|
1
|
1 #!/home/jjjjia/.conda/envs/py36/bin/python
|
|
|
2
|
|
|
3 #$ -S /home/jjjjia/.conda/envs/py36/bin/python
|
|
|
4 #$ -V # Pass environment variables to the job
|
|
|
5 #$ -N CPO_pipeline # Replace with a more specific job name
|
|
|
6 #$ -wd /home/jjjjia/testCases # Use the current working dir
|
|
|
7 #$ -pe smp 8 # Parallel Environment (how many cores)
|
|
|
8 #$ -l h_vmem=11G # Memory (RAM) allocation *per core*
|
|
|
9 #$ -e ./logs/$JOB_ID.err
|
|
|
10 #$ -o ./logs/$JOB_ID.log
|
|
|
11 #$ -m ea
|
|
|
12 #$ -M bja20@sfu.ca
|
|
|
13
|
|
|
14 #~/scripts/pipeline.py -i BC11-Kpn005_S2 -f /data/jjjjia/R1/BC11-Kpn005_S2_L001_R1_001.fastq.gz -r /data/jjjjia/R2/BC11-Kpn005_S2_L001_R2_001.fastq.gz -o pipelineResultsQsub -e "Klebsiella pneumoniae"
|
|
|
15
|
|
|
16 import subprocess
|
|
|
17 import pandas
|
|
|
18 import optparse
|
|
|
19 import os
|
|
|
20 import datetime
|
|
|
21 import sys
|
|
|
22 import time
|
|
|
23 import urllib.request
|
|
|
24 import gzip
|
|
|
25 import collections
|
|
|
26 import json
|
|
|
27 import numpy
|
|
|
28
|
|
|
29
|
|
2
|
30 debug = False #True #debug skips the shell scripts and also dump out a ton of debugging messages
|
|
1
|
31
|
|
|
32 if not debug:
|
|
|
33 #parses some parameters
|
|
|
34 parser = optparse.OptionParser("Usage: %prog [options] arg1 arg2 ...")
|
|
|
35 #required
|
|
|
36 parser.add_option("-i", "--id", dest="id", type="string", help="identifier of the isolate")
|
|
|
37 parser.add_option("-a", "--assembly", dest="assemblyPath", type="string", help="absolute file path to contigs fasta")
|
|
|
38 parser.add_option("-c", "--card-db", dest="cardDB", default = "/home/jjjjia/databases/card202.json", type="string", help="absolute path to card reference database")
|
|
|
39 parser.add_option("-o", "--output", dest="output", default='./', type="string", help="absolute path to output folder")
|
|
|
40 parser.add_option("-e", "--expected", dest="expectedSpecies", default="NA/NA/NA", type="string", help="expected species of the isolate")
|
|
|
41
|
|
|
42 #optionals
|
|
|
43 parser.add_option("-k", "--script-path", dest="scriptDir", default="/home/jjjjia/scripts", type="string", help="absolute file path to this script folder")
|
|
|
44 parser.add_option("-b", "--update-abricate-path", dest="updateAbPath", default = "", type="string", help="absolute file path to fasta sequence used for abricate database")
|
|
|
45 parser.add_option("-m", "--update-abricate-dbname", dest="updateAbName", default = "default", type="string", help="name of abricate database to update")
|
|
|
46 parser.add_option("-u", "--update-mlst", dest="updateMLST", default = "False", type="string", help="True = update MLST")
|
|
|
47 #used for parsing
|
|
|
48 parser.add_option("-s", "--mlst-scheme", dest="mlst", default= "/home/jjjjia/databases/scheme_species_map.tab", type="string", help="absolute file path to mlst scheme")
|
|
|
49
|
|
|
50
|
|
|
51 #parallelization, useless, these are hard coded to 8cores/64G RAM
|
|
|
52 #parser.add_option("-t", "--threads", dest="threads", default=8, type="int", help="number of cpu to use")
|
|
|
53 #parser.add_option("-p", "--memory", dest="memory", default=64, type="int", help="memory to use in GB")
|
|
|
54
|
|
|
55 (options,args) = parser.parse_args()
|
|
|
56 #if len(args) != 8:
|
|
|
57 #parser.error("incorrect number of arguments, all 7 is required")
|
|
|
58 curDir = os.getcwd()
|
|
|
59 outputDir = options.output
|
|
|
60 expectedSpecies = options.expectedSpecies
|
|
|
61 mlstScheme = options.mlst
|
|
|
62 tempDir = outputDir + "/shovillTemp"
|
|
|
63 scriptDir = options.scriptDir
|
|
|
64 updateAbName = options.updateAbName
|
|
|
65 updateAbPath = options.updateAbPath
|
|
|
66 updateMLST = options.updateMLST
|
|
|
67 cardDB=options.cardDB
|
|
|
68 assemblyPath=options.assemblyPath
|
|
|
69 ID = options.id
|
|
|
70 else:
|
|
|
71 manifestDir = ""
|
|
|
72 curDir = os.getcwd()
|
|
|
73 outputDir = "pipelineTest"
|
|
|
74 expectedSpecies = "Escherichia coli"
|
|
|
75 #threads = 8
|
|
|
76 #memory = 64
|
|
|
77 mlstScheme = outputDir + "/scheme_species_map.tab"
|
|
|
78 tempDir= outputDir + "/shovillTemp"
|
|
|
79 scriptDir = "~/scripts"
|
|
|
80 updateAbName = "cpo"
|
|
|
81 updateAbPath = "~/database/bccdcCPO.seq"
|
|
|
82 updateMLST = True
|
|
|
83 assemblyPath = "./"
|
|
|
84 cardDB = "./"
|
|
|
85 ID = "BC11-Kpn005_S2"
|
|
|
86
|
|
|
87 #region result objects
|
|
|
88 #define some objects to store values from results
|
|
|
89 #//TODO this is not the proper way of get/set private object variables. every value has manually assigned defaults intead of specified in init(). Also, use property(def getVar, def setVar).
|
|
|
90 class starFinders(object):
|
|
|
91 def __init__(self):
|
|
|
92 self.file = ""
|
|
|
93 self.sequence = ""
|
|
|
94 self.start = 0
|
|
|
95 self.end = 0
|
|
|
96 self.gene = ""
|
|
|
97 self.shortGene = ""
|
|
|
98 self.coverage = ""
|
|
|
99 self.coverage_map = ""
|
|
|
100 self.gaps = ""
|
|
|
101 self.pCoverage = 100.00
|
|
|
102 self.pIdentity = 100.00
|
|
|
103 self.database = ""
|
|
|
104 self.accession = ""
|
|
|
105 self.product = ""
|
|
|
106 self.source = "chromosome"
|
|
|
107 self.row = ""
|
|
|
108
|
|
|
109 class PlasFlowResult(object):
|
|
|
110 def __init__(self):
|
|
|
111 self.sequence = ""
|
|
|
112 self.length = 0
|
|
|
113 self.label = ""
|
|
|
114 self.confidence = 0
|
|
|
115 self.usefulRow = ""
|
|
|
116 self.row = ""
|
|
|
117
|
|
|
118 class MlstResult(object):
|
|
|
119 def __init__(self):
|
|
|
120 self.file = ""
|
|
|
121 self.speciesID = ""
|
|
|
122 self.seqType = 0
|
|
|
123 self.scheme = ""
|
|
|
124 self.species = ""
|
|
|
125 self.row=""
|
|
|
126
|
|
|
127 class mobsuiteResult(object):
|
|
|
128 def __init__(self):
|
|
|
129 self.file_id = ""
|
|
|
130 self.cluster_id = ""
|
|
|
131 self.contig_id = ""
|
|
|
132 self.contig_num = 0
|
|
|
133 self.contig_length = 0
|
|
|
134 self.circularity_status = ""
|
|
|
135 self.rep_type = ""
|
|
|
136 self.rep_type_accession = ""
|
|
|
137 self.relaxase_type = ""
|
|
|
138 self.relaxase_type_accession = ""
|
|
|
139 self.mash_nearest_neighbor = ""
|
|
|
140 self.mash_neighbor_distance = 0.00
|
|
|
141 self.repetitive_dna_id = ""
|
|
|
142 self.match_type = ""
|
|
|
143 self.score = 0
|
|
|
144 self.contig_match_start = 0
|
|
|
145 self.contig_match_end = 0
|
|
|
146 self.row = ""
|
|
|
147
|
|
|
148 class mobsuitePlasmids(object):
|
|
|
149 def __init__(self):
|
|
|
150 self.file_id = ""
|
|
|
151 self.num_contigs = 0
|
|
|
152 self.total_length = 0
|
|
|
153 self.gc = ""
|
|
|
154 self.rep_types = ""
|
|
|
155 self.rep_typeAccession = ""
|
|
|
156 self.relaxase_type= ""
|
|
|
157 self.relaxase_type_accession = ""
|
|
|
158 self.mpf_type = ""
|
|
|
159 self.mpf_type_accession= ""
|
|
|
160 self.orit_type = ""
|
|
|
161 self.orit_accession = ""
|
|
|
162 self.PredictedMobility = ""
|
|
|
163 self.mash_nearest_neighbor = ""
|
|
|
164 self.mash_neighbor_distance = 0.00
|
|
|
165 self.mash_neighbor_cluster= 0
|
|
|
166 self.row = ""
|
|
|
167 class RGIResult(object):
|
|
|
168 def __init__(self):
|
|
|
169 self.ORF_ID = ""
|
|
|
170 self.Contig = ""
|
|
|
171 self.Start = -1
|
|
|
172 self.Stop = -1
|
|
|
173 self.Orientation = ""
|
|
|
174 self.Cut_Off = ""
|
|
|
175 self.Pass_Bitscore = 100000
|
|
|
176 self.Best_Hit_Bitscore = 0.00
|
|
|
177 self.Best_Hit_ARO = ""
|
|
|
178 self.Best_Identities = 0.00
|
|
|
179 self.ARO = 0
|
|
|
180 self.Model_type = ""
|
|
|
181 self.SNPs_in_Best_Hit_ARO = ""
|
|
|
182 self.Other_SNPs = ""
|
|
|
183 self.Drug_Class = ""
|
|
|
184 self.Resistance_Mechanism = ""
|
|
|
185 self.AMR_Gene_Family = ""
|
|
|
186 self.Predicted_DNA = ""
|
|
|
187 self.Predicted_Protein = ""
|
|
|
188 self.CARD_Protein_Sequence = ""
|
|
|
189 self.Percentage_Length_of_Reference_Sequence = 0.00
|
|
|
190 self.ID = ""
|
|
|
191 self.Model_ID = 0
|
|
|
192 self.source = ""
|
|
|
193 self.row = ""
|
|
|
194
|
|
|
195 #endregion
|
|
|
196
|
|
|
197 #region useful functions
|
|
|
198 def read(path):
|
|
|
199 return [line.rstrip('\n') for line in open(path)]
|
|
|
200 def execute(command):
|
|
|
201 process = subprocess.Popen(command, shell=False, cwd=curDir, universal_newlines=True, stdout=subprocess.PIPE, stderr=subprocess.STDOUT)
|
|
|
202
|
|
|
203 # Poll process for new output until finished
|
|
|
204 while True:
|
|
|
205 nextline = process.stdout.readline()
|
|
|
206 if nextline == '' and process.poll() is not None:
|
|
|
207 break
|
|
|
208 sys.stdout.write(nextline)
|
|
|
209 sys.stdout.flush()
|
|
|
210
|
|
|
211 output = process.communicate()[0]
|
|
|
212 exitCode = process.returncode
|
|
|
213
|
|
|
214 if (exitCode == 0):
|
|
|
215 return output
|
|
|
216 else:
|
|
|
217 raise subprocess.CalledProcessError(exitCode, command)
|
|
|
218 def httpGetFile(url, filepath=""):
|
|
|
219 if (filepath == ""):
|
|
|
220 return urllib.request.urlretrieve(url)
|
|
|
221 else:
|
|
|
222 urllib.request.urlretrieve(url, filepath)
|
|
|
223 return True
|
|
|
224 def gunzip(inputpath="", outputpath=""):
|
|
|
225 if (outputpath == ""):
|
|
|
226 with gzip.open(inputpath, 'rb') as f:
|
|
|
227 gzContent = f.read()
|
|
|
228 return gzContent
|
|
|
229 else:
|
|
|
230 with gzip.open(inputpath, 'rb') as f:
|
|
|
231 gzContent = f.read()
|
|
|
232 with open(outputpath, 'wb') as out:
|
|
|
233 out.write(gzContent)
|
|
|
234 return True
|
|
|
235 def ToJson(dictObject, outputPath):
|
|
|
236 outDir = outputDir + '/summary/' + ID + ".json/"
|
|
|
237 if not (os.path.exists(outDir)):
|
|
|
238 os.makedirs(outDir)
|
|
|
239 with open(outDir + outputPath, 'w') as f:
|
|
|
240 json.dump([ob.__dict__ for ob in dictObject.values()], f, ensure_ascii=False)
|
|
|
241 #endregion
|
|
|
242
|
|
|
243 #region functions to parse result files
|
|
|
244 def ParseMLSTResult(pathToMLSTResult):
|
|
|
245 _mlstResult = {}
|
|
|
246 scheme = pandas.read_csv(mlstScheme, delimiter='\t', header=0)
|
|
|
247 scheme = scheme.replace(numpy.nan, '', regex=True)
|
|
|
248
|
|
|
249 taxon = {}
|
|
|
250 #record the scheme as a dictionary
|
|
|
251 taxon["-"] = "No MLST Match"
|
|
|
252 for i in range(len(scheme.index)):
|
|
|
253 key = scheme.iloc[i,0]
|
|
|
254 if (str(scheme.iloc[i,2]) == "nan"):
|
|
|
255 value = str(scheme.iloc[i,1])
|
|
|
256 else:
|
|
|
257 value = str(scheme.iloc[i,1]) + " " + str(scheme.iloc[i,2])
|
|
|
258
|
|
|
259 if (key in taxon.keys()):
|
|
|
260 taxon[key] = taxon.get(key) + ";" + value
|
|
|
261 else:
|
|
|
262 taxon[key] = value
|
|
|
263 #read in the mlst result
|
|
|
264 mlst = pandas.read_csv(pathToMLSTResult, delimiter='\t', header=None)
|
|
|
265 _mlstHit = MlstResult()
|
|
|
266
|
|
|
267 _mlstHit.file = mlst.iloc[0,0]
|
|
|
268 _mlstHit.speciesID = (mlst.iloc[0,1])
|
|
|
269 _mlstHit.seqType = str(mlst.iloc[0,2])
|
|
|
270 for i in range(3, len(mlst.columns)):
|
|
|
271 _mlstHit.scheme += mlst.iloc[0,i] + ";"
|
|
|
272 _mlstHit.species = taxon[_mlstHit.speciesID]
|
|
|
273 _mlstHit.row = "\t".join(str(x) for x in mlst.ix[0].tolist())
|
|
|
274 _mlstResult[_mlstHit.speciesID]=_mlstHit
|
|
|
275
|
|
|
276 return _mlstResult
|
|
|
277
|
|
|
278 def ParsePlasmidFinderResult(pathToPlasmidFinderResult):
|
|
|
279 #pipelineTest/contigs/BC110-Kpn005.fa contig00019 45455 45758 IncFIC(FII)_1 8-308/499 ========/=..... 8/11 59.52 75.65 plasmidfinder AP001918 IncFIC(FII)_1__AP001918
|
|
|
280 #example resfinder:
|
|
|
281 #pipelineTest/contigs/BC110-Kpn005.fa contig00038 256 1053 OXA-181 1-798/798 =============== 0/0 100.00 100.00 bccdc AEP16366.1 OXA-48 family carbapenem-hydrolyzing class D beta-lactamase OXA-181
|
|
|
282
|
|
|
283 _pFinder = {} #***********************
|
|
|
284 plasmidFinder = pandas.read_csv(pathToPlasmidFinderResult, delimiter='\t', header=0)
|
|
|
285 plasmidFinder = plasmidFinder.replace(numpy.nan, '', regex=True)
|
|
|
286
|
|
|
287
|
|
|
288 for i in range(len(plasmidFinder.index)):
|
|
|
289 pf = starFinders()
|
|
|
290 pf.file = str(plasmidFinder.iloc[i,0])
|
|
|
291 pf.sequence = str(plasmidFinder.iloc[i,1])
|
|
|
292 pf.start = int(plasmidFinder.iloc[i,2])
|
|
|
293 pf.end = int(plasmidFinder.iloc[i,3])
|
|
|
294 pf.gene = str(plasmidFinder.iloc[i,4])
|
|
|
295 pf.shortGene = pf.gene[:pf.gene.index("_")]
|
|
|
296 pf.coverage = str(plasmidFinder.iloc[i,5])
|
|
|
297 pf.coverage_map = str(plasmidFinder.iloc[i,6])
|
|
|
298 pf.gaps = str(plasmidFinder.iloc[i,7])
|
|
|
299 pf.pCoverage = float(plasmidFinder.iloc[i,8])
|
|
|
300 pf.pIdentity = float(plasmidFinder.iloc[i,9])
|
|
|
301 pf.database = str(plasmidFinder.iloc[i,10])
|
|
|
302 pf.accession = str(plasmidFinder.iloc[i,11])
|
|
|
303 pf.product = str(plasmidFinder.iloc[i,12])
|
|
|
304 pf.source = "plasmid"
|
|
|
305 pf.row = "\t".join(str(x) for x in plasmidFinder.ix[i].tolist())
|
|
|
306 _pFinder[pf.gene]=pf
|
|
|
307 #row = "\t".join(str(x) for x in plasmidFinder.ix[i].tolist())
|
|
|
308 #plasmidFinderContigs.append(str(plasmidFinder.iloc[i,1]))
|
|
|
309 #origins.append(str(plasmidFinder.iloc[i,4][:plasmidFinder.iloc[i,4].index("_")]))
|
|
|
310 return _pFinder
|
|
|
311
|
|
|
312 def ParseMobsuiteResult(pathToMobsuiteResult):
|
|
|
313 _mobsuite = {}
|
|
|
314 mResult = pandas.read_csv(pathToMobsuiteResult, delimiter='\t', header=0)
|
|
|
315 mResult = mResult.replace(numpy.nan, '', regex=True)
|
|
|
316
|
|
|
317 for i in range(len(mResult.index)):
|
|
|
318 mr = mobsuiteResult()
|
|
|
319 mr.file_id = str(mResult.iloc[i,0])
|
|
|
320 mr.cluster_id = str(mResult.iloc[i,1])
|
|
|
321 if (mr.cluster_id == "chromosome"):
|
|
|
322 break
|
|
|
323 mr.contig_id = str(mResult.iloc[i,2])
|
|
|
324 mr.contig_num = mr.contig_id[(mr.contig_id.find("contig")+6):mr.contig_id.find("_len=")]
|
|
|
325 mr.contig_length = int(mResult.iloc[i,3])
|
|
|
326 mr.circularity_status = str(mResult.iloc[i,4])
|
|
|
327 mr.rep_type = str(mResult.iloc[i,5])
|
|
|
328 mr.rep_type_accession = str(mResult.iloc[i,6])
|
|
|
329 mr.relaxase_type = str(mResult.iloc[i,7])
|
|
|
330 mr.relaxase_type_accession = str(mResult.iloc[i,8])
|
|
|
331 mr.mash_nearest_neighbor = str(mResult.iloc[i,9])
|
|
|
332 mr.mash_neighbor_distance = float(mResult.iloc[i,10])
|
|
|
333 mr.repetitive_dna_id = str(mResult.iloc[i,11])
|
|
|
334 mr.match_type = str(mResult.iloc[i,12])
|
|
|
335 if (mr.match_type == ""):
|
|
|
336 mr.score = -1
|
|
|
337 mr.contig_match_start = -1
|
|
|
338 mr.contig_match_end = -1
|
|
|
339 else:
|
|
|
340 mr.score = int(mResult.iloc[i,13])
|
|
|
341 mr.contig_match_start = int(mResult.iloc[i,14])
|
|
|
342 mr.contig_match_end = int(mResult.iloc[i,15])
|
|
|
343 mr.row = "\t".join(str(x) for x in mResult.ix[i].tolist())
|
|
|
344 _mobsuite[mr.contig_id]=(mr)
|
|
|
345 return _mobsuite
|
|
|
346
|
|
|
347 def ParseMobsuitePlasmids(pathToMobsuiteResult):
|
|
|
348 _mobsuite = {}
|
|
|
349 mResults = pandas.read_csv(pathToMobsuiteResult, delimiter='\t', header=0)
|
|
|
350 mResults = mResults.replace(numpy.nan, '', regex=True)
|
|
|
351
|
|
|
352 for i in range(len(mResults.index)):
|
|
|
353 mr = mobsuitePlasmids()
|
|
|
354 mr.file_id = str(mResults.iloc[i,0])
|
|
|
355 mr.num_contigs = int(mResults.iloc[i,1])
|
|
|
356 mr.total_length = int(mResults.iloc[i,2])
|
|
|
357 mr.gc = int(mResults.iloc[i,3])
|
|
|
358 mr.rep_types = str(mResults.iloc[i,4])
|
|
|
359 mr.rep_typeAccession = str(mResults.iloc[i,5])
|
|
|
360 mr.relaxase_type = str(mResults.iloc[i,6])
|
|
|
361 mr.relaxase_type_accession = str(mResults.iloc[i,7])
|
|
|
362 mr.mpf_type = str(mResults.iloc[i,8])
|
|
|
363 mr.mpf_type_accession = str(mResults.iloc[i,9])
|
|
|
364 mr.orit_type = str(mResults.iloc[i,10])
|
|
|
365 mr.orit_accession = str(mResults.iloc[i,11])
|
|
|
366 mr.PredictedMobility = str(mResults.iloc[i,12])
|
|
|
367 mr.mash_nearest_neighbor = str(mResults.iloc[i,13])
|
|
|
368 mr.mash_neighbor_distance = float(mResults.iloc[i,14])
|
|
|
369 mr.mash_neighbor_cluster = int(mResults.iloc[i,15])
|
|
|
370 mr.row = "\t".join(str(x) for x in mResults.ix[i].tolist())
|
|
|
371 _mobsuite[mr.file_id] = mr
|
|
|
372 return _mobsuite
|
|
|
373
|
|
|
374 def ParseResFinderResult(pathToResFinderResults, plasmidContigs, likelyPlasmidContigs):
|
|
|
375 _rFinder = {}
|
|
|
376 resFinder = pandas.read_csv(pathToResFinderResults, delimiter='\t', header=0)
|
|
|
377 resFinder = resFinder.replace(numpy.nan, '', regex=True)
|
|
|
378
|
|
|
379 for i in range(len(resFinder.index)):
|
|
|
380 rf = starFinders()
|
|
|
381 rf.file = str(resFinder.iloc[i,0])
|
|
|
382 rf.sequence = str(resFinder.iloc[i,1])
|
|
|
383 rf.start = int(resFinder.iloc[i,2])
|
|
|
384 rf.end = int(resFinder.iloc[i,3])
|
|
|
385 rf.gene = str(resFinder.iloc[i,4])
|
|
|
386 rf.shortGene = rf.gene
|
|
|
387 rf.coverage = str(resFinder.iloc[i,5])
|
|
|
388 rf.coverage_map = str(resFinder.iloc[i,6])
|
|
|
389 rf.gaps = str(resFinder.iloc[i,7])
|
|
|
390 rf.pCoverage = float(resFinder.iloc[i,8])
|
|
|
391 rf.pIdentity = float(resFinder.iloc[i,9])
|
|
|
392 rf.database = str(resFinder.iloc[i,10])
|
|
|
393 rf.accession = str(resFinder.iloc[i,11])
|
|
|
394 rf.product = str(resFinder.iloc[i,12])
|
|
|
395 rf.row = "\t".join(str(x) for x in resFinder.ix[i].tolist())
|
|
|
396 if (rf.sequence[6:] in plasmidContigs):
|
|
|
397 rf.source = "plasmid"
|
|
|
398 elif (rf.sequence[6:] in likelyPlasmidContigs):
|
|
|
399 rf.source = "likely plasmid"
|
|
|
400 else:
|
|
|
401 rf.source = "likely chromosome"
|
|
|
402 _rFinder[rf.gene]=rf
|
|
|
403 return _rFinder
|
|
|
404
|
|
|
405 def ParseRGIResult(pathToRGIResults, plasmidContigs, likelyPlasmidContigs):
|
|
|
406 _rgiR = {}
|
|
|
407 RGI = pandas.read_csv(pathToRGIResults, delimiter='\t', header=0)
|
|
|
408 RGI = RGI.replace(numpy.nan, '', regex=True)
|
|
|
409
|
|
|
410 for i in range(len(RGI.index)):
|
|
|
411 r = RGIResult()
|
|
|
412 r.ORF_ID = str(RGI.iloc[i,0])
|
|
|
413 r.Contig = str(RGI.iloc[i,1])
|
|
|
414 r.Contig_Num = r.Contig[6:r.Contig.find("_")]
|
|
|
415 r.Start = int(RGI.iloc[i,2])
|
|
|
416 r.Stop = int(RGI.iloc[i,3])
|
|
|
417 r.Orientation = str(RGI.iloc[i,4])
|
|
|
418 r.Cut_Off = str(RGI.iloc[i,5])
|
|
|
419 r.Pass_Bitscore = int(RGI.iloc[i,6])
|
|
|
420 r.Best_Hit_Bitscore = float(RGI.iloc[i,7])
|
|
|
421 r.Best_Hit_ARO = str(RGI.iloc[i,8])
|
|
|
422 r.Best_Identities = float(RGI.iloc[i,9])
|
|
|
423 r.ARO = int(RGI.iloc[i,10])
|
|
|
424 r.Model_type = str(RGI.iloc[i,11])
|
|
|
425 r.SNPs_in_Best_Hit_ARO = str(RGI.iloc[i,12])
|
|
|
426 r.Other_SNPs = str(RGI.iloc[i,13])
|
|
|
427 r.Drug_Class = str(RGI.iloc[i,14])
|
|
|
428 r.Resistance_Mechanism = str(RGI.iloc[i,15])
|
|
|
429 r.AMR_Gene_Family = str(RGI.iloc[i,16])
|
|
|
430 r.Predicted_DNA = str(RGI.iloc[i,17])
|
|
|
431 r.Predicted_Protein = str(RGI.iloc[i,18])
|
|
|
432 r.CARD_Protein_Sequence = str(RGI.iloc[i,19])
|
|
|
433 r.Percentage_Length_of_Reference_Sequence = float(RGI.iloc[i,20])
|
|
|
434 r.ID = str(RGI.iloc[i,21])
|
|
|
435 r.Model_ID = int(RGI.iloc[i,22])
|
|
|
436 r.row = "\t".join(str(x) for x in RGI.ix[i].tolist())
|
|
|
437 if (r.Contig_Num in plasmidContigs):
|
|
|
438 r.source = "plasmid"
|
|
|
439 elif (r.Contig_Num in likelyPlasmidContigs):
|
|
|
440 r.source = "likely plasmid"
|
|
|
441 else:
|
|
|
442 r.source = "likely chromosome"
|
|
|
443 _rgiR[r.Model_ID]=r
|
|
|
444 return _rgiR
|
|
|
445 #endregion
|
|
|
446
|
|
|
447 def Main():
|
|
|
448 notes = []
|
|
|
449 #init the output list
|
|
|
450 output = []
|
|
|
451 jsonOutput = []
|
|
|
452
|
|
|
453 print(str(datetime.datetime.now()) + "\n\nID: " + ID + "\nAssembly: " + assemblyPath)
|
|
|
454 output.append(str(datetime.datetime.now()) + "\n\nID: " + ID + "\nAssembly: " + assemblyPath)
|
|
|
455
|
|
|
456 #region update databases if update=true
|
|
|
457 if not debug:
|
|
|
458 #update databases if necessary
|
|
|
459 if not (updateAbPath == "" and updateAbName == "default"):
|
|
|
460 print("updating abricate database: " + updateAbName + " @fasta path: " + updateAbPath)
|
|
|
461 cmd = [scriptDir + "/pipeline_updateAbricateDB.sh", updateAbPath, updateAbName]
|
|
|
462 update = execute(cmd)
|
|
|
463 if (updateMLST.lower() == "true"):
|
|
|
464 print("updating mlst database... ")
|
|
|
465 cmd = [scriptDir + "/pipeline_updateMLST.sh"]
|
|
|
466 update = execute(cmd)
|
|
|
467 #endregion
|
|
|
468
|
|
|
469 print("step 3: parsing the mlst results")
|
|
|
470
|
|
|
471 print("performing MLST")
|
|
|
472 #region call the mlst script
|
|
|
473 if not debug:
|
|
|
474 print("running pipeline_prediction.sh")
|
|
|
475 #input parameters: 1=ID 2 = assemblyPath, 3= outputdir, 4=card.json
|
|
|
476 cmd = [scriptDir + "/pipeline_prediction.sh", ID, assemblyPath, outputDir, cardDB]
|
|
|
477 result = execute(cmd)
|
|
|
478 #endregion
|
|
|
479
|
|
|
480 #region parse the mlst results
|
|
|
481 print("step 3: parsing mlst, plasmid, and amr results")
|
|
|
482
|
|
|
483 print("identifying MLST")
|
|
|
484 pathToMLSTScheme = outputDir + "/predictions/" + ID + ".mlst"
|
|
|
485 mlstHit = ParseMLSTResult(pathToMLSTScheme)#***********************
|
|
|
486 ToJson(mlstHit, "mlst.json") #write it to a json output
|
|
|
487 mlstHit = list(mlstHit.values())[0]
|
|
|
488
|
|
|
489 #endregion
|
|
|
490
|
|
|
491 #region parse mobsuite, resfinder and rgi results
|
|
|
492 print("identifying plasmid contigs and amr genes")
|
|
|
493
|
|
|
494 plasmidContigs = []
|
|
|
495 likelyPlasmidContigs = []
|
|
|
496 origins = []
|
|
|
497
|
|
|
498 #parse mobsuite results
|
|
|
499 mSuite = ParseMobsuiteResult(outputDir + "/predictions/" + ID + ".recon/contig_report.txt")#*************
|
|
|
500 ToJson(mSuite, "mobsuite.json") #*************
|
|
|
501 mSuitePlasmids = ParseMobsuitePlasmids(outputDir + "/predictions/" + ID + ".recon/mobtyper_aggregate_report.txt")#*************
|
|
|
502 ToJson(mSuitePlasmids, "mobsuitePlasmids.json") #*************
|
|
|
503
|
|
|
504 for key in mSuite:
|
|
|
505 if mSuite[key].contig_num not in plasmidContigs and mSuite[key].contig_num not in likelyPlasmidContigs:
|
|
|
506 if not (mSuite[key].rep_type == ''):
|
|
|
507 plasmidContigs.append(mSuite[key].contig_num)
|
|
|
508 else:
|
|
|
509 likelyPlasmidContigs.append(mSuite[key].contig_num)
|
|
|
510 for key in mSuite:
|
|
|
511 if mSuite[key].rep_type not in origins:
|
|
|
512 origins.append(mSuite[key].rep_type)
|
|
|
513
|
|
|
514 #parse resfinder AMR results
|
|
|
515 rFinder = ParseResFinderResult(outputDir + "/predictions/" + ID + ".cp", plasmidContigs, likelyPlasmidContigs) #**********************
|
|
|
516 ToJson(rFinder, "resfinder.json") #*************
|
|
|
517
|
|
|
518 rgiAMR = ParseRGIResult(outputDir + "/predictions/" + ID + ".rgi.txt", plasmidContigs, likelyPlasmidContigs)#***********************
|
|
|
519 ToJson(rgiAMR, "rgi.json") #*************
|
|
|
520
|
|
|
521 carbapenamases = []
|
|
|
522 amrGenes = []
|
|
|
523 for keys in rFinder:
|
|
|
524 carbapenamases.append(rFinder[keys].shortGene + "(" + rFinder[keys].source + ")")
|
|
|
525 for keys in rgiAMR:
|
|
|
526 if (rgiAMR[keys].Drug_Class.find("carbapenem") > -1):
|
|
|
527 if (rgiAMR[keys].Best_Hit_ARO not in carbapenamases):
|
|
|
528 carbapenamases.append(rgiAMR[keys].Best_Hit_ARO+ "(" + rgiAMR[keys].source + ")")
|
|
|
529 else:
|
|
|
530 if (rgiAMR[keys].Best_Hit_ARO not in amrGenes):
|
|
|
531 amrGenes.append(rgiAMR[keys].Best_Hit_ARO+ "(" + rgiAMR[keys].source + ")")
|
|
|
532 #endregion
|
|
|
533
|
|
|
534 #region output parsed mlst information
|
|
|
535 print("formatting mlst outputs")
|
|
|
536 output.append("\n\n\n~~~~~~~MLST summary~~~~~~~")
|
|
|
537 output.append("MLST determined species: " + mlstHit.species)
|
|
|
538 output.append("\nMLST Details: ")
|
|
|
539 output.append(mlstHit.row)
|
|
|
540
|
|
|
541 output.append("\nMLST information: ")
|
|
|
542 if (mlstHit.species == expectedSpecies):
|
|
|
543 output.append("MLST determined species is the same as expected species")
|
|
|
544 #notes.append("MLST determined species is the same as expected species")
|
|
|
545 else:
|
|
|
546 output.append("!!!MLST determined species is NOT the same as expected species, contamination? mislabeling?")
|
|
|
547 notes.append("MLST: Not expected species. Possible contamination or mislabeling")
|
|
|
548
|
|
|
549 #endregion
|
|
|
550
|
|
|
551 #region output the parsed plasmid/amr results
|
|
|
552 output.append("\n\n\n~~~~~~~~Plasmids~~~~~~~~\n")
|
|
|
553
|
|
|
554 output.append("predicted plasmid origins: ")
|
|
|
555 output.append(";".join(origins))
|
|
|
556
|
|
|
557 output.append("\ndefinitely plasmid contigs")
|
|
|
558 output.append(";".join(plasmidContigs))
|
|
|
559
|
|
|
560 output.append("\nlikely plasmid contigs")
|
|
|
561 output.append(";".join(likelyPlasmidContigs))
|
|
|
562
|
|
|
563 output.append("\nmob-suite prediction details: ")
|
|
|
564 for key in mSuite:
|
|
|
565 output.append(mSuite[key].row)
|
|
|
566
|
|
|
567 output.append("\n\n\n~~~~~~~~AMR Genes~~~~~~~~\n")
|
|
|
568 output.append("predicted carbapenamase Genes: ")
|
|
|
569 output.append(",".join(carbapenamases))
|
|
|
570 output.append("other RGI AMR Genes: ")
|
|
|
571 for key in rgiAMR:
|
|
|
572 output.append(rgiAMR[key].Best_Hit_ARO + "(" + rgiAMR[key].source + ")")
|
|
|
573
|
|
|
574 output.append("\nDetails about the carbapenamase Genes: ")
|
|
|
575 for key in rFinder:
|
|
|
576 output.append(rFinder[key].row)
|
|
|
577 output.append("\nDetails about the RGI AMR Genes: ")
|
|
|
578 for key in rgiAMR:
|
|
|
579 output.append(rgiAMR[key].row)
|
|
|
580
|
|
|
581 #write summary to a file
|
|
|
582 summaryDir = outputDir + "/summary/" + ID
|
|
|
583 out = open(summaryDir + ".txt", 'w')
|
|
|
584 for item in output:
|
|
|
585 out.write("%s\n" % item)
|
|
|
586
|
|
|
587
|
|
|
588 #TSV output
|
|
|
589 tsvOut = []
|
|
|
590 tsvOut.append("ID\tExpected Species\tMLST Species\tSequence Type\tMLST Scheme\tCarbapenem Resistance Genes\tOther AMR Genes\tTotal Plasmids\tPlasmids ID\tNum_Contigs\tPlasmid Length\tPlasmid RepType\tPlasmid Mobility\tNearest Reference\tDefinitely Plasmid Contigs\tLikely Plasmid Contigs")
|
|
|
591 #start with ID
|
|
|
592 temp = ""
|
|
|
593 temp += (ID + "\t")
|
|
|
594 temp += expectedSpecies + "\t"
|
|
|
595
|
|
|
596 #move into MLST
|
|
|
597 temp += mlstHit.species + "\t"
|
|
|
598 temp += str(mlstHit.seqType) + "\t"
|
|
|
599 temp += mlstHit.scheme + "\t"
|
|
|
600
|
|
|
601 #now onto AMR genes
|
|
|
602 temp += ";".join(carbapenamases) + "\t"
|
|
|
603 temp += ";".join(amrGenes) + "\t"
|
|
|
604
|
|
|
605 #lastly plasmids
|
|
|
606 temp+= str(len(mSuitePlasmids)) + "\t"
|
|
|
607 plasmidID = ""
|
|
|
608 contigs = ""
|
|
|
609 lengths = ""
|
|
|
610 rep_type = ""
|
|
|
611 mobility = ""
|
|
|
612 neighbour = ""
|
|
|
613 for keys in mSuitePlasmids:
|
|
|
614 plasmidID += str(mSuitePlasmids[keys].mash_neighbor_cluster) + ";"
|
|
|
615 contigs += str(mSuitePlasmids[keys].num_contigs) + ";"
|
|
|
616 lengths += str(mSuitePlasmids[keys].total_length) + ";"
|
|
|
617 rep_type += str(mSuitePlasmids[keys].rep_types) + ";"
|
|
|
618 mobility += str(mSuitePlasmids[keys].PredictedMobility) + ";"
|
|
|
619 neighbour += str(mSuitePlasmids[keys].mash_nearest_neighbor) + ";"
|
|
|
620 temp += plasmidID + "\t" + contigs + "\t" + lengths + "\t" + rep_type + "\t" + mobility + "\t" + neighbour + "\t"
|
|
|
621 temp += ";".join(plasmidContigs) + "\t"
|
|
|
622 temp += ";".join(likelyPlasmidContigs)
|
|
|
623 tsvOut.append(temp)
|
|
|
624
|
|
|
625 summaryDir = outputDir + "/summary/" + ID
|
|
|
626 out = open(summaryDir + ".tsv", 'w')
|
|
|
627 for item in tsvOut:
|
|
|
628 out.write("%s\n" % item)
|
|
|
629 #endregion
|
|
|
630
|
|
|
631
|
|
|
632 start = time.time()#time the analysis
|
|
|
633 print("Starting workflow...")
|
|
|
634 #analysis time
|
|
|
635 Main()
|
|
|
636
|
|
|
637 end = time.time()
|
|
2
|
638 print("Finished!\nThe analysis used: " + str(end-start) + " seconds")
|
