def LoadFrame(ori_file, mode='b'):
test_frame = pd.read_csv(ori_file)
test_frame = test_frame.astype('str')
if mode == 'b':
test_frame['Reference_Allele'] = test_frame['Reference_Allele'] + \
test_frame['Tumor_Seq_Allele2']
test_frame['Reference_Allele'].loc[test_frame['Reference_Allele'] ==
'GT'] = 'CA'
test_frame['Reference_Allele'].loc[test_frame['Reference_Allele'] ==
'GC'] = 'CG'
test_frame['Reference_Allele'].loc[test_frame['Reference_Allele'] ==
'GA'] = 'CT'
test_frame['Reference_Allele'].loc[test_frame['Reference_Allele'] ==
'AT'] = 'TA'
test_frame['Reference_Allele'].loc[test_frame['Reference_Allele'] ==
'TC'] = 'AG'
test_frame['Reference_Allele'].loc[test_frame['Reference_Allele'] ==
'AC'] = 'TG'
test_frame = SATA_PRETREAT.GetSurvTime(test_frame)
test_frame = SATA_PRETREAT.NeuPreteate(test_frame)
classed_frame = SATA_PRETREAT.Classifier(test_frame)
classed_frame.drop(
classed_frame[classed_frame['class_result'] == 0].index,
inplace=True)
classed_frame = SATA_PRETREAT.DropEmptyData(classed_frame,
column='to_last_known_alive')
return classed_frame
def VisiableSeq(test_frame):
print('File_Loaded')
test_frame = test_frame.astype('str')
test_frame['Reference_Allele'] = test_frame['Reference_Allele'] + \
test_frame['Tumor_Seq_Allele2']
test_frame['Reference_Allele'].loc[test_frame['Reference_Allele'] == 'GT'] = 'CA'
test_frame['Reference_Allele'].loc[test_frame['Reference_Allele'] == 'GC'] = 'CG'
test_frame['Reference_Allele'].loc[test_frame['Reference_Allele'] == 'GA'] = 'CT'
test_frame['Reference_Allele'].loc[test_frame['Reference_Allele'] == 'AT'] = 'TA'
test_frame['Reference_Allele'].loc[test_frame['Reference_Allele'] == 'AG'] = 'TC'
test_frame['Reference_Allele'].loc[test_frame['Reference_Allele'] == 'AC'] = 'TG'
print("********************************")
# print(list(set(test_frame['Reference_Allele'])))
test_frame = SATA_PRETREAT.GetSurvTime(test_frame)
test_frame = SATA_PRETREAT.NeuPreteate(test_frame)
classed_frame = SATA_PRETREAT.Classifier(test_frame)
classed_frame.drop(
classed_frame[classed_frame['class_result'] == 0].index, inplace=True)
classed_frame.reset_index(drop=True, inplace=True)
print('visiable frame loaded')
pre_list = []
post_list = []
for i in range(len(classed_frame)):
try:
pre_list.append(str(WHOLE_SEQUENCE.get_seq(classed_frame.Chromosome[i],
int(
classed_frame.Start_Position[i])-EXT,
int(classed_frame.Start_Position[i])-1)).lower())
except ValueError:
print('!!!!!!!', i, ' wrong.')
pre_list.append('nan')
try:
post_list.append(str(WHOLE_SEQUENCE.get_seq(classed_frame.Chromosome[i],
int(
classed_frame.Start_Position[i])+1,
int(classed_frame.Start_Position[i])+EXT)).lower())
except ValueError:
print('!!!!!!!', i, ' wrong.')
post_list.append('nan')
if i % 10000 == 0:
print(i, ' has finished.')
pre_list = np.array(pre_list)
post_list = np.array(post_list)
pre_list = pre_list.reshape(len(classed_frame), -1)
post_list = post_list.reshape(len(classed_frame), -1)
classed_frame['front_seq'] = pre_list
classed_frame['behind_seq'] = post_list
print(pd.crosstab(classed_frame['class_result'],
classed_frame['Reference_Allele']))
return classed_frame
def _DropEmptyCol(self, drop_col=[]):
"""
paramater:
|-> drop_col: a list of columns that needed to delete empty
"""
new_frame = self.dataframe
for c in drop_col:
new_frame = SATA_PRETREAT.DropEmptyData(new_frame, column=c)
return new_frame
def ChiSeqTest(self, chisq_col='', recomb=[]):
_sata_frame = self._DropEmptyCol(drop_col=[chisq_col])
chiseq_frame = SATA_PRETREAT.ChiSquarePretreat(
self.dataframe, column=chisq_col, class_column=self.class_col)
chiseq_frame = chiseq_frame.fillna(0)
if recomb == []:
chiseq_array = np.array(chiseq_frame)
else:
chiseq_array = np.array(chiseq_frame[recomb])
print(chiseq_array)
chiseq_value = scipy.stats.chi2_contingency(chiseq_array)
chiseq_frame['chi2'] = chiseq_value[0]
chiseq_frame['P'] = chiseq_value[1]
chiseq_frame['df'] = chiseq_value[2]
return chiseq_frame
def CalSataItem(dataframe, outtxtfile, WITH_TYPICAL, output_f, label='base'):
sata_dict = cfg.clicfeat_dict
for cal_item in SATA_LIST:
calcframe = SATA_PRETREAT.DropEmptyData(dataframe, column=cal_item)
print('{:*^120}'.format(cal_item + '_' + label), file=outtxtfile)
print('{:*^120}'.format(cal_item + '_' + label))
if sata_dict[cal_item] == 'ori':
satasclass = SATA_METHODS.SataMethod(dataframe=calcframe,
class_col='class_result')
try:
chiasclass = satasclass.ChiSeqTest(chisq_col=cal_item)
print(chiasclass, file=outtxtfile)
OutputFrame(chiasclass,
in_f=output_f,
filename=cal_item + '_' + label + '_as_class')
except UnboundLocalError:
print(cal_item, ' is loss', file=outtxtfile)
sataasitem = SATA_METHODS.SataMethod(dataframe=calcframe,
class_col=cal_item)
try:
chiasitem = sataasitem.ChiSeqTest(chisq_col='class_result')
print(chiasitem, file=outtxtfile)
OutputFrame(chiasitem,
in_f=output_f,
filename=cal_item + '_' + label + '_as_item')
except UnboundLocalError:
print(cal_item, ' is loss', file=outtxtfile)
elif sata_dict[cal_item] == 'mid':
if DRAW_BOX == True:
mean_col = []
std_col = []
col_col = []
for cl in cfg.class_list_costom:
kk = calcframe[cal_item][calcframe['class_result'] == cl]
kk_count = kk.count()
col_col.append(cl + ': Num=' + str(kk_count))
kk = np.array(kk).astype(float).astype(int)
kk = abs(kk)
if kk.size == 0:
std_col.append(0)
mean_col.append(0)
else:
mean_col.append(np.mean(kk))
std_col.append(np.std(kk))
print(std_col)
plt.errorbar(range(len(col_col)),
mean_col,
yerr=std_col,
fmt='o')
plt.xticks(range(len(col_col)), col_col, rotation=90)
try:
plt.savefig(OUTPUT_PATH + output_f.split('.')[0] + '/' +
str(WITH_TYPICAL) + '-' + cal_item + '.tif')
except FileNotFoundError:
os.mkdir(OUTPUT_PATH + output_f.split('.')[0] + '/')
plt.savefig(OUTPUT_PATH + output_f.split('.')[0] + '/' +
WITH_TYPICAL + '-' + cal_item + '.tif')
plt.close()
sataasmid = SATA_METHODS.SataMethod(dataframe=calcframe)
itemdescribe = sataasmid.Describe(descb_col=cal_item)
itemttest = sataasmid.CalTTest(describeframe=itemdescribe)
print(itemdescribe, file=outtxtfile)
print(itemttest, file=outtxtfile)
OutputFrame(itemdescribe,
in_f=output_f,
filename=cal_item + '_' + label + '_describe')
OutputFrame(itemttest,
in_f=output_f,
filename=cal_item + '_' + label + '_ttest')
if cal_item == 'to_last_known_alive':
print('{:*^120}'.format('logrank'), file=outtxtfile)
lr_frame = sataasmid.LogRankTest(PATH +
output_f.split('.')[0] + '/' +
label + '_logrank')
OutputFrame(lr_frame,
in_f=output_f,
filename=label + '_logrank')
print(lr_frame, file=outtxtfile)
print(lr_frame)
else:
for i in sata_dict[cal_item]:
for j in sata_dict[cal_item][i]:
calcframe[cal_item][calcframe[cal_item] == j] = i
calcframe.drop(calcframe[calcframe[cal_item] == 'delete'].index,
inplace=True)
satacaledcalss = SATA_METHODS.SataMethod(dataframe=calcframe,
class_col='class_result')
try:
chicaledclass = satacaledcalss.ChiSeqTest(chisq_col=cal_item)
print(chicaledclass, file=outtxtfile)
OutputFrame(chicaledclass,
in_f=output_f,
filename=cal_item + '_' + label + '_as_class')
except UnboundLocalError:
print(cal_item, ' is loss', file=outtxtfile)
satacaleditem = SATA_METHODS.SataMethod(dataframe=calcframe,
class_col=cal_item)
try:
chicaleditem = satacaleditem.ChiSeqTest(
chisq_col='class_result')
print(chicaleditem, file=outtxtfile)
OutputFrame(chicaleditem,
in_f=output_f,
filename=cal_item + '_' + label + '_as_item')
except UnboundLocalError:
print(cal_item, ' is loss.', file=outtxtfile)
print("{:=^120}".format('finish'))
print("{:=^120}".format('finish'), file=outtxtfile)
def CalSataItem(dataframe, outtxtfile, output_f, label='base'):
sata_dict = cfg.clicfeat_dict
for cal_item in SATA_LIST:
calcframe = SATA_PRETREAT.DropEmptyData(dataframe, column=cal_item)
print('{:*^120}'.format(cal_item + '_' + label), file=outtxtfile)
print('{:*^120}'.format(cal_item + '_' + label))
if sata_dict[cal_item] == 'ori':
satasclass = SATA_METHODS.SataMethod(dataframe=calcframe,
class_col='class_result')
try:
chiasclass = satasclass.ChiSeqTest(chisq_col=cal_item)
print(chiasclass, file=outtxtfile)
OutputFrame(chiasclass,
in_f=output_f,
filename=cal_item + '_' + label + '_as_class')
except UnboundLocalError:
print(cal_item, ' is loss', file=outtxtfile)
sataasitem = SATA_METHODS.SataMethod(dataframe=calcframe,
class_col=cal_item)
try:
chiasitem = sataasitem.ChiSeqTest(chisq_col='class_result')
print(chiasitem, file=outtxtfile)
OutputFrame(chiasitem,
in_f=output_f,
filename=cal_item + '_' + label + '_as_item')
except UnboundLocalError:
print(cal_item, ' is loss', file=outtxtfile)
elif sata_dict[cal_item] == 'mid':
sataasmid = SATA_METHODS.SataMethod(dataframe=calcframe)
itemdescribe = sataasmid.Describe(descb_col=cal_item)
itemttest = sataasmid.CalTTest(describeframe=itemdescribe)
print(itemdescribe, file=outtxtfile)
print(itemttest, file=outtxtfile)
OutputFrame(itemdescribe,
in_f=output_f,
filename=cal_item + '_' + label + '_describe')
OutputFrame(itemttest,
in_f=output_f,
filename=cal_item + '_' + label + '_ttest')
if cal_item == 'to_last_known_alive':
print('{:*^120}'.format('logrank'), file=outtxtfile)
lr_frame = sataasmid.LogRankTest(PATH +
output_f.split('.')[0] + '/' +
label + '_logrank')
OutputFrame(lr_frame,
in_f=output_f,
filename=label + '_logrank')
print(lr_frame, file=outtxtfile)
print(lr_frame)
else:
for i in sata_dict[cal_item]:
for j in sata_dict[cal_item][i]:
calcframe[cal_item][calcframe[cal_item] == j] = i
calcframe.drop(calcframe[calcframe[cal_item] == 'delete'].index,
inplace=True)
satacaledcalss = SATA_METHODS.SataMethod(dataframe=calcframe,
class_col='class_result')
try:
chicaledclass = satacaledcalss.ChiSeqTest(chisq_col=cal_item)
print(chicaledclass, file=outtxtfile)
OutputFrame(chicaledclass,
in_f=output_f,
filename=cal_item + '_' + label + '_as_class')
except UnboundLocalError:
print(cal_item, ' is loss', file=outtxtfile)
satacaleditem = SATA_METHODS.SataMethod(dataframe=calcframe,
class_col=cal_item)
try:
chicaleditem = satacaleditem.ChiSeqTest(
chisq_col='class_result')
print(chicaleditem, file=outtxtfile)
OutputFrame(chicaleditem,
in_f=output_f,
filename=cal_item + '_' + label + '_as_item')
except UnboundLocalError:
print(cal_item, ' is loss.', file=outtxtfile)
print("{:=^120}".format('finish'))
print("{:=^120}".format('finish'), file=outtxtfile)
outtxtfile.close()