-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathexomedepth_and_arrays.py
310 lines (204 loc) · 13.6 KB
/
exomedepth_and_arrays.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
'''
ALBA MALAGON MARQUEZ
Final Degree Project: 'STRATEGY FOR ACCURATE CNV DETECTION AND ML ALGORITHM FOR CLASSIFYING NGS VARIANTS'
Script that filters CNVs from Exome Depth and arrays with enough coverage, and stores the coincident and not coincident calls between ExomeDepth and arrays.
USAGE: python3 exomedepth_and_arrays.py
'''
from CNV_libraries_variables import *
starttime = timeit.default_timer()
# D E F I N I N G V A R I A B L E S ----------------------------------------------------------------------
alldataGFval = GFval
alldataEDval = EDval
alldataarray = array
# R E A D I N G F I L E S ----------------------------------------------------------------------------------
#reading arrays designs
design60_37 = pd.read_csv(design[0], sep='\t')
design60_24 = pd.read_csv(design[1], sep='\t')
designPrePost = pd.read_csv(design[2], sep='\t')
design400 = pd.read_csv(design[3], sep='\t')
design180 = pd.read_csv(design[4], sep='\t')
design1M = pd.read_csv(design[5], sep='\t')
#reading samples with array and exoma
exomaa = pd.read_csv(exoma[0],sep=';')
#reading the pseudogenes dataset
pseudogenes = pd.read_csv(pseudogenes[0], sep='\t')
#reading the segmental duplications
segdups = pd.read_csv(SD[0], sep='\t')
#reading exoma data
capexoma = pd.read_csv(captura_ex[0],sep='\t')
color_print("### GENERATING THE WHOLE DATASET ###", color='red')
y=1
# creating an empty dataset where all Ok.Debayan files will be added
column_names = ['chr', 'start', 'end', 'width', 'type', 'nexons', 'BF',
'reads.expected', 'reads.observed', 'reads.ratio', 'SegDup',
'exons.hg19', 'RefSeqGenes', 'OMIM', 'OMIMFeb2017', 'NqExDels',
'NqExDups', 'qExDels', 'qExDups','Sample']
dataset = pd.DataFrame(columns=column_names)
#Adding ExomeDepth CNVs data
for path in alldataEDval:
color_print('Adding EXOME DEPTH data {} ({}/{})'.format(path.rsplit('/')[-1].rsplit('_')[0], y, len(alldataEDval)), color='yellow')
variable_length=len(dataset)
dataset = dataset.append(pd.read_csv(str(path), sep="\t"),ignore_index=True, sort=False) #raises and error if the file is open!! BE CAREFUL!
dataset.loc[variable_length:len(dataset),['Sample']] = path.rsplit('/')[-1].rsplit('_')[0]
y+=1
dataset.to_csv('{}/datasetCNV_VALIDATION.csv'.format(os.getcwd()), header=True, sep="\t")
dataset = pd.read_csv('{}/datasetCNV_VALIDATION.csv'.format(os.getcwd()), sep="\t")
dataset.drop(['OMIM', 'OMIMFeb2017', 'NqExDels',
'NqExDups', 'qExDels', 'qExDups', 'Unnamed: 0'], axis=1, inplace=True)
# C O V E R E D C N V S E X O M E D E P T H -------------------------------------------------------------
#creating three new columns to keep track of the probes, coverage and the array type of each CNV
dataset['PROBES'] = [0] * len(dataset)
dataset['COVERAGE'] = 'False'
dataset["array_type"] = ['.'] * len(dataset)
# analysing each CNV individually
for index, cnv in dataset.iterrows():
chro = cnv['chr']
start = cnv['start']
end = cnv['end']
typee = cnv['type']
allchr='chr'+str(cnv['chr'])
#adding the array type corresponding to the current sample
array=exomaa[(exomaa['Mostra'] == dataset['Sample'][index])]
dataset['array_type'][index] = array['Tipus_prova_array'].iloc[0]
#extracting the number of probes the CNV contains taking into account its design type
if (dataset['array_type'][index] == 'qChip Post') or (dataset['array_type'][index] == 'qChip Pre v1.1 Complete') or (dataset['array_type'][index] == 'qChip Pre v1.1 Targeted'):
num_probes=designPrePost[(start<=designPrePost["start"]) & (end>=designPrePost["end"]) & (allchr==designPrePost["chr"])]
if dataset['array_type'][index] == 'qChip 400K':
num_probes=design400[(start<=design400["start"]) & (end>=design400["end"]) & (allchr==design400["chr"])]
if dataset['array_type'][index] == 'qChip 1M':
num_probes=design1M[(start<=design1M["start"]) & (end>=design1M["end"]) & (allchr==design1M["chr"])]
if dataset['array_type'][index] == 'qChip 180K':
num_probes=design180[(start<=design180["start"]) & (end>=design180["end"]) & (allchr==design180["chr"])]
if dataset['array_type'][index] == 'qChip Exon 8x60K (085824)':
num_probes=design60_24[(start<=design60_24["start"]) & (end>=design60_24["end"]) & (allchr==design60_24["chr"])]
if dataset['array_type'][index] == 'qChip Exon 8x60K (085337)':
num_probes=design60_37[(start<=design60_37["start"]) & (end>=design60_37["end"]) & (allchr==design60_37["chr"])]
if len(num_probes)>0:
#adding the number of probes
dataset['PROBES'][index] = str(len(num_probes))
#assigning the coverage to True if the CNV has a minimum of three probes
if len(num_probes) >= 3:
dataset['COVERAGE'][index] = 'True'
#filtering only those CNVs covered
datasetCOVERED = pd.DataFrame(dataset, columns=dataset.columns)
datasetCOVERED = datasetCOVERED[~datasetCOVERED.COVERAGE.str.contains('False')]
datasetCOVERED.reset_index(inplace=True)
dataset.to_csv('{}/datasetCNV_VALIDATION_v2.csv'.format(os.getcwd()), header=True, sep="\t")
datasetCOVERED.to_csv('{}/datasetCNV_VALIDATION_COVERED.csv'.format(os.getcwd()), header=True, sep="\t")
# generating individual datasets with CNVs covered per sample
individual = pd.DataFrame(columns=datasetCOVERED.columns)
for sample in datasetCOVERED['Sample'].unique():
individual=individual.append(datasetCOVERED[datasetCOVERED['Sample']==sample],ignore_index=True, sort=False)
individual.to_csv('/Users/alba/Desktop/QGEN/cnv/EDcobertesIND/{}_EDcob.csv'.format(sample), header=True, sep="\t")
individual = pd.DataFrame(columns=datasetCOVERED.columns)
# C O V E R E D C N V S F R O M A R R A Y S -------------------------------------------------------------
color_print("### GENERATING THE WHOLE DATASET ###", color='red')
y=1
# creating an empty dataset where all CNVs from arrays files will be added
column_names = ['chrom','band','start','end','size','probes','type','genes','CNV %','Comments','Sample']
datasetCNVsARRAYS = pd.DataFrame(columns=column_names)
#Adding arrays CNVs data
for path in alldataarray:
color_print('Adding ARRAY data {} ({}/{})'.format(path.rsplit('/')[-1].rsplit('_')[2], y, len(alldataarray)), color='cyan')
variable_length=len(datasetCNVsARRAYS)
datasetCNVsARRAYS = datasetCNVsARRAYS.append(pd.read_csv(str(path), sep="\t"),ignore_index=True, sort=False) #raises and error if the file is open!! BE CAREFUL!
datasetCNVsARRAYS.loc[variable_length:len(datasetCNVsARRAYS),['Sample']] = path.rsplit('/')[-1].rsplit('_')[2]
y+=1
datasetCNVsARRAYS.to_csv('{}/datasetCNVsARRAYS.csv'.format(os.getcwd()), header=True, sep="\t")
#creating three new columns to keep track of the number of exons and the coverage of each CNV
datasetCNVsARRAYS['numexons'] = [0] * len(datasetCNVsARRAYS)
datasetCNVsARRAYS['COVERAGE'] = 'False'
#analysing each CNV individually
for i, cnva in datasetCNVsARRAYS.iterrows():
chro = cnva['chrom']
start = cnva['start']
end = cnva['end']
typee = cnva['type']
#extracting the number of exons the CNV contains from the exoma dataset
numexons = capexoma[(start<=capexoma["start"]) & (end>=capexoma["end"]) & (chro==capexoma["chr"])]
if len(numexons)>0:
#adding the number of exons and assigning the coverage to True if it has at least one exon
datasetCNVsARRAYS['numexons'][i] = str(len(numexons))
datasetCNVsARRAYS['COVERAGE'][i] = 'True'
#filtering only those CNVs covered
datasetCNVsARRAYSCOVERED = pd.DataFrame(datasetCNVsARRAYS, columns=datasetCNVsARRAYS.columns)
datasetCNVsARRAYSCOVERED = datasetCNVsARRAYSCOVERED[~datasetCNVsARRAYSCOVERED.COVERAGE.str.contains('False')]
datasetCNVsARRAYSCOVERED.reset_index(inplace=True)
datasetCNVsARRAYS.to_csv('{}/datasetCNVsARRAYS_v2.csv'.format(os.getcwd()), header=True, sep="\t")
datasetCNVsARRAYSCOVERED.to_csv('{}/datasetCNVsARRAYS_COVERED.csv'.format(os.getcwd()), header=True, sep="\t")
# V A L I D A T I O N E X O M E D E P T H A N D A R R A Y S --------------------------------------------
# using only the CNVs variants coming from ExomeDepth and arrays sufficiently covered
datasetCNVsARRAYS_COVERED = pd.read_csv('{}/datasetCNVsARRAYS_COVERED.csv'.format(os.getcwd()), sep="\t")
datasetCOB = pd.read_csv('{}/datasetCNV_VALIDATION_COVERED.csv'.format(os.getcwd()), sep="\t")
datasetCOB.drop(['Unnamed: 0'], axis=1, inplace=True)
datasetCNVsARRAYS_COVERED.drop(['Unnamed: 0'], axis=1, inplace=True)
truePdata = pd.DataFrame(columns = datasetCOB.columns)
falseNdata = pd.DataFrame(columns = ['chrom','band','start','end','size','probes','type','genes','CNV %','Comments'])
falsePdata = pd.DataFrame(columns = datasetCOB.columns)
segduppseudodata = pd.DataFrame(columns = datasetCOB.columns)
# matching names: duplication == gain and deletion == loss
datasetCNVsARRAYS_COVERED['type'][datasetCNVsARRAYS_COVERED.type.str.contains('gain')] = 'duplication'
datasetCNVsARRAYS_COVERED['type'][datasetCNVsARRAYS_COVERED.type.str.contains('loss')] = 'deletion'
a=0
#analysing each CNV coming from ARRAYS individually
for ia, arrayy in datasetCNVsARRAYS_COVERED.iterrows():
a+=1
startarrayy = arrayy['start']
endarrayy = arrayy['end']
chrarrayy = arrayy['chrom']
typearrayy = arrayy['type']
samplearrayy = arrayy['Sample']
ARRAYinED1 = datasetCOB[(samplearrayy==datasetCOB['Sample']) & (startarrayy<=datasetCOB['start']) & (endarrayy>=datasetCOB['end']) & (typearrayy==datasetCOB['type']) & (chrarrayy[-1]==datasetCOB['chr'])]
ARRAYinED2 = datasetCOB[((samplearrayy==datasetCOB['Sample']) & (startarrayy<datasetCOB['start']) & (endarrayy>=datasetCOB['start']) & (endarrayy<=datasetCOB['end']) & (typearrayy==datasetCOB['type']) & (chrarrayy[-1]==datasetCOB['chr']))]
ARRAYinED3 = datasetCOB[((samplearrayy==datasetCOB['Sample']) & (startarrayy<=datasetCOB['end']) & (startarrayy>=datasetCOB['start']) & (endarrayy>datasetCOB['end']) & (typearrayy==datasetCOB['type']) & (chrarrayy[-1]==datasetCOB['chr']))]
if (len(ARRAYinED1) == 0) and (len(ARRAYinED2) == 0) and (len(ARRAYinED3) == 0):
#false negative data corresponds to those Exome Depth CNVs that are not predicted, but they are for arrays
falseNdata = falseNdata.append(arrayy)
y=0
#analysing each CNV coming from EXOMEDEPTH individually
for i, cnvcob in datasetCOB.iterrows():
y+=1
idcnv = cnvcob['Sample']
startcnv = cnvcob['start']
endcnv = cnvcob['end']
chrcnv = 'chr'+str(cnvcob['chr'])
typecnv = cnvcob['type']
samplecnv = cnvcob['Sample']
EDinARRAY1 = datasetCNVsARRAYS_COVERED[(samplecnv==datasetCNVsARRAYS_COVERED['Sample']) & (startcnv>=datasetCNVsARRAYS_COVERED['start']) & (endcnv<=datasetCNVsARRAYS_COVERED['end']) & (chrcnv==datasetCNVsARRAYS_COVERED['chrom']) & (typecnv==datasetCNVsARRAYS_COVERED['type'])]
EDinARRAY2 = datasetCNVsARRAYS_COVERED[((samplecnv==datasetCNVsARRAYS_COVERED['Sample']) & (startcnv<datasetCNVsARRAYS_COVERED['start']) & (endcnv>=datasetCNVsARRAYS_COVERED['start']) & (endcnv<=datasetCNVsARRAYS_COVERED['end']) & (chrcnv==datasetCNVsARRAYS_COVERED['chrom']) & (typecnv==datasetCNVsARRAYS_COVERED['type']))]
EDinARRAY3 = datasetCNVsARRAYS_COVERED[((samplecnv==datasetCNVsARRAYS_COVERED['Sample']) & (startcnv<=datasetCNVsARRAYS_COVERED['end']) & (startcnv>=datasetCNVsARRAYS_COVERED['start']) & (endcnv>datasetCNVsARRAYS_COVERED['end']) & (chrcnv==datasetCNVsARRAYS_COVERED['chrom']) & (typecnv==datasetCNVsARRAYS_COVERED['type']))]
#SEGMENTAL DUPLICATIONS: filtering CNVs falling within segmental duplications
sd1 = segdups[(startcnv>=segdups["chromStart"]) & (endcnv<=segdups["chromEnd"]) & (chrcnv==segdups["chrom"])]
'''
sd2 = segdups[(start<segdups["chromStart"]) & (end>segdups["chromEnd"]) & (allchr==segdups["chrom"])]
sd3 = segdups[((start<segdups["chromStart"]) & (end>=segdups["chromStart"]) & (end<=segdups["chromEnd"]) & (allchr==segdups["chrom"]))]
sd4 = segdups[((start<=segdups["chromEnd"]) & (start>=segdups["chromStart"]) & (end>segdups["chromEnd"]) & (allchr==segdups["chrom"]))]
'''
#considering pseudogenes
pseudo=pseudogenes[(pseudogenes['cdsStart']>=startcnv) & (pseudogenes['cdsEnd']<=endcnv) & (pseudogenes['#chrom']==chrcnv)]
pseudo.reset_index(inplace=True)
#CNVs containing a superdup or a pseudogene
if (len(sd1) > 0) or (len(pseudo) > 0):
segduppseudodata = segduppseudodata.append(cnvcob)
else:
if (len(EDinARRAY1) > 0) or (len(EDinARRAY2) > 0) or (len(EDinARRAY3) > 0):
#true positive data corresponds to those CNVs predicted by ExomeDepth and by arrays (coincide)
truePdata = truePdata.append(cnvcob)
else:
#false positive data corresponds to those CNVs predicted by ExomeDepth but not by arrays (not coincide)
falsePdata = falsePdata.append(cnvcob)
color_print('TRUE POSITIVE: {}'.format(len(truePdata)), color='green')
color_print('FALSE POSITIVE: {}'.format(len(falsePdata)), color='red')
color_print('FALSE NEGATIVE: {}'.format(len(falseNdata)), color='magenta')
color_print('SEGDUPS AND PSEUDOGENES: {}'.format(len(segduppseudodata)), color='white')
print('Total number of CNVs from arrays:', ia)
print('Total number of CNVs from Exome Depth covered:', y)
truePdata.to_csv('{}/truePdata.csv'.format(os.getcwd()), header=True, sep="\t")
falsePdata.to_csv('{}/falsePdata.csv'.format(os.getcwd()), header=True, sep="\t")
falseNdata.to_csv('{}/falseNdata.csv'.format(os.getcwd()), header=True, sep="\t")
segduppseudodata.to_csv('{}/segduppseudodata.csv'.format(os.getcwd()), header=True, sep="\t")
print()
print()
print()
stoptime = timeit.default_timer()
print('\n\nTime:',stoptime - starttime)