def run_and_return(command):
if system_platform() == 'Windows':
return run_and_return_winfix(command)
else:
with Sultan.load() as s:
result = s.java('-jar ChunkyLauncher.jar -' + command).run()
log(str(result.stdout))
temp = [x.replace('\t', '') for x in result.stderr] # remove \t
return temp
print('ERROR! Choice shoule be `0` or `1`!')
temp_time = time.time() # 计个时看看
name = aao_login(stuID, stuPwd, retry_cnt)
print('\nMeow~下面开始获取{}课表啦！\n'.format({0: '个人', 1: '班级'}.get(choice)))
courseTable = getCourseTable(choice=choice)
list_lessonObj = parseCourseTable(courseTable)
print('课表获取完成，下面开始生成iCal日历文件啦！')
cal = create_ics(list_lessonObj, semester_start_date)
print('日历生成完成，下面开始导出啦！\n')
export_ics(cal, semester_year, semester, stuID) # Export `.ics` file
if not args.notxt: # 若命令行参数含`--notxt`则不导出
exportCourseTable(list_lessonObj, semester_year, semester,
stuID) # Export `.txt` file
if not args.noxlsx: # 若命令行参数含`--noxlsx`则不导出
print('\n开始生成xlsx表格文件！ ')
xlsx = create_xls(list_lessonObj, semester_year, semester, stuID)
print('xlsx文件生成完成，开始导出！')
export_xls(xlsx, semester_year, semester,
stuID) # Export `.xlsx` file
print('\n导出完成，累计用时：', time.time() - temp_time, 's')
print("Thanks for your use! 欢迎来GitHub上点个Star呢！")
except Exception as e:
print("ERROR! 欢迎在GitHub上提出issue & Pull Request!")
print(e)
finally:
session.cookies.clear() # 清一下cookie
if system_platform(
) == 'Windows': # Fix Linux `sh: 1: pause: not found` bug
os.system('pause')
def mr_test(expdf, outcdf, opts, logger):
logger.logger.info(
'>>>>>>>>>>MR analyses for {0} and {1} start<<<<<<<<<<'.format(
opts.exp_name, opts.out_name))
expdf['RS'] = expdf['RS'].str.lower()
outcdf['RS'] = outcdf['RS'].str.lower()
expdf.to_csv('exposure.txt',
sep='\t',
na_rep='NA',
float_format='%g',
encoding='utf-8',
index=False)
outcdf.to_csv('outcome.txt',
sep='\t',
na_rep='NA',
float_format='%g',
encoding='utf-8',
index=False)
pleiotropy(opts, logger)
mr_result = pd.DataFrame({
'outcome': [],
'exposure': [],
'method': [],
'nsnp': [],
'b': [],
'se': [],
'pval': []
})
mr_heter = pd.DataFrame()
mr_pleio = pd.DataFrame()
mr_data = pd.DataFrame()
mr_pleiosnp = pd.DataFrame()
r_errf = tempfile.TemporaryFile(mode='w+', encoding='utf-8')
if opts.heterogeneity != 'none':
if os.path.exists('mr.pleiosnp'):
mr_pleiosnp = pd.read_csv('mr.pleiosnp',
sep='\s+',
header=0,
dtype=str)
try:
os.remove('mr.pleiosnp')
except:
pass
if 'mr_gsmr' in opts.mr_method:
if expdf.shape[0] < 5:
logger.logger.info('No enough SNPs for GSMR analysis.')
else:
logger.logger.info(
'Notes: make sure that SNP frequency are accurate for GSMR analysis'
)
refdir = os.path.split(os.path.realpath(__file__))[0]
sysstr = system_platform()
gctascript = os.path.join(refdir, 'refsource',
'gcta_{0}'.format(sysstr), 'gcta64')
with open('gsmr_ref_data.txt', 'w') as writer:
writer.write('\n'.join([
os.path.join(refdir, 'refsource', '1000G_EUR_Phase3_plink',
'1000G.EUR.QC.{0}'.format(str(i)))
for i in range(1, 23)
]))
with open('gsmr_exposure.txt', 'w') as writer:
writer.write('\t'.join([opts.exp_name, 'exposure.txt']) + '\n')
with open('gsmr_outcome.txt', 'w') as writer:
writer.write('\t'.join([opts.out_name, 'outcome.txt']) + '\n')
try:
commands = '{0} --mbfile gsmr_ref_data.txt --gsmr-file gsmr_exposure.txt gsmr_outcome.txt --gsmr-direction 0 --out gsmr_result_pleio --heidi-thresh 0 --effect-plot --diff-freq 1 --clump-r2 1 --gwas-thresh 1 --gsmr-snp-min 5'.format(
gctascript)
content1 = subprocess.check_output(commands,
stderr=None,
shell=True,
bufsize=0)
'''gsmr_result = pd.read_csv('gsmr_result_pleio.gsmr', sep='\s+', header=0,dtype=str,
names=['exposure', 'outcome', 'b', 'se', 'pval', 'nsnp', 'global_heidi_outlier'])'''
gsmr_result = pd.read_csv(
'gsmr_result_pleio.gsmr',
sep='\s+',
header=0,
dtype=str,
names=['exposure', 'outcome', 'b', 'se', 'pval', 'nsnp'])
gsmr_result['method'] = 'gsmr' * gsmr_result.shape[0]
gsmr_result = gsmr_result[[
'outcome', 'exposure', 'method', 'nsnp', 'b', 'se', 'pval'
]].copy()
mr_result = pd.concat([mr_result, gsmr_result], axis=0)
#logger.logger.info(content1.decode('utf-8'))
if opts.heterogeneity == 'none':
r_errf.seek(0)
r_errf.truncate()
content2 = subprocess.check_output(
'{0} --mbfile gsmr_ref_data.txt --gsmr-file gsmr_exposure.txt gsmr_outcome.txt --gsmr-direction 0 --out gsmr_result_nonpleio --heidi-thresh 0.01 --effect-plot --diff-freq 1 --clump-r2 1 --gwas-thresh 1 --gsmr-snp-min 5'
.format(gctascript),
stderr=r_errf,
shell=True,
bufsize=0)
#logger.logger.info(content2.decode('utf-8'))
gsmr_result = pd.read_csv('gsmr_result_nonpleio.gsmr',
sep='\s+',
header=0,
names=[
'exposure', 'outcome', 'b',
'se', 'pval', 'nsnp'
],
dtype=str)
gsmr_result[
'method'] = 'gsmr-outlier-correction' * gsmr_result.shape[
0]
gsmr_result = gsmr_result[[
'outcome', 'exposure', 'method', 'nsnp', 'b', 'se',
'pval'
]].copy()
mr_result = pd.concat([mr_result, gsmr_result], axis=0)
except subprocess.CalledProcessError as e:
r_errf.flush()
r_errf.seek(0)
logger.logger.info('Error info:\n{0}'.format(r_errf.read()))
logger.logger.info("GSMR analysis Failed\nSkiping...")
r_errf.seek(0)
r_errf.truncate()
try:
os.remove('gsmr_ref_data.txt')
os.remove('gsmr_exposure.txt')
os.remove('gsmr_outcome.txt')
os.remove('gsmr_result_nonpleio.gsmr')
os.remove('gsmr_result_pleio.gsmr')
os.remove('gsmr_result_nonpleio.log')
os.remove('gsmr_result_pleio.log')
os.remove('gsmr_result_nonpleio.pleiosnps')
except:
pass
if set([
"mr_wald_ratio", "mr_two_sample_ml mr_egger_regression",
"mr_egger_regression_bootstrap", "mr_simple_median",
"mr_weighted_median", "mr_penalised_weighted_median",
"Penalised weighted median", "mr_ivw", "mr_ivw_radial",
"mr_ivw_mre", "mr_ivw_fe", "mr_simple_mode", "mr_weighted_mode",
"mr_weighted_mode_nome", "mr_simple_mode_nome", "mr_raps",
"mr_sign", "mr_uwr"
]) & set(opts.mr_method):
try:
#写提示，SNP少某些MR方法无法实现
output = subprocess.check_output(
'Rscript mr_test.R {0} {1} {2} {3} {4} {5}'.format(
'exposure.txt', 'outcome.txt', opts.exp_name,
opts.out_name, ','.join(set(opts.mr_method)), 'no' if
opts.heterogeneity in ['mr_presso', 'mr_gsmr'] else 'yes'),
stderr=r_errf,
shell=True,
bufsize=0)
#logger.logger.info(output.decode('utf-8'))
if os.path.exists('mr.result'):
twosamplemr_result = pd.read_csv('mr.result',
sep='\s+',
header=0,
dtype=str)
mr_result = pd.concat([mr_result, twosamplemr_result], axis=0)
os.remove('mr.result')
if os.path.exists('mr.heter'):
twosamplemr_heter = pd.read_csv('mr.heter',
sep='\s+',
header=0,
dtype=str)
mr_heter = pd.concat([mr_heter, twosamplemr_heter], axis=0)
os.remove('mr.heter')
if os.path.exists('mr.pleio'):
twosamplemr_pleio = pd.read_csv('mr.pleio',
sep='\s+',
header=0,
dtype=str)
mr_pleio = pd.concat([mr_pleio, twosamplemr_pleio], axis=0)
os.remove('mr.pleio')
if os.path.exists('mr.data'):
twosamplemr_data = pd.read_csv('mr.data',
sep='\s+',
header=0,
dtype=str)
mr_data = pd.concat([mr_data, twosamplemr_data], axis=0)
os.remove('mr.data')
except subprocess.CalledProcessError as e:
r_errf.flush()
r_errf.seek(0)
logger.logger.info('Error info:\n{0}'.format(r_errf.read()))
logger.logger.info("MR analysis Failed.\nSkiping...")
except FileNotFoundError as e1:
pass
try:
os.remove('exposure.txt')
os.remove('outcome.txt')
except:
pass
r_errf.close()
if mr_result.empty:
logger.logger.info('\n=====No MR results=====\n')
else:
logger.logger.info('\n=====Main MR results=====\n')
logger.logger.info(str(mr_result))
if mr_heter.empty:
logger.logger.info('\n=====No heterogeneity test results=====\n')
else:
logger.logger.info(
'\n=====Main heterogeneity test for MR results=====\n')
logger.logger.info(str(mr_heter))
if mr_pleio.empty:
logger.logger.info('\n=====No Egger pleiotropy test result=====\n\n')
else:
logger.logger.info('\n--Egger pleiotropy: Intercetpt--\n\n')
logger.logger.info(str(mr_pleio))
logger.logger.info(
'\n\n>>>>>>>>>>MR analysis for {0} and {1} finished!<<<<<<<<<<\n'.
format(opts.exp_name, opts.out_name))
return mr_result, mr_heter, mr_pleio, mr_data, mr_pleiosnp
def pleiotropy(opts, logger):
r_errf = tempfile.TemporaryFile(mode='w+', encoding='utf-8')
try:
os.remove('mr.pleiosnp')
except:
pass
if opts.heterogeneity == 'none':
logger.logger.info("Notes: no pleiotropy test is adopted!")
elif opts.heterogeneity == 'mr_presso':
logger.logger.info('Pleiotropy test for {0} using {1}'.format(
opts.exp_name, opts.heterogeneity))
try:
output = subprocess.check_output(
'Rscript pleiotropy_test.R {0} {1} {2} {3} {4}'.format(
'exposure.txt', 'outcome.txt', opts.exp_name,
opts.out_name, opts.heterogeneity),
stderr=r_errf,
shell=True,
bufsize=0)
#logger.logger.info(output.decode('utf-8'))
presso_pleiosnp = pd.read_csv('mr.pleiosnp',
sep='\s+',
header=0,
dtype=str)
if not presso_pleiosnp.empty:
logger.logger.info(
'Pleiotropic SNPs for {0} and {1} using MR-PRESSO global test!'
.format(opts.exp_name, opts.out_name))
logger.logger.info(str(presso_pleiosnp))
else:
logger.logger.info(
'No pleiotropic SNPs for {0} and {1} identified by MR-PRESSO global test.'
.format(opts.exp_name, opts.out_name))
except subprocess.CalledProcessError as e:
r_errf.flush()
r_errf.seek(0)
logger.logger.info('Error info:\n{0}'.format(r_errf.read()))
logger.logger.info("Pleiotropy test Failed.\nSkiping...")
r_errf.seek(0)
r_errf.truncate()
except FileNotFoundError as e1:
pass
elif opts.heterogeneity == 'mr_gsmr':
logger.logger.info('Pleiotropy test for {0} using {1}'.format(
opts.exp_name, opts.heterogeneity))
if pd.read_csv('exposure.txt', sep='\s+', header=0,
dtype=str).shape[0] < 5:
logger.logger.info(
'No enough SNPs to perform GSMR HEIDI outlier test')
return
refdir = os.path.split(os.path.realpath(__file__))[0]
sysstr = system_platform()
gctascript = os.path.join(refdir, 'refsource',
'gcta_{0}'.format(sysstr), 'gcta64')
try:
with open('gsmr_ref_data.txt', 'w') as writer:
writer.write('\n'.join([
os.path.join(refdir, 'refsource', '1000G_EUR_Phase3_plink',
'1000G.EUR.QC.{0}'.format(str(i)))
for i in range(1, 23)
]))
with open('gsmr_exposure.txt', 'w') as writer:
writer.write('\t'.join([opts.exp_name, 'exposure.txt']) + '\n')
with open('gsmr_outcome.txt', 'w') as writer:
writer.write('\t'.join([opts.out_name, 'outcome.txt']) + '\n')
command = '{0} --mbfile gsmr_ref_data.txt --gsmr-file gsmr_exposure.txt gsmr_outcome.txt --gsmr-direction 0 --out gsmr_result_nonpleio --heidi-thresh 0.01 --diff-freq 1 --clump-r2 1 --gwas-thresh 1 --gsmr-snp-min 5'.format(
gctascript)
content2 = subprocess.check_output(command,
stderr=r_errf,
shell=True,
bufsize=0)
#logger.logger.info(content2.decode('utf-8'))
if os.path.exists('gsmr_result_nonpleio.pleio_snps'):
gsmr_pleiosnp = pd.read_csv(
'gsmr_result_nonpleio.pleio_snps',
sep='\s+',
header=None,
names=['exposure', 'outcome', 'RS'],
dtype=str)
gsmr_pleiosnp[
'method'] = 'gsmr-outlier-correction' * gsmr_pleiosnp.shape[
0]
logger.logger.info(
'Pleiotropic SNPs for {0} and {1} using GSMR HEIDI outlier test!'
.format(opts.exp_name, opts.out_name))
gsmr_pleiosnp = gsmr_pleiosnp[[
'exposure', 'outcome', 'method', 'RS'
]]
logger.logger.info(str(gsmr_pleiosnp))
gsmr_pleiosnp.to_csv('mr.pleiosnp',
header=True,
index=False,
sep='\t',
na_rep='NA',
float_format='%g',
encoding='utf-8')
for x in ['exposure.txt', 'outcome.txt']:
df = pd.read_csv(x, sep='\s+', header=0, dtype=str)
df = df[df['RS'].isin(gsmr_pleiosnp['RS']) == False]
df.to_csv(x,
header=True,
sep='\t',
na_rep='NA',
float_format='%g',
encoding='utf-8',
index=False)
else:
logger.logger.info(
'No pleiotropic SNPs for {0} and {1} identified by GSMR HEIDI outlier test'
.format(opts.exp_name, opts.out_name))
except subprocess.CalledProcessError as e:
r_errf.flush()
r_errf.seek(0)
logger.logger.info('Error info:\n{0}'.format(r_errf.read()))
logger.logger.info("Pleiotropy test Failed.\nSkiping...")
r_errf.seek(0)
r_errf.truncate()
try:
os.remove('gsmr_ref_data.txt')
os.remove('gsmr_exposure.txt')
os.remove('gsmr_outcome.txt')
os.remove('gsmr_result_nonpleio.pleio_snps')
os.remove('gsmr_result_nonpleio.gsmr')
os.remove('gsmr_result_nonpleio.gsmr')
os.remove('gsmr_result_nonpleio.badsnps')
except:
pass
r_errf.close()
else:
logger.logger.info('No pleiotropy test for {0} using {1}'.format(
opts.exp_name, opts.heterogeneity))
def clumping(opts, expdf, refdir, logger):
plinkreport = expdf.loc[:, ['RS', 'P']]
plinkreport['RS'] = plinkreport['RS'].str.lower()
plinkreport.to_csv('plinkreport',
sep='\t',
header=['SNP', 'P'],
index=False)
independent = []
if opts.clump and opts.origin == 'offline':
for i in range(1, 23):
try:
errf = tempfile.TemporaryFile()
sysstr = system_platform()
plinkscript = os.path.join(refdir, 'refsource',
'plink_{0}'.format(sysstr), 'plink')
geno = os.path.join(refdir, 'refsource',
'1000G_EUR_Phase3_plink',
'1000G.EUR.QC.{0}'.format(str(i)))
content = subprocess.check_output(
'{0} --bfile {1} --clump {2} --clump-p1 {3} --clump-r2 {4} --clump-kb {5} '
.format(plinkscript, geno, 'plinkreport', opts.clump_p1,
opts.r2, opts.windowsize),
shell=True,
stderr=errf,
bufsize=0)
errf.flush()
errf.seek(0)
if 'Warning: No significant --clump results' in errf.read(
).decode('utf-8'):
logger.logger.info(
'No significant SNPs in chromosome {0}'.format(str(i)))
continue
else:
pass
#logger.logger.info(content.decode('utf-8')[-82:-35])
errf.close()
except subprocess.CalledProcessError as err1:
logger.logger.warning(
'Warning: plink clumping algorithm for chromosome {0} failed. Please check .log file'
.format(str(i)))
errf.flush()
errf.seek(0)
logger.logger.warning(errf.read().decode('utf-8'))
continue
with open('plink.clumped', 'r') as reader:
snps = [
x.strip().split()[2] for x in reader.readlines()[1:]
if not x.isspace()
]
independent.extend(snps)
logger.logger.info("{0} independent SNPs after clumping".format(
len(independent)))
try:
os.remove('plink.clumped')
os.remove('plinkreport')
os.remove('plink.log')
os.remove('plink.nosex')
except:
pass
elif opts.clump and opts.origin == 'online':
logger.logger.info('Not developed now.\nExiting...')
exit(1)
pass
# TODO 在线clumping
else:
independent = list(expdf['RS'])
pd.Series(independent).to_csv('include',
header=False,
sep='\t',
na_rep='NA',
float_format='%g',
encoding='utf-8',
index=False)
return independent