, "r") as html:
64 html_parser.feed("".join(html.readlines()))
65
66 pdg_filename = re.search(regex, "".join(html_parser.href_data)).group(0)
67 pdg_release = pdg_filename.rstrip(suffix)
68 pdg_metadata_url = "/".join([url, pdg_filename])
69 pdg_release = pdg_filename.rstrip(suffix)
70 dest_dir = os.path.join(db_path, pdg_release)
71
72 logging.info(f"Found NCBI PDG file:\n{pdg_metadata_url}")
73
74 if (
75 not overwrite
76 and re.match(r".+?\.metadata\.tsv$", pdg_filename)
77 and os.path.exists(dest_dir)
78 ):
79 logging.error(f"DB path\n{dest_dir}\nalready exists. Please use -f to overwrite.")
80 exit(1)
81 elif overwrite and not re.match(r".+?\.reference_target\.cluster_list\.tsv$", pdg_filename):
82 shutil.rmtree(dest_dir, ignore_errors=True) if os.path.exists(dest_dir) else None
83 os.makedirs(dest_dir)
84 elif (
85 not overwrite
86 and re.match(r".+?\.metadata\.tsv$", pdg_filename)
87 and not os.path.exists(dest_dir)
88 ):
89 os.makedirs(dest_dir)
90
91 tsv_at = os.path.join(dest_dir, pdg_filename)
92 logging.info(f"Saving to:\n{tsv_at}")
93
94 with urlopen(pdg_metadata_url, context=contxt) as response:
95 with open(tsv_at, "w") as tsv:
96 tsv.writelines(response.read().decode("utf-8"))
97
98 html.close()
99 tmp_html_file.close()
100 os.unlink(tmp_html_file.name)
101 tsv.close()
102 response.close()
103
104 return tsv_at, dest_dir
105
106
107 def main() -> None:
108 """
109 This script is part of the `bettercallsal_db` Nextflow workflow and is only
110 tested on POSIX sytems.
111 It:
112 1. Downloads the latest NCBI Pathogens Release metadata file, which
113 looks like PDGXXXXXXXXXX.2504.metadata.csv and also the SNP cluster
114 information file which looks like PDGXXXXXXXXXX.2504.reference_target.cluster_list.tsv
115 2. Generates a new metadata file with only required information such as
116 computed_serotype, isolates GenBank or RefSeq downloadable genome FASTA
117 URL.
118 """
119
120 prog_name = os.path.basename(inspect.stack()[0].filename)
121
122 parser = argparse.ArgumentParser(
123 prog=prog_name, description=main.__doc__, formatter_class=MultiArgFormatClasses
124 )
125
126 # required = parser.add_argument_group("required arguments")
127
128 parser.add_argument(
129 "-db",
130 dest="db_path",
131 default=os.getcwd(),
132 required=False,
133 help="Absolute UNIX path to a path where all results files are\nstored.",
134 )
135 parser.add_argument(
136 "-f",
137 dest="overwrite_db",
138 default=False,
139 required=False,
140 action="store_true",
141 help="Force overwrite a PDG release directory at DB path\nmentioned with -db.",
142 )
143 parser.add_argument(
144 "-org",
145 dest="organism",
146 default="Salmonella",
147 required=False,
148 help="The organism to create the DB flat files\nfor.",
149 )
150 parser.add_argument(
151 "-rel",
152 dest="release",
153 default=False,
154 required=False,
155 help="If you get a 404 error, try mentioning the actual release identifier.\n"
156 + "Ex: For Salmonella, you can get the release identifier by going to:\n"
157 + " https://ftp.ncbi.nlm.nih.gov/pathogen/Results/Salmonella\n"
158 + "Ex: If you want metadata beloginging to release PDG000000002.2507, then you\n"
159 + " would use this command-line option as:\n -rel PDG000000002.2507",
160 )
161
162 args = parser.parse_args()
163 db_path = args.db_path
164 org = args.organism
165 overwrite = args.overwrite_db
166 release = args.release
167 ncbi_pathogens_loc = "/".join(
168 ["https://ftp.ncbi.nlm.nih.gov/pathogen/Results", org, "latest_snps"]
169 )
170
171 if not db_path:
172 db_path = os.getcwd()
173
174 # Save metadata
175 file, dest_dir = dl_pdg(
176 db_path=db_path,
177 url="/".join([ncbi_pathogens_loc, "Metadata"]),
178 regex=re.compile(r"PDG\d+\.\d+\.metadata\.tsv"),
179 suffix=".metadata.tsv",
180 overwrite=overwrite,
181 release=release,
182 )
183
184 # Save cluster to target mapping
185 dl_pdg(
186 db_path=db_path,
187 url="/".join([ncbi_pathogens_loc, "Clusters"]),
188 regex=re.compile(r"PDG\d+\.\d+\.reference_target\.cluster_list\.tsv"),
189 suffix="reference_target\.cluster_list\.tsv",
190 overwrite=overwrite,
191 release=release,
192 )
193
194 # Create accs.txt for dataformat to fetch required ACC fields
195 accs_file = os.path.join(dest_dir, "accs_all.txt")
196 with open(file, "r") as pdg_metadata_fh:
197 with open(accs_file, "w") as accs_fh:
198 for line in pdg_metadata_fh.readlines():
199 if re.match(r"^#", line) or line in ["\n", "\n\r", "\r"]:
200 continue
201 cols = line.strip().split("\t")
202 asm_acc = cols[9]
203 accs_fh.write(f"{asm_acc}\n") if (asm_acc != "NULL") else None
204 accs_fh.close()
205 pdg_metadata_fh.close()
206
207 logging.info("Finished writing accessions for dataformat tool.")
208
209
210 if __name__ == "__main__":
211 main()