Mercurial > repos > charles-bernard > data_manager_build_alfa_indexes
comparison data_manager_build_alfa_indexes/data_manager/data_manager_build_alfa_indexes.py @ 0:016200d4e379 draft
Uploaded
author | charles-bernard |
---|---|
date | Thu, 27 Oct 2016 05:15:05 -0400 |
parents | |
children | 6f0be85be8fb |
comparison
equal
deleted
inserted
replaced
-1:000000000000 | 0:016200d4e379 |
---|---|
1 #!/usr/bin/python | |
2 | |
3 import sys | |
4 import shutil | |
5 import re | |
6 import urllib2 | |
7 import subprocess | |
8 import gzip | |
9 import os | |
10 import tempfile | |
11 import logging | |
12 from optparse import OptionParser | |
13 from galaxy.util.json import from_json_string, to_json_string | |
14 | |
15 | |
16 def get_arg(): | |
17 parser = OptionParser() | |
18 parser.add_option("-e", "--ensembl", dest = 'ensembl_info', action = "store", nargs = 2, metavar = ("kingdom", "species_name"), type = "str") | |
19 parser.add_option("-o", "--output", dest='output_filename', action="store", nargs = 1, metavar = 'JSON_FILE') | |
20 (options, args) = parser.parse_args() | |
21 return options, args | |
22 | |
23 def cleanup_before_exit(tmp_dir): | |
24 if tmp_dir and os.path.exists(tmp_dir): | |
25 shutil.rmtree(tmp_dir) | |
26 | |
27 def get_page_content(url): | |
28 req = urllib2.Request(url) | |
29 page = urllib2.urlopen(req) | |
30 return page.read() | |
31 | |
32 | |
33 def download_file(link, local_file_name): | |
34 req = urllib2.Request(link) | |
35 src_file = urllib2.urlopen(req) | |
36 local_file = open(local_file_name, 'wb') | |
37 local_file.write(src_file.read()) | |
38 local_file.close() | |
39 | |
40 | |
41 def uncompress_gz(gz_file_name, uncompressed_file_name): | |
42 logging.info("____________________________________________________________") | |
43 logging.info("*** Uncompressing %s" % gz_file_name) | |
44 uncompressed_file = open(uncompressed_file_name, 'wb') | |
45 with gzip.open(gz_file_name, 'rb') as src_file: | |
46 uncompressed_file.write(src_file.read()) | |
47 uncompressed_file.close() | |
48 logging.info("-> Uncompressed !\n") | |
49 | |
50 | |
51 def add_data_table_entry( data_manager_dict, data_table_entry ): | |
52 data_manager_dict['data_tables'] = data_manager_dict.get( 'data_tables', {} ) | |
53 data_manager_dict['data_tables']['alfa_indexes'] = data_manager_dict['data_tables'].get( 'alfa_indexes', data_table_entry ) | |
54 return data_manager_dict | |
55 | |
56 | |
57 def standardize_species_name(species_name): | |
58 standard_species_name = re.sub(r'[)]$', '', species_name) | |
59 standard_species_name = re.sub(r'[ _),-.(=]+ *', '_', standard_species_name) | |
60 return standard_species_name.lower() | |
61 | |
62 | |
63 def get_ensembl_url_root(kingdom): | |
64 logging.info("____________________________________________________________") | |
65 logging.info("*** Determining Ensembl ftp root url") | |
66 if kingdom == 'vertebrates': | |
67 root = 'ftp://ftp.ensembl.org/pub/current_gtf/' | |
68 else: | |
69 root = 'ftp://ftp.ensemblgenomes.org/pub/%s/current/' % kingdom | |
70 logging.info("-> Determined !\n") | |
71 return root | |
72 | |
73 | |
74 def test_ensembl_species_exists(kingdom, url, species_name): | |
75 logging.info("____________________________________________________________") | |
76 logging.info ("*** Testing whether %s is referenced in Ensembl %s" % (species_name, kingdom)) | |
77 list_species_file_name = 'species_Ensembl%s%s.txt' % (kingdom[0].upper(), kingdom[1:]) | |
78 if kingdom=='vertebrates': | |
79 download_file(url, list_species_file_name) | |
80 else: | |
81 download_file(url + list_species_file_name, list_species_file_name) | |
82 | |
83 grep_result = subprocess.Popen(['grep', species_name, list_species_file_name], stdout=subprocess.PIPE, stderr=subprocess.STDOUT) | |
84 species_lines_matched, grep_error = grep_result.communicate() | |
85 if grep_error != None or species_lines_matched == "": | |
86 msg = 'The species \'%s\' is not referenced on Ensembl (%s)' % (species_name, kingdom) | |
87 logging.critical(msg) | |
88 sys.exit(msg) | |
89 | |
90 species_lines = species_lines_matched.split('\n') | |
91 del species_lines[-1] | |
92 nb_lines = len(species_lines) | |
93 | |
94 if nb_lines == 1: | |
95 columns = species_lines[0].split('\t') | |
96 found_species_name = columns[1] | |
97 if species_name != found_species_name: | |
98 logging.info('-> \'%s\' has been replace with the complete species name \'%s\'' % (species_name, found_species_name)) | |
99 return found_species_name, species_lines_matched | |
100 logging.info("-> Referenced !\n") | |
101 return species_name, species_lines_matched | |
102 else: | |
103 list_species = [''] * nb_lines | |
104 for i in range(0, nb_lines): | |
105 columns = species_lines[i].split('\t') | |
106 list_species[i] = columns[1] | |
107 exact_match = re.search('^%s$' % species_name, list_species[i]) | |
108 if exact_match: | |
109 logging.info("-> Referenced !\n") | |
110 return species_name, species_lines[i] | |
111 msg = 'The string \'%s\' has been matched against the list of Ensembl Species but is not a complete species name.\nPlease retry with one of the following species names:\n%s' % (species_name, list_species[0:]) | |
112 logging.critical(msg) | |
113 sys.exit(msg) | |
114 | |
115 | |
116 def get_ensembl_collection(kingdom, species_line): | |
117 logging.info("*** Extracting the %s_collection of the species" % kingdom) | |
118 collection_regex = re.compile('%s_.+_collection' % kingdom.lower()) | |
119 collection_match = re.search(collection_regex, species_line) | |
120 if not collection_match: | |
121 logging.info("-> Skiped: this species is not classified in a Ensembl %s collection\n" % kingdom) | |
122 return None | |
123 logging.info("-> Extracted !\n") | |
124 return collection_match.group(0) | |
125 | |
126 | |
127 def get_ensembl_gtf_archive_name(url_dir, species_name): | |
128 logging.info("____________________________________________________________") | |
129 logging.info("*** Extracting the gtf archive name of %s" % species_name) | |
130 gtf_archive_regex = re.compile('%s\..*\.[0-9]+\.gtf\.gz' % species_name, flags = re.IGNORECASE) | |
131 dir_content = get_page_content(url_dir) | |
132 gtf_archive_match = re.search(gtf_archive_regex, dir_content) | |
133 if not gtf_archive_match: | |
134 sys.exit('The species is referenced on Ensembl but error of nomenclature led to download failure') | |
135 gtf_archive_name = gtf_archive_match.group(0) | |
136 logging.info("-> Extracted !\n") | |
137 return gtf_archive_name | |
138 | |
139 | |
140 def get_ensembl_gtf_archive(kingdom, url, species_name, species_line): | |
141 if kingdom != 'vertebrates': | |
142 url = url + 'gtf/' | |
143 if kingdom == 'bacteria' or kingdom == 'protists' or kingdom == 'fungi': | |
144 collection = get_ensembl_collection(kingdom, species_line) | |
145 if collection != None: | |
146 url = url + "%s/" % collection | |
147 final_url = url + species_name + '/' | |
148 gtf_archive_name = get_ensembl_gtf_archive_name(final_url, species_name) | |
149 logging.info("____________________________________________________________") | |
150 logging.info("*** Download the gtf archive of %s" % species_name) | |
151 download_file(final_url + gtf_archive_name, gtf_archive_name) | |
152 logging.info("-> Downloaded !\n") | |
153 return gtf_archive_name | |
154 | |
155 | |
156 def generate_alfa_indexes(path_to_alfa, gtf_file_name): | |
157 logging.info("____________________________________________________________") | |
158 logging.info("*** Generating alfa indexes from %s" % gtf_file_name) | |
159 alfa_result = subprocess.Popen(['python', path_to_alfa + 'ALFA.py', '-a', gtf_file_name], stdout=subprocess.PIPE, stderr=subprocess.STDOUT) | |
160 alfa_out, alfa_err = alfa_result.communicate() | |
161 if alfa_err != None and not re.search('### End of program', alfa_err): | |
162 msg = 'Generation Failed due an alfa error: %s' % (alfa_err) | |
163 logging.critical(msg) | |
164 sys.exit(msg) | |
165 logging.info("-> Generated !\n") | |
166 | |
167 | |
168 def get_data_table_new_entry(gtf_archive_name): | |
169 info_list = gtf_archive_name.split('.') | |
170 species = info_list[0] | |
171 version = info_list[1] | |
172 release = info_list[2] | |
173 value = '%s_%s_%s' % (species, version, release) | |
174 dbkey = value | |
175 name = '%s: %s (release %s)' % (species, version, release) | |
176 prefix = '%s.%s.%s' % (species, version, release) | |
177 entry_dict = { 'species': species, 'version': version, 'release': release, 'value': value, 'dbkey': dbkey, 'name': name, 'prefix': prefix } | |
178 return entry_dict | |
179 | |
180 | |
181 def main(): | |
182 options, args = get_arg() | |
183 galaxy_root_dir = args[0] | |
184 | |
185 path_to_alfa = os.path.join(galaxy_root_dir, 'tools/alfa/') | |
186 path_to_tmp_dir = os.path.join(galaxy_root_dir, 'database/tmp/') | |
187 | |
188 if options.output_filename == None: | |
189 msg = 'No json output file specified' | |
190 logging.critical(msg) | |
191 sys.exit(msg) | |
192 output_filename = options.output_filename | |
193 params = from_json_string(open(output_filename).read()) | |
194 target_directory = params['output_data'][0]['extra_files_path'] | |
195 os.mkdir(target_directory) | |
196 | |
197 tmp_dir = tempfile.mkdtemp(prefix='tmp', suffix='', dir=path_to_tmp_dir) | |
198 os.chdir(tmp_dir) | |
199 log_file_name = 'galaxy_log_report.log' | |
200 logging.basicConfig(level=logging.INFO, filename=log_file_name, filemode="a+", format='%(message)s') | |
201 data_manager_dict = {} | |
202 | |
203 if options.ensembl_info: | |
204 kingdom, species_name = options.ensembl_info | |
205 species_name = standardize_species_name(species_name) | |
206 url = get_ensembl_url_root(kingdom) | |
207 species_name, species_line = test_ensembl_species_exists(kingdom, url, species_name) | |
208 gtf_archive_name = get_ensembl_gtf_archive(kingdom, url, species_name, species_line) | |
209 data_table_entry = get_data_table_new_entry(gtf_archive_name) | |
210 gtf_file_name = '%s.gtf' % data_table_entry['prefix'] | |
211 uncompress_gz(gtf_archive_name, gtf_file_name) | |
212 generate_alfa_indexes(path_to_alfa, gtf_file_name) | |
213 stranded_index_name = '%s.stranded.index' % data_table_entry['prefix'] | |
214 unstranded_index_name = '%s.unstranded.index' % data_table_entry['prefix'] | |
215 add_data_table_entry(data_manager_dict, data_table_entry) | |
216 | |
217 logging.info("____________________________________________________________") | |
218 logging.info("*** General Info") | |
219 logging.info("TMP DIR:\t%s" % tmp_dir) | |
220 logging.info("TARGET DIR:\t%s" % target_directory) | |
221 logging.info("URL ROOT:\t%s" % url) | |
222 logging.info("SPECIES:\t%s" % data_table_entry['species']) | |
223 logging.info("VERSION:\t%s" % data_table_entry['version']) | |
224 logging.info("RELEASE:\t%s" % data_table_entry['release']) | |
225 logging.info("VALUE:\t%s" % data_table_entry['value']) | |
226 logging.info("DBKEY:\t%s" % data_table_entry['dbkey']) | |
227 logging.info("NAME:\t%s" % data_table_entry['name']) | |
228 logging.info("PREFIX:\t%s" % data_table_entry['prefix']) | |
229 logging.info("____________________________________________________________") | |
230 logging.info("*** Intial dictionary") | |
231 logging.info("%s" % params) | |
232 | |
233 shutil.copyfile(stranded_index_name, os.path.join(target_directory, stranded_index_name)) | |
234 shutil.copyfile(unstranded_index_name, os.path.join(target_directory, unstranded_index_name)) | |
235 shutil.copyfile(log_file_name, os.path.join(target_directory, log_file_name)) | |
236 | |
237 cleanup_before_exit(tmp_dir) | |
238 | |
239 open(output_filename, 'wb').write(to_json_string(data_manager_dict)) | |
240 main() |