Mercurial > repos > in_silico > cravat_score_and_annotate
comparison cravat_submit/cravat_submit.py @ 15:0835042eb731 draft
Uploaded
author | in_silico |
---|---|
date | Mon, 30 Jul 2018 13:32:05 -0400 |
parents | |
children |
comparison
equal
deleted
inserted
replaced
14:45b91fdd18ce | 15:0835042eb731 |
---|---|
1 from __future__ import print_function | |
2 import requests | |
3 import json | |
4 import time | |
5 from urllib import urlretrieve | |
6 # from urllib.request import urlretrieve | |
7 import sys | |
8 import csv | |
9 import argparse | |
10 | |
11 """ | |
12 Tool's email: | |
13 usernmae: cravatgalaxy@gmail.com | |
14 password: chicken_quesadilla | |
15 """ | |
16 | |
17 email = 'cravatgalaxy@gmail.com' | |
18 | |
19 class CravatSubmissionException(Exception): | |
20 def __init__(self, message): | |
21 super(CravatSubmissionException, self).__init__(message) | |
22 | |
23 class CravatSubmission(object): | |
24 | |
25 def get_cmd_args(self, argv): | |
26 parser = argparse.ArgumentParser() | |
27 parser.add_argument('path', | |
28 help="Path to python module") | |
29 parser.add_argument('--input', | |
30 '-i', | |
31 required = True, | |
32 help='Input path to a cravat file for querying',) | |
33 parser.add_argument('--output', | |
34 '-o', | |
35 default = None, | |
36 help = 'Output path to write results from query') | |
37 parser.add_argument('--analysis', | |
38 '-a', | |
39 required=True, | |
40 help = "Cravat analysis. Should be 'VEST', 'CHASM', 'NONE', or 'VEST;CHASM'") | |
41 return parser.parse_args(argv) | |
42 | |
43 def is_valid_analysis(self, analysis): | |
44 """: Test if analysis is a recognized value""" | |
45 analyses = ["VEST", "CHASM", "VEST;CHASM", ""] | |
46 return analysis in analyses | |
47 | |
48 def is_skippable(self, s): | |
49 """: Test if a line (str or list/tuple) is skippable, a.k.a. a header or blank line""" | |
50 if not isinstance(s, str): | |
51 raise CravatSubmissionException("is_skippable accepts a string") | |
52 skippable = s == "" \ | |
53 or s[0] == "#" \ | |
54 or s.startswith('"#For more information on CRAVAT') \ | |
55 or s.isspace() | |
56 return skippable | |
57 | |
58 def parse(self, s, sep='\t'): | |
59 """: Convert string line to an array of values""" | |
60 return s.strip().split(sep) | |
61 | |
62 def unparse(self, array, sep='\t', newline='\n'): | |
63 """: Convert an array of values to a writable string line""" | |
64 return sep.join([str(i) for i in array]) + newline | |
65 | |
66 def get_headers(self, path, pattern='Input line', sep='\t'): | |
67 """: Get the headers from a Results/Details file obtained from by a finished Cravat submission""" | |
68 with open(path, 'r') as f: | |
69 for line in f: | |
70 if line.startswith(pattern): | |
71 return self.parse(line) | |
72 return None | |
73 | |
74 def create_index(self, path, prop='Input line'): | |
75 """ | |
76 : Create an index of seek/tell positions in file associated to a line value. Used to record | |
77 : the location of lines betwen two files that are associated with each other without reading entire | |
78 : files into memory. | |
79 """ | |
80 headers = self.get_headers(path) | |
81 if prop not in headers: | |
82 raise CravatSubmissionException("Index retrievel property not found in headers") | |
83 prop_loc = headers.index(prop) | |
84 index = {} | |
85 with open(path, 'r') as f: | |
86 pos = 0 | |
87 line = f.readline() | |
88 while line != "": | |
89 if not self.is_skippable(line): | |
90 parsed = self.parse(line) | |
91 if not parsed == headers: | |
92 index[parsed[prop_loc]] = pos | |
93 pos = f.tell() | |
94 line = f.readline() | |
95 return index | |
96 | |
97 def get_header_val_dict(self, headers, vals): | |
98 """: Associate an array of header keys to an array of values.""" | |
99 return { header:val for (header, val) in zip(headers, vals) } | |
100 | |
101 def write_results(self, results_path, details_path, out_path, write_headers=True): | |
102 """ | |
103 : Using the paths to the Results and Details file from a Cravat Sumbission, | |
104 : write the output file. | |
105 """ | |
106 results_headers = self.get_headers(results_path) | |
107 details_headers = self.get_headers(details_path) | |
108 if results_headers == None \ | |
109 or details_headers == None: | |
110 raise CravatSubmissionException("Unable to intepret headers in Results or Details submission files") | |
111 headers = results_headers | |
112 headers.extend(filter(lambda x: x not in headers, details_headers)) | |
113 results_index = self.create_index(results_path) | |
114 details_index = self.create_index(details_path) | |
115 with open(results_path, 'r') as results_file, \ | |
116 open(details_path, 'r') as details_file, \ | |
117 open(out_path, 'w') as out_file: | |
118 if write_headers: | |
119 out_file.write(self.unparse(headers)) | |
120 for line_id, file_pos in results_index.items(): | |
121 results_file.seek(file_pos) | |
122 results_vals = self.parse(results_file.readline()) | |
123 results_dict = self.get_header_val_dict(results_headers, results_vals) | |
124 if line_id in details_index: | |
125 details_file.seek(details_index[line_id]) | |
126 details_vals = self.parse(details_file.readline()) | |
127 details_dict = self.get_header_val_dict(details_headers, details_vals) | |
128 # On a repeated entry, the Details value will overwrite Results value | |
129 results_dict.update(details_dict) | |
130 line = [ results_dict.get(header, 'None') for header in headers ] | |
131 out_file.write(self.unparse(line)) | |
132 | |
133 def submit(self, in_path, analysis): | |
134 """: Make a POST request to submit a job to production CRAVAT server.""" | |
135 if not self.is_valid_analysis(analysis): | |
136 raise ValueError("Did not get valid analyses.") | |
137 # Create post request to submit job to CRAVAT production server | |
138 submit = requests.post('http://cravat.us/CRAVAT/rest/service/submit', | |
139 files={'inputfile' : open(in_path)}, | |
140 data={'email' : email, | |
141 'analyses' : analysis}) | |
142 # Check job run status in loop until status is 'Success' | |
143 jobid = json.loads(submit.text)['jobid'] | |
144 while True: | |
145 check = requests.get('http://cravat.us/CRAVAT/rest/service/status', params={'jobid': jobid}) | |
146 status = json.loads(check.text)['status'] | |
147 print(status) | |
148 if status == 'Success': | |
149 break | |
150 else: | |
151 time.sleep(2) | |
152 # Download completed job results to local files | |
153 timestamp = time.strftime("%Y-%m-%d_%H-%M-%S_") | |
154 results_path = 'Z_Variant_Result' + timestamp + '.tsv' | |
155 details_path = 'Z_Additional_Details' + timestamp + '.tsv' | |
156 urlretrieve("http://cravat.us/CRAVAT/results/" + jobid + "/" + "Variant.Result.tsv", | |
157 filename=results_path) | |
158 urlretrieve("http://cravat.us/CRAVAT/results/" + jobid + "/" + "Variant_Additional_Details.Result.tsv", | |
159 filename=details_path) | |
160 return results_path, details_path | |
161 | |
162 if __name__ == "__main__": | |
163 submission = CravatSubmission() | |
164 cmd_args = submission.get_cmd_args(sys.argv) | |
165 # Galaxy converts semi-colons to X's. Switch it back | |
166 analysis = cmd_args.analysis | |
167 if analysis == "VESTXCHASM": | |
168 analysis = "VEST;CHASM" | |
169 results_path, details_path = submission.submit(cmd_args.input, analysis) | |
170 #submission.write_results('Results_test.tsv', 'Details_test.tsv', 'Out_test.tsv') | |
171 submission.write_results(results_path, details_path, cmd_args.output) |