def __init__(self, *args, **kwargs):
DataParser.__init__(self, *args, **kwargs)
if self.data is None:
raise ValueError('IPAQ Parser: Data not loaded')
self.type = ''
# if 'type' in kwargs:
# self.type = kwargs.get('type')
# self.fields = self.dbi.getFields(self.type)
# self.info = self.dbi.getInfo(self.type)
# elif self.etype is not None:
# self.type = self.etype
self.data.dropna(axis=1, how='all', inplace=True)
# self.data = self.data.filter(regex="^[^78]")
fields = [
'sitting', 'walking_days', 'walking_time', 'moderate_days',
'moderate_time', 'vigorous_days', 'vigorous_time', 'pa', 'mvpa'
]
self.fields = fields
self.data = self.data[['ID'] + fields]
# self.data.columns = ['ID'] + fields
self.data['ID'] = self.data.apply(lambda x: stripspaces(x, 0), axis=1)
# self.data['ID'] = self.data['ID'].apply(lambda x: getID(x))
self.sortSubjects('ID')
print('Data load complete')
if self.info is None:
self.info = {'prefix': 'IP', 'xsitype': 'opex:ipaq'}
def __init__(self, *args):
DataParser.__init__(self, *args)
fields = join(self.resource_dir, "dexa_fields.xlsx")
#Replace field headers
if access(fields, R_OK):
self.fields = pd.read_excel(fields, header=0, sheetname='dexa_fields')
df_header = pd.read_excel(fields, header=0, sheetname='dexa_header')
self.header = df_header['concatenated'].tolist()
self.data.columns = self.header
print(("Loaded rows=", len(self.data['ID'])))
#extract subject info
df_subj = self.data.iloc[:,0:4]
df_subj['SubjectID'] = df_subj.apply(lambda x: stripspaces(x, 'ID'), axis=1)
#Split data into intervals
self.intervals = {0:'BASELINE', 3:'MIDPOINT',6:'ENDPOINT', 9:'MID-FOLLOW-UP', 12:'FOLLOW-UP'}
self.df = dict()
for i,intval in list(self.intervals.items()):
cols = [c for c in self.header if c.startswith(intval)]
simplecols = []
for col in cols:
cparts = col.split("_")
simplecols.append("_".join(cparts[1:]))
self.df[i] = pd.concat([df_subj,self.data[cols]], axis=1)
self.df[i].columns = df_subj.columns.tolist() + simplecols
#self.df[i].reindex(df_subj.columns.tolist() + simplecols, fill_value='')
if DEBUG:
msg ="Interval=%s data=%d" % (intval, len(self.df[i]))
print(msg)
self.sortSubjects('SubjectID')
else:
raise ValueError("Cannot access fields file: %s" % fields)
def __init__(self, *args):
# super(AmunetParser, self).__init__(*args) - PYTHON V3
DataParser.__init__(self, *args)
self.dates = dict()
self.subjects = dict()
self.interval = None
self.sortSubjects()
def __init__(self, *args):
DataParser.__init__(self, *args)
#Replace field headers
self.fields = ['depression', 'anxiety', 'stress']
ncols = []
for ix in range(0, 13, 3):
ncols += [c + '_' + str(ix) for c in self.fields]
dropcols = [] #remove check columns
#
self.data.set_index(list(self.data)[0], inplace=True)
for n in range(3, len(self.data.columns), 12):
start = n
end = n + 9
dropcols += self.data.columns.tolist()[start:end]
print(('Selecting Totals columns for ', dropcols))
df = self.data.drop(columns=dropcols)
# #check num cols match and delete end cols in case blank ones have been included
if len(['ID'] + ncols) < len(df.columns):
df = df[df.columns[0:len(ncols)]]
df.reset_index(inplace=True)
df.columns = ['ID'] + ncols
df.set_index('ID', inplace=True)
self.data = df
# #sort subjects
self.data['SubjectID'] = self.data.index
self.sortSubjects('SubjectID')
def __init__(self, *args):
DataParser.__init__(self, *args)
# cleanup subjects
self.data['ID'] = self.data.apply(lambda x: stripspaces(x, 0), axis=1)
# self.data['ID'] = self.data['ID'].apply(lambda x: getID(x))
if self.info is None:
self.info = {'prefix': 'GDN', 'xsitype': 'opex:godin'}
# Replace field headers
self.fields = ['strenuous', 'moderate', 'light', 'total', 'sweat']
fields = ['strenuous', 'moderate', 'light', 'total', 'sweat']
cols = [
'ID', 'Strenuous', 'Moderate', 'Light',
'Totalleisureactivityscore', 'Sweat(1,2,or3)'
]
ncols = ['SubjectID']
renamecols = dict(list(zip(cols, ncols + fields)))
df = self.data.iloc[:, 0:7]
df.dropna(axis=0, how='any', thresh=5,
inplace=True) # remove all empty rows
df.fillna(999, inplace=True) # replace any remaining na with 999
df.rename(columns=renamecols, inplace=True)
df.reindex()
self.data = df
# sort subjects
self.sortSubjects('SubjectID')
print('Data load complete')
def __init__(self, inputdir, fieldsfile='fmri.xlsx', *args):
DataParser.__init__(self, *args)
self.inputdir = inputdir
self.fieldsfile = os.path.join(self.resource_dir, 'fields', fieldsfile)
self.lookup()
self.getdata()
self.sortSubjects('Subject')
def __init__(self, *args):
DataParser.__init__(self, *args)
try:
self.cantabNewFields()
self.sortSubjects('Participant ID')
except Exception as e:
print(e)
raise e
def __init__(self, *args):
DataParser.__init__(self, *args)
#Replace field headers
self.fields = [
'calories_burned', 'steps', 'distance', 'floors', 'min_sed',
'min_lightact', 'min_fairact', 'min_veryact', 'act_calories'
]
self.sortSubjects('Subject')
print('Data load complete')
def __init__(self, inputdir, filename, *args):
DataParser.__init__(self, *args)
self.inputdir = inputdir
self.filename = filename
self.get_date()
self.fields = ['q' + str(i) for i in range(1, 42)]
self.fields_comments = ['q42a', 'q42b', 'q42c']
self.score_fcas()
self.sortSubjects(subjectfield='Subject')
def __init__(self, *args):
DataParser.__init__(self, *args)
cols = ["AttentionOrientation", "Memory", "Fluency", "Language", "Visuospatial", "MMSE", "ACERTotal"]
self.fields = ['attention', 'memory', 'fluency', 'language', 'visuospatial', 'MMSE', 'total']
df = self.data
df = df[['ID', 'TimePoint'] + cols]
# df['ID'] = df['ID'].apply(lambda x: getID(x))
df.columns=['ID', 'interval'] + self.fields
self.data = df
self.sortSubjects()
def setUp(self):
try:
datafile = join(ROOTDATADIR, 'blood', 'MULTIPLEX',
'2018-02-01 1058VB 1021LB 1107 1114.xlsx')
sheet = 0
skip = 1
header = None
etype = 'MULTIPLEX'
self.dp = None
self.dp = DataParser(datafile, sheet, skip, header, etype)
except Exception as e:
print(e)
def __init__(self, inputdir, *args):
DataParser.__init__(self, *args)
self.inputdir = inputdir
self.etype = basename(dirname(self.inputdir))
self.interval = basename(inputdir)[0:-1]
fields = ['CA1','CA2','DG','CA3','misc','SUB','ERC','BA35','BA36','PHC','sulcus', 'Hippoc']
flist = [[side + '_' + f for f in fields] for side in ['left', 'right']]
self.fields = ['icv'] + [item for sublist in flist for item in sublist] + ['Total_Hippoc']
self.parse()
self.sortSubjects('Subject')
def __init__(self, *args):
DataParser.__init__(self, *args)
# cleanup subjects
self.data['ID'] = self.data.apply(lambda x: stripspaces(x, 0), axis=1)
# self.data['ID'] = self.data['ID'].apply(lambda x: getID(x))
df = self.data
# Replace field headers
self.fields = ['current', 'past', 'total']
columns = ['CurrentResult', 'PastResult', 'TotalResult']
df.rename(columns=dict(list(zip(columns, self.fields))), inplace=True)
self.data = df
self.sortSubjects('ID')
print('Data load complete')
def __init__(self, *args):
DataParser.__init__(self, *args)
path = r'resources\fields'
fields = pd.read_csv(join(
path, 'accel_fields.csv'))['ACCELEROMETRY'].values.tolist()
self.type = basename(dirname(self.datafile))
self.location = basename(dirname(dirname(self.datafile)))
print(self.location)
if self.type == 'month':
fields = [f for f in fields if f not in ['day', 'valid_day']]
self.fields = fields
self.data = self.data[['Subject', 'interval'] + self.fields]
self.sortSubjects(subjectfield='Subject')
class TestDataparser(unittest.TestCase):
def setUp(self):
try:
datafile = join(ROOTDATADIR, 'blood', 'MULTIPLEX',
'2018-02-01 1058VB 1021LB 1107 1114.xlsx')
sheet = 0
skip = 1
header = None
etype = 'MULTIPLEX'
self.dp = None
self.dp = DataParser(datafile, sheet, skip, header, etype)
except Exception as e:
print(e)
def tearDown(self):
if self.dp is not None:
self.dp.dbi.closeconn()
def test_info(self):
if self.dp is not None:
expected = {
'prefix': u'MPX',
'xsitype': u'opex:bloodMultiplexData'
}
self.assertDictEqual(expected, self.dp.info)
def test_fields(self):
if self.dp is not None:
expected = [
u'GH', u'Leptin', u'BDNF', u'IGFBP7', u'IL1', u'IL4', u'IL6',
u'IL10'
]
self.assertListEqual(expected, self.dp.fields)
def test_prefix(self):
if self.dp is not None and self.dp.info is not None:
data = self.dp.getPrefix()
expected = 'MPX'
self.assertEqual(expected, data)
def test_xsd(self):
if self.dp is not None and self.dp.info is not None:
data = self.dp.getxsd()
expected = 'opex:bloodMultiplexData'
self.assertEqual(expected, data)
def __init__(self, *args, **kwargs):
DataParser.__init__(self, *args)
#Maybe empty sheet
if self.data.empty or len(self.data.columns) <= 1:
msg = "No data available"
raise ValueError(msg)
# cleanup subjects
self.data['ID'] = self.data.apply(lambda x: stripspaces(x, 0), axis=1)
if self.info is None:
self.info = {'prefix': 'INS', 'xsitype': 'opex:insomnia'}
# Replace field headers
self.fieldmap = {
'q1': 'Q1',
'q2': 'Q2',
'q3': 'Q3',
'q4': 'Q4',
'q5': 'Q5',
'q6': 'Q6',
'q7': 'Q7',
'total': 'TotalScore'
}
cols = [
'ID', self.fieldmap['q1'], self.fieldmap['q2'],
self.fieldmap['q3'], self.fieldmap['q4'], self.fieldmap['q5'],
self.fieldmap['q6'], self.fieldmap['q7'], self.fieldmap['total']
]
# self.fieldmap = {'total': 'TotalScore'}
self.fields = list(self.fieldmap.keys())
ncols = ['SubjectID'] + self.fields
# cols = ['ID', self.fieldmap['total']]
# zeros have been entered when should be blank
self.data[self.fieldmap['total']] = self.data.apply(
lambda x: self.nodatarow(x, self.fieldmap['total']), axis=1)
self.data[cols[1:]] = self.data.apply(
lambda x: self.nodatarow(x, cols[1:]), axis=1)
df = self.data[cols]
df = df.astype(object) # convert to object
df.columns = ncols
df.reindex()
self.data = df
# sort subjects
self.sortSubjects('SubjectID')
print('Data load complete')
def __init__(self, *args):
DataParser.__init__(self, *args)
df = self.data
df['Subject'] = df['subjname'].apply(
lambda x: re.findall('(?<=sub-)(.*)(?=_)', x)[0])
df['interval'] = df['subjname'].apply(
lambda x: int(re.findall('(?<=ses-)(.*)', x)[0]))
renamecols = {
'ICV': 'icv',
'right_subiculum': 'right_SUB',
'left_subiculum': 'left_SUB'
}
df.rename(columns=renamecols, inplace=True)
self.cols = [c for c in self.fields if c in df.columns]
self.data = df[['Subject', 'interval'] + self.cols]
self.sortSubjects('Subject')
def __init__(self, *args):
DataParser.__init__(self, *args)
# self.convertScores()
# self.scoreSF()
# pop_fields = ['GenPop_PF', 'GenPop_RP', 'GenPop_BP', 'GenPop_GH', 'GenPop_VT', 'GenPop_SF', 'GenPop_RE',
# 'GenPop_MH', 'GenPop_PCS', 'GenPop_MCS']
self.fields = ['PF', 'RP', 'BP', 'GH', 'VT', 'SF', 'RE', 'MH', 'PCS', 'MCS']
df = self.data
df[['Subject', 'interval']] = df.pop('RecordID').str.split('\\s', 1, expand=True)
df['interval'] = df['interval'].apply(extract_interval)
# df['Subject'] = df['Subject'].apply(lambda x: getID(x))
self.data = df
# self.data = df[['Subject', 'interval'] + self.fields]
self.sortSubjects('Subject')
def setUp(self):
try:
datafile = join(ROOTDATADIR, 'cantab',
'RowBySession_HealthyBrains_20180504.csv')
sheet = 0
skip = 0
header = None
etype = 'CANTAB'
self.dp = None
self.dp = DataParser(datafile, sheet, skip, header, etype)
except Exception as e:
print(e)
def __init__(self, *args):
DataParser.__init__(self, *args)
# cleanup subjects
self.data['ID'] = self.data.apply(lambda x: stripspaces(x, 0), axis=1)
# self.data['ID'] = self.data['ID'].apply(lambda x: getID(x))
if self.info is None:
self.info = {'prefix': 'GDN', 'xsitype': 'opex:paces'}
# Replace field headers
self.fields = ['q' + str(i)
for i in range(1, 9)] + ['total', 'enjoy_percent']
columns = [
'Q1', 'Q2', 'Q3', 'Q4', 'Q5', 'Q6', 'Q7', 'Q8', 'SumTotal',
'%Enjoyment'
]
renamecols = dict(list(zip(columns, self.fields)))
self.data.rename(columns=renamecols, inplace=True)
self.sortSubjects('ID')
print('Data load complete')
def __init__(self, *args):
DataParser.__init__(self, *args)
# Maybe empty sheet
if self.data.empty or len(self.data.columns) <= 1:
msg = "No data available"
raise ValueError(msg)
# cleanup subjects
self.data['ID'] = self.data.apply(lambda x: stripspaces(x, 0), axis=1)
# self.data['ID'] = self.data['ID'].apply(lambda x: getID(x))
if self.info is None:
self.info = {'prefix': 'PSQ', 'xsitype': 'opex:psqi'}
# Replace field headers
self.fields = ['c'+str(i) for i in range(1,8)] + ['total']
ncols = ['SubjectID'] + self.fields
cols = ['ID'] + [c for c in self.data.columns if (isinstance(c,str) or isinstance(c,str)) and c.startswith('Component')] + ['total']
df = self.data[cols]
df.columns = ncols
df.reindex()
self.data = df
# sort subjects
self.sortSubjects('SubjectID')
print('Data load complete')
def __init__(self, inputdir, inputsubdir, datafile, testonly=False):
DataParser.__init__(self, etype='COSMED')
self.inputdir = inputdir
self.testonly = testonly
# Load fields
fieldsfile = join(self.getResourceDir(), "cosmed_fields.xlsx")
self.subjectdataloc = pd.read_excel(fieldsfile,
header=0,
sheet_name='cosmed')
self.fields = pd.read_excel(fieldsfile,
header=0,
sheet_name='cosmed_xnat')
self.datafields = pd.read_excel(fieldsfile,
header=0,
sheet_name='cosmed_data')
# Get list of subjects - parse individual files
self.subjects = dict()
self.files = glob.glob(join(inputsubdir, "*.xlsx"))
# create an output dir for processed files
pdir = join(inputsubdir, 'processed')
if not isdir(pdir):
mkdir(pdir)
# Load efficiency data from single file
self.effdata_cols = {
'0': [9, 12],
'3': [13, 16],
'6': [17, 20],
'9': [21, 24],
'12': [25, 28]
}
self.effdata = self.__loadEfficiencydata(datafile)
# Load data from files
self.loaded = self.__loadData()
def __init__(self, *args):
DataParser.__init__(self, *args)
self.training_file = pd.ExcelFile(self.datafile)
self.type = re.findall('(?<=Diary-)[A-Z]{3}', basename(self.datafile))[0]
self.extraction = {'AIT': extract_AIT, 'MIT': extract_MIT, 'LIT': extract_LIT}
self.getData()
def __init__(self, inputdir, *args):
DataParser.__init__(self, *args)
self.inputdir = inputdir
self.getData()
self.sortSubjects('Subject')
def __init__(self, *args):
DataParser.__init__(self, *args)
self.get_data()
self.sortSubjects('Subject')
def __init__(self, *args, **kwargs):
DataParser.__init__(self, *args)
if self.data is None:
raise ValueError('BloodParser: Data not loaded')
self.type = ''
if 'type' in kwargs:
self.type = kwargs.get('type')
self.fields = self.dbi.getFields(self.type)
self.info = self.dbi.getInfo(self.type)
# self.fields = self.getFieldsFromFile(self.type)
elif self.etype is not None:
self.type = self.etype
print('Rename Headers for ', self.type)
## Rename columns in dataframe
if self.type == 'IGF':
colnames = {'Date': 'A_Date',
'Participant ID ': 'Participant ID',
'Timepoint': 'Sample ID',
'IGF-1': 'IGF1'}
self.data = self.data.rename(index=str, columns=colnames)
elif self.type == 'SOMATO':
colnames = {'Date': 'A_Date',
'Participant ID ': 'Participant ID',
'Timepoint': 'Sample ID',
'Somatostatin': 'somatostatin'}
self.data = self.data.rename(index=str, columns=colnames)
elif self.type == 'BDNF':
colnames = {'Date': 'A_Date',
'Participant ID ': 'Participant ID',
'Timepoint': 'Sample ID'}
self.data = self.data.rename(index=str, columns=colnames)
elif self.type == 'MULTIPLEX':
colnames = {'Date': 'A_Date',
'Participant ID ': 'Participant ID',
'Timepoint': 'Sample ID',
'IGFBP-7': 'IGFBP7'}
self.data = self.data.rename(index=str, columns=colnames)
elif self.type == 'INFLAM':
print('Headers for ', self.type)
colnames = {'Date': 'A_Date',
'Participant ID ': 'Participant ID',
'Timepoint': 'Sample ID',
u'IFN\u03b3': 'ifngamma',
'IL-10': 'il10',
'IL-12(p70)': 'il12p70',
u'IL-1\u03b2': 'il1beta',
'IL-6': 'il6',
'IL-8': 'il8cxcl8',
u'TNF\u03B1': 'tnfalpha'
}
self.data = self.data.rename(index=str, columns=colnames)
elif self.type == 'ELISAS':
colnames = {'Date': 'A_Date',
'Participant ID ': 'Participant ID',
'Timepoint': 'Sample ID',
'Beta-H (ng/ul)': 'BetaHydroxy'}
self.data = self.data.rename(index=str, columns=colnames)
elif self.type == 'COBAS':
# Name unnamed columns to field names
if self.fields[0] not in self.data.columns:
colnames = {}
v = 1
for i in range(len(self.fields)):
colnames['Value.' + str(v)] = self.fields[i]
v = v + 2
else:
colnames = {'Date': 'A_Date',
'Participant ID ': 'Participant ID',
'Timepoint': 'Sample ID',
'Prolactin': 'Prolactin',
'Insulin': 'Insulin',
'HGH': 'HGH',
'Cortisol': 'Cortisol'}
self.data = self.data.rename(index=str, columns=colnames)
print('Colnames: ', self.data.columns.tolist())
# Insert Row Number column
if 'R_No.' not in self.data.columns:
self.data.insert(0, 'R_No.', list(range(len(self.data))))
# Remove NaT rows
i = self.data.query('A_Date =="NaT"')
if not i.empty:
self.data.drop(i.index[0], inplace=True)
print('NaT row dropped')
# Organize data into subjects
subjectfield = 'Participant ID'
if subjectfield not in self.data.columns:
raise ValueError('Subject ID field not present: ', subjectfield)
self.data[subjectfield] = self.data[subjectfield].str.replace(" ", "")
self.sortSubjects(subjectfield)
if self.subjects is not None:
print('BloodParser: subjects loaded successfully')
self.subjectfield = subjectfield
def __init__(self, *args):
DataParser.__init__(self, *args)
# self.opex = pd.read_csv(join(self.resource_dir, 'opex.csv'))
self.expts = dict()
def __init__(self, **kwargs):
# super(AmunetParser, self).__init__(*args) - PYTHON V3
DataParser.__init__(self, **kwargs)
self.data = missingData(self.datafile)
self.sortSubjects('Subject')
