forked from homijan/apm-cnv
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathgenerateTFPN.py
executable file
·209 lines (197 loc) · 10.8 KB
/
generateTFPN.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
#!/usr/bin/env python3
import openpyxl
import sys
from CNAdefs import *
from weightedmeanvalue import weightedMeanValues
# Clinical result of karyotype or fish
TP = 'TP'; FP = 'FP'; FN = 'FN'; TN = 'TN'; NA = 'NA'
def updateSheets(row, colNames, w_values, value_sheet, TFPN_sheet, gainThreshold, deletionThreshold):
# User provides gain/deletion thresholds based on the data
# w_values is a dictionary useing cna keys
for cna in w_values:
value = w_values[cna]
kcna = f'k{cna}'
fcna = f'f{cna}'
tcna = f't{cna}'
value_sheet.cell(row, colNames[kcna]).value = value
value_sheet.cell(row, colNames[fcna]).value = value
value_sheet.cell(row, colNames[tcna]).value = value
tcna_value = ref_sheet.cell(row, colNames[tcna]).value
# tcna of loss is t5q, t7q, etc, and gain is t1q, ttrisomy8 , and ttrisomy12 (skip leading 't')
if (tcna[1:]=='1q' or (tcna[1:]=='trisomy8' or tcna[1:]=='trisomy12')):
# Gain true-false-positive-negative logic
if (tcna_value==1 and value > gainThreshold):
TFPN_sheet.cell(row, colNames[tcna]).value = TP
elif (tcna_value==0 and value > gainThreshold):
TFPN_sheet.cell(row, colNames[tcna]).value = FP
elif (tcna_value==1 and value <= gainThreshold):
TFPN_sheet.cell(row, colNames[tcna]).value = FN
elif (tcna_value==0 and value <= gainThreshold):
TFPN_sheet.cell(row, colNames[tcna]).value = TN
else:
TFPN_sheet.cell(row, colNames[tcna]).value = NA
else:
# Deletion true-false-positive-negative logic
if (tcna_value==1 and value < deletionThreshold):
TFPN_sheet.cell(row, colNames[tcna]).value = TP
elif (tcna_value==0 and value < deletionThreshold):
TFPN_sheet.cell(row, colNames[tcna]).value = FP
elif (tcna_value==1 and value >= deletionThreshold):
TFPN_sheet.cell(row, colNames[tcna]).value = FN
elif (tcna_value==0 and value >= deletionThreshold):
TFPN_sheet.cell(row, colNames[tcna]).value = TN
else:
TFPN_sheet.cell(row, colNames[tcna]).value = NA
def getSEtable(diagnosis, cnas, TFPN_sheet):
# Names to indexes of TFPN sheet
colNames = {}
for i in range(TFPN_sheet.max_column):
col = i + 1
colName = TFPN_sheet.cell(1, col).value
colNames[colName] = col
# Create table
table = {}
table['names'] = ['Lesion', 'FN', 'FP', 'TN', 'TP', 'Sensitivity', 'Specificity', 'PPV', 'NPV', 'Accuracy']
for cna in cnas:
tcna = f't{cna}'
col = colNames[tcna]
FNcount = 0; FPcount = 0; TNcount = 0; TPcount = 0
for i in range(200):
row = i + 2
if (TFPN_sheet.cell(row, col).value==FN and TFPN_sheet.cell(row, colNames['Diagnosis']).value==diagnosis):
FNcount = FNcount + 1
elif (TFPN_sheet.cell(row, col).value==FP and TFPN_sheet.cell(row, colNames['Diagnosis']).value==diagnosis):
FPcount = FPcount + 1
elif (TFPN_sheet.cell(row, col).value==TN and TFPN_sheet.cell(row, colNames['Diagnosis']).value==diagnosis):
TNcount = TNcount + 1
elif (TFPN_sheet.cell(row, col).value==TP and TFPN_sheet.cell(row, colNames['Diagnosis']).value==diagnosis):
TPcount = TPcount + 1
eps = 1e-32
table[tcna] = [cna, FNcount, FPcount, TNcount, TPcount, TPcount / (TPcount + FNcount + eps), TNcount / (TNcount + FPcount + eps), TPcount / (TPcount + FPcount + eps), TNcount / (TNcount + FNcount + eps), (TPcount + TNcount) / (TPcount + FPcount + TNcount + FNcount + eps)]
return table
def writeSETable(diagnosis, cnas, TFPN_sheet, tableFileName):
tableSE = getSEtable(diagnosis, cnas, TFPN_sheet)
#print(f'method {method}')
print(f'Writing table {tableFileName}')
with open(tableFileName, 'w') as f:
for item in tableSE['names']:
f.write(str(item))
f.write(' ')
f.write('\n')
for cna in cnas:
tcna = 't'+cna
for item in tableSE[tcna]:
f.write(str(item))
f.write(' ')
f.write('\n')
ref_table_name = 'tables/CNV_allcases_r.xlsx'
ref_book = openpyxl.load_workbook(ref_table_name, read_only=True)
ref_sheet = ref_book.active
z_table_name = 'tables/CNV_zscore.xlsx'
z_book = openpyxl.load_workbook(z_table_name)
z_sheet = z_book.active
TFPN_table_name = 'tables/CNV_TFPN.xlsx'
TFPN_book = openpyxl.load_workbook(TFPN_table_name)
TFPN_sheet = TFPN_book.active
Nrows = ref_sheet.max_row
Ncolumns = ref_sheet.max_column
colNames = {}
for i in range(Ncolumns):
col = i + 1
colName = ref_sheet.cell(1, col).value
colNames[colName] = col
# TODO add comments
# Input parametrs
# First provide one of the cnv methods to be used: canary-kurtz-cytobands, canary-kurtz-arms, canary-mse-cytobands, canary-mse-newcytobands, canary-mse-arms, wisecondor, testcnvkit, testichor
cnvMethod = 'canary-kurtz-arms'
if (len(sys.argv)>=2):
cnvMethod = sys.argv[1]
updateTables = False
if (len(sys.argv)>=3):
if (sys.argv[2]=='yes'):
updateTables = True
if (updateTables):
for i in range(200):#range(Nrows-1):
row = i + 2
HLabel = ref_sheet.cell(row, colNames['HSTAMP_Label']).value
# Default weighted meanvalue (redefine if needed in particular case)
defaultValue = 'NA'
# File name and column names used in canary-kurtz output cytobands
if (cnvMethod=='canary-kurtz-cytobands'):
fileName = f'/drive3/dkurtz/HEMESTAMP/CANARy/samples/output/Sample_{HLabel}-T1_Tumor.SegmentedGenome.cytobands-noXY.on-off-combined.txt'
chrColName = "chrNum"; intChromValue = True; startColName = "Start"; endColName = "End"; valueColName = "combinedStoufferZL2CNR"
gainThreshold = 1.96; deletionThreshold = -1.96
# File name and column names used in canary-kurtz output arms
elif (cnvMethod=='canary-kurtz-arms'):
fileName = f'/drive3/dkurtz/HEMESTAMP/CANARy/samples/output/Sample_{HLabel}-T1_Tumor.SegmentedGenome.arms.on-off-combined.txt'
chrColName = "chrNum"; intChromValue = True; startColName = "Start"; endColName = "End"; valueColName = "combinedStoufferZL2CNR"
gainThreshold = 1.96; deletionThreshold = -1.96
# File name and column names used in canary-mse output cytobands
elif (cnvMethod=='canary-mse-cytobands'):
fileName = f'canary-python/results-canary/results-canary-mse/Sample_{HLabel}-T1_Tumor.cnvZscores'
chrColName = "#chr"; intChromValue = False; startColName = "start"; endColName = "end"; valueColName = "gc.corrected.norm.log.std.index.zWeighted.Final"
gainThreshold = 1.96; deletionThreshold = -1.96
# File name and column names used in canary-mse new output cytobands
elif (cnvMethod=='canary-mse-newcytobands'):
fileName = f'/drive3/mse/CNV/Alicia/results-canary2_new/Sample_{HLabel}-T1_Tumor.cnvZscores'
chrColName = "#chr"; intChromValue = False; startColName = "start"; endColName = "end"; valueColName = "gc.corrected.norm.log.std.index.zWeighted.Final"
gainThreshold = 1.96; deletionThreshold = -1.96
# File name and column names used in canary-mse output arms
elif (cnvMethod=='canary-mse-arms'):
fileName = f'/drive3/mse/CNV/Alicia/results-canary5/Sample_{HLabel}-T1_Tumor.cnvZscores'
chrColName = "#chr"; intChromValue = False; startColName = "start"; endColName = "end"; valueColName = "gc.corrected.norm.log.std.index.zWeighted.Final"
gainThreshold = 1.96; deletionThreshold = -1.96
# File name and column names used in wisecondor output
elif (cnvMethod=='wisecondor'):
fileName = f'wisecondor/testSamples/Sample_{HLabel}-T1_Tumor.sorted.samtools-deduped.sorted.offtarget.std.txt'
chrColName = "chrNum"; intChromValue = True; startColName = "Start"; endColName = "End"; valueColName = "z-score"; defaultValue = 0
gainThreshold = 1.96; deletionThreshold = -1.96
# File name and column names used in cnvki-cnst output. Note that cnvkit uses copynumber rather than z-score
elif (cnvMethod=='cnvkit-cns'):
fileName = f'cnvkit/results-cnn-tumor/Sample_{HLabel}-T1_Tumor.samtools.call.cns'
chrColName = "chromosome"; intChromValue = False; startColName = "start"; endColName = "end"; valueColName = "cn"
gainThreshold = 2.0; deletionThreshold = 2.0
# File name and column names used in cnvkit-cnr output. Note that cnvkit uses copynumber rather than z-score
elif (cnvMethod=='cnvkit-cnr'):
fileName = f'cnvkit/results-cnn-tumor/Sample_{HLabel}-T1_Tumor.samtools.call.cnr'
chrColName = "chromosome"; intChromValue = False; startColName = "start"; endColName = "end"; valueColName = "cn"
gainThreshold = 2.0; deletionThreshold = 2.0
# File name and column names used in ichorcna-cns output. Note that ichorcna uses copynumber rather than z-score
elif (cnvMethod=='ichorcna-cns'):
fileName = f'ichorcna/results-ichorcna/{HLabel}.seg'
chrColName = "chr"; intChromValue = True; startColName = "start"; endColName = "end"; valueColName = "copy.number"
gainThreshold = 2.0; deletionThreshold = 2.0
# File name and column names used in ichorcna-cnr output. Note that ichorcna uses copynumber rather than z-score
elif (cnvMethod=='ichorcna-cnr'):
fileName = f'ichorcna/results-ichorcna/{HLabel}.cna.seg'
chrColName = "chr"; intChromValue = True; startColName = "start"; endColName = "end"; valueColName = f'{HLabel}.copy.number'
gainThreshold = 2.0; deletionThreshold = 2.0
#
else:
print(f'Provided cnv method {cnvMethod} not in the supported list: canary-kurtz-cytobands, canary-kurtz-arms, canary-mse-cytobands, canary-mse-newcytobands, canary-mse-arms, wisecondor, cnvkit-cns, cnvkit-cnr, ichorcna-cns,ichorcna-cnr')
# Obtain zscore for every cna from CNAdefs
print(f'fileName: {fileName}')
w_zscores = weightedMeanValues(CNA, cnas, fileName, chrColName, startColName, endColName, valueColName, intChromValue, defaultValue)
updateSheets(row, colNames, w_zscores, z_sheet, TFPN_sheet, gainThreshold, deletionThreshold)
print(f'{cnvMethod} {HLabel} {w_zscores}')
# Save the updated excel table
z_book.save(z_table_name)
# Comment what. why, how is saved?
z_book.save(f'tables/{cnvMethod}-Zscore_table.xlsx')
TFPN_book.save(TFPN_table_name)
# Writing the resulting tables
TFPN_sheets = {}
TFPN_sheets[cnvMethod] = TFPN_sheet
for method in TFPN_sheets:
# Define CLL diagnosis, the list of its CNAs, and the name of the written table
diagnosis = 'CLL'; cnas = ['11q', '13q', '17p', 'trisomy12']
tableFileName = 'tables/'+method+'-'+diagnosis+'-tableSE.txt'
writeSETable(diagnosis, cnas, TFPN_sheets[method], tableFileName)
# Define MDS diagnosis, the list of its CNAs, and the name of the written table
diagnosis = 'MDS'; cnas = ['5q', '7q', '20q', 'trisomy8']
tableFileName = 'tables/'+method+'-'+diagnosis+'-tableSE.txt'
writeSETable(diagnosis, cnas, TFPN_sheets[method], tableFileName)
# Define MM diagnosis, the list of its CNAs, and the name of the written table
diagnosis = 'MM'; cnas = ['1p', '1q', '13q', '17p']
tableFileName = 'tables/'+method+'-'+diagnosis+'-tableSE.txt'
writeSETable(diagnosis, cnas, TFPN_sheets[method], tableFileName)