Mercurial > repos > fabio > sbtas_se

comparison retrieve.py @ 0:00d6e82d74e9 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded 20180122
author fabio
date Mon, 22 Jan 2018 16:41:50 -0500
parents
children 4291c9d1ff07
comparison
equal deleted inserted replaced
-1:000000000000 0:00d6e82d74e9
1 #!/usr/bin/env python
2
3 # NCBI SRA Tools
4 # https://galaxyproject.org/tutorials/upload/
5
6 import os
7 import optparse
8 from subprocess import Popen, PIPE
9
10 db_key = "?";
11 sra_instant_url = "ftp://ftp-trace.ncbi.nlm.nih.gov/sra/sra-instant/reads/ByRun/sra/SRR/";
12
13 def convertSRA(tmp_dir, accession_number, data_format):
14 absolute_tmp_dir = os.path.abspath(tmp_dir);
15 sra_file_path = os.path.join(absolute_tmp_dir, accession_number+".sra");
16 if os.path.isdir(absolute_tmp_dir) and os.path.exists(sra_file_path):
17 process = None;
18 if data_format == ".fasta.gz":
19 process = Popen(["fastq-dump", "--fasta", "--gzip", sra_file_path, "--outdir", absolute_tmp_dir], stdout=PIPE);
20 elif data_format == ".fastq.gz":
21 process = Popen(["fastq-dump", "--gzip", sra_file_path, "--outdir", absolute_tmp_dir], stdout=PIPE);
22 elif data_format == ".fasta":
23 process = Popen(["fastq-dump", "--fasta", sra_file_path, "--outdir", absolute_tmp_dir], stdout=PIPE);
24 elif data_format == ".fastq":
25 process = Popen(["fastq-dump", sra_file_path, "--outdir", absolute_tmp_dir], stdout=PIPE);
26 else:
27 process = None;
28 if process is not None:
29 (output, err) = process.communicate();
30 if err:
31 # kill the process
32 # kill_process(process.pid);
33 # remove any trace of the output file
34 an_file_path = os.path.join(tmp_dir, accession_number+data_format);
35 if os.path.exists(an_file_path):
36 os.unlink(an_file_path);
37 # try to restart the process
38 return downloadAccessionData(tmp_dir, accession_number, data_format);
39 #exit_code = process.wait();
40 return os.path.join(tmp_dir, accession_number+data_format);
41 return "";
42
43 def downloadAccessionData(accession_number, accession_path, appdata_path, data_format, limit=10):
44 split = accession_number[:6];
45 srr_path = sra_instant_url+split+"/"+accession_number+"/"+accession_number+".sra";
46 sra_file_path = os.path.join(appdata_path, accession_number+".sra");
47 process = Popen(['wget', srr_path, "--output-document="+sra_file_path], stdout=PIPE);
48 (output, err) = process.communicate();
49 if err:
50 # remove any trace of the output file
51 if os.path.exists(an_file_path):
52 os.unlink(an_file_path);
53 # try to restart the process
54 if limit > 0:
55 return downloadAccessionData(accession_number, accession_path, appdata_path, data_format, limit-1);
56 return -1;
57 if os.path.exists(sra_file_path):
58 converted_file_path = convertSRA(appdata_path, accession_number, data_format);
59 if os.path.exists(converted_file_path):
60 os.rename(converted_file_path, accession_path);
61 os.unlink(sra_file_path);
62 return 0;
63
64 def process_accessions( options, args ):
65 # create appdata dir if it does not exist
66 appdata_path = options.appdata;
67 if not os.path.exists(appdata_path):
68 os.makedirs(appdata_path);
69 data_format = options.dataformat;
70 '''
71 # Collection test
72 test_file_name = "Test Collection" + "_" + "SRRtest" + "_" + data_format[1:] + "_" + db_key;
73 test_file_path = os.path.join(appdata_path, test_file_name);
74 file = open(test_file_path, "w");
75 file.write("Hello World");
76 file.close();
77 '''
78 # read inputs
79 comma_sep_file_paths = options.files;
80 #print("files: "+str(comma_sep_file_paths)+" - "+str(type(comma_sep_file_paths)));
81 # check if options.files contains at least one file path
82 if comma_sep_file_paths is not None:
83 # split file paths
84 file_paths = comma_sep_file_paths.split(",");
85 # split file names
86 comma_sep_file_names = str(options.names);
87 #print("names: "+str(comma_sep_file_names));
88 file_names = comma_sep_file_names.split(",");
89 # populate a dictionary with the files containing the sequences to query
90 for idx, file_path in enumerate(file_paths):
91 file_name = file_names[idx];
92 #print(file_name + ": " + file_path);
93 with open(file_path) as accessions:
94 for line in accessions:
95 if line.strip() != "" and not line.startswith(">"):
96 accession_number = line.strip();
97 filename_with_collection_prefix = file_name + "_" + accession_number + "_" + data_format[1:] + "_" + db_key;
98 accession_path = os.path.join(appdata_path, filename_with_collection_prefix)
99 # download fastq filte related to accession_number
100 downloadAccessionData( accession_number, accession_path, appdata_path, data_format );
101 return 0;
102
103 def __main__():
104 # Parse the command line options
105 usage = "Usage: retrieve.py --files comma_sep_file_paths --names comma_seq_file_names --format data_format --appdata folder_name";
106 parser = optparse.OptionParser(usage = usage);
107 parser.add_option("-f", "--files", type="string",
108 action="store", dest="files", help="comma separated files path");
109 parser.add_option("-n", "--names", type="string",
110 action="store", dest="names", help="comma separated names associated to the files specified in --files");
111 parser.add_option("-e", "--format", type="string",
112 action="store", dest="dataformat", help="data format");
113 parser.add_option("-a", "--appdata", type="string",
114 action="store", dest="appdata", help="appdata folder name");
115 parser.add_option("-v", "--version", action="store_true", dest="version",
116 default=False, help="display version and exit");
117 (options, args) = parser.parse_args();
118 if options.version:
119 print __version__;
120 else:
121 return process_accessions( options, args );
122
123 if __name__ == "__main__": __main__()