def file_callback(attrname, old, new):
# decode data
raw_contents = file_source.data['file_contents'][0]
prefix, b64_contents = raw_contents.split(',', 1)
file_contents = base64.b64decode(b64_contents).decode('utf-8-sig',
errors='ignore')
file_io = StringIO(file_contents)
# update dataframe
global experiment_df
experiment_df = pd.read_csv(file_io)
experiment_df.columns = [str(i) for i in list(experiment_df)]
# update database
global experiment_db
experiment_db = dict(model='Michaelis-Menten')
xmin = experiment_df[experiment_df.columns[0]].values[0]
xmax = experiment_df[experiment_df.columns[0]].values[-1]
for s in experiment_df.columns[1:]:
if np.max(experiment_df[s]) > 0.:
df = experiment_df[[experiment_df.columns[0], s]].dropna(axis=0)
experiment_db[s] = ck.progress_curve(df, xmin, xmax)
experiment_db[s + '_fit'] = 0
# reload page
load_page(experiment_df, experiment_db)
def update():
# get selections
fit_routine = fit_button.active
scalex = len(scalex_box.active)
model_eq = model_select.value
subtract = subtract_select.value
sample = sample_select.value
transform = transform_input.value
offset = offset_input.value
top = top_fix.value
bottom = bottom_fix.value
slope = slope_fix.value
threshold = threshold_slider.value
start = start_time.value
end = end_time.value
warning.visible = False
warning_source.data = pd.DataFrame(data=dict(x=[], y=[], t=[]))
circles.visible = False
circles_source.data = pd.DataFrame(data=dict(x=[], y=[]))
model.title.text = model_eq
if 'x' in transform:
raw.yaxis.axis_label = 'Concentration'
else:
raw.yaxis.axis_label = 'Concentration'
# update database
experiment_db[sample + '_fit'] = fit_routine
experiment_db['model'] = model_eq
pdf = experiment_df[[experiment_df.columns[0], sample]]
experiment_db[sample] = ck.progress_curve(pdf, start, end)
if fit_routine == 3:
raw.title.text = "Schnell-Mendoza Fit"
resi.title.text = "Schnell-Mendoza Fit Residuals"
raw.title.text = "Schnell-Mendoza Fit"
resi.title.text = "Schnell-Mendoza Fit Residuals"
model_select.value = 'Michaelis-Menten'
scalex_box.active = []
model_data_source.data = pd.DataFrame(
data=dict(xt=[], yt=[], et=[], n=[], ct=[])).to_dict('list')
model_plot_source.data = pd.DataFrame(
data=dict(xp=[], yp=[], u=[], l=[], ep=[], cp=[])).to_dict('list')
ic_source.data = pd.DataFrame(data=dict(
label=[], Bottom=[], Top=[], Slope=[], p50=[])).to_dict('list')
if 'x' in transform:
raw_data, model_result, fit_data, varea_data = ck.sm_fit(
experiment_db).fit(sample, transform, subtract)
model_fit_source.data = pd.DataFrame(
data=dict(x=model_result['xfit'],
y=model_result['yfit'])).to_dict('list')
varea_source.data = pd.DataFrame(data=dict(
x=varea_data['x'], r1=varea_data['r1'],
r2=varea_data['r2'])).to_dict('list')
mm_source.data = pd.DataFrame(
data=dict(label=['Fit Value', 'Std. Error'],
Km=fit_data['Km'],
Vmax=fit_data['Vmax']),
index=['value', 'error']).to_dict('list')
raw_source.data = pd.DataFrame(data=dict(
x=raw_data['x'],
y=raw_data['y'],
yr=raw_data['resi'],
yfit=raw_data['yfit'],
)).to_dict('list')
else:
warning.visible = True
warning_source.data = pd.DataFrame(data=dict(
x=[0],
y=[0],
t=[
'Please enter transform equation! \nMust convert signal to [substrate] \nin Schnell-Mendoza mode (e.g. via \nx/6.22/0.45/0.001 for sample data). \nNote: this transform may need \nto be inverted through multiplying \nby -1 when analyzing experiments \nthat measure increasing product \nconcentration over time)'
]))
circles.visible = True
circles_source.data = pd.DataFrame(
data=dict(x=[-.05, -.05, 1.6, 1.6], y=[0, 0.6, 0, 0.6]))
raw.x_range = Range1d(-0.1, 2.5)
raw_source.data = pd.DataFrame(data=dict(
x=[],
y=[],
yr=[],
yfit=[],
)).to_dict('list')
model_fit_source.data = pd.DataFrame(
data=dict(x=[], y=[])).to_dict('list')
varea_source.data = pd.DataFrame(
data=dict(x=[], r1=[], r2=[])).to_dict('list')
mm_source.data = pd.DataFrame(data=dict(label=['', ''],
Km=['', ''],
Vmax=['', '']),
index=['value',
'error']).to_dict('list')
else:
raw.title.text = "Initial Rate Fit"
resi.title.text = "Initial Rate Fit Residuals"
varea_source.data = pd.DataFrame(
data=dict(x=[], r1=[], r2=[])).to_dict('list')
raw.yaxis.axis_label = 'Signal'
if fit_routine == 0:
progress_data = experiment_db[sample].spline()
experiment_db[sample].spline = progress_data
elif fit_routine == 1:
progress_data = experiment_db[sample].linear()
experiment_db[sample].linear = progress_data
elif fit_routine == 2:
try:
offset = float(offset)
except:
pass
progress_data = experiment_db[sample].logarithmic(offset)
experiment_db[sample].logarithmic = progress_data
raw_source.data = pd.DataFrame(
data=dict(x=progress_data['x'],
y=progress_data['y'],
yr=progress_data['resi'],
yfit=progress_data['yfit'])).to_dict('list')
# model analysis
if len(list(experiment_df)) > 2:
model_dict = ck.kinetic_model(experiment_db)
model_result = model_dict.model(subtract, transform, threshold,
bottom, top, slope, scalex, offset)
model_data_source.data = pd.DataFrame(
data=dict(xt=model_result['xt'],
yt=model_result['yt'],
et=model_result['et'],
n=model_result['n'],
ct=model_result['ct'])).to_dict('list')
model_plot_source.data = pd.DataFrame(
data=dict(xp=model_result['xp'],
yp=model_result['yp'],
u=model_result['u'],
l=model_result['l'],
ep=model_result['ep'],
cp=model_result['cp'])).to_dict('list')
model_fit_source.data = pd.DataFrame(
data=dict(x=model_result['xfit'],
y=model_result['yfit'])).to_dict('list')
if experiment_db['model'] == 'Michaelis-Menten':
mm_source.data = pd.DataFrame(
data=dict(label=['Fit Value', 'Std. Error'],
Km=model_result['Km'],
Vmax=model_result['Vmax']),
index=['value', 'error']).to_dict('list')
ic_source.data = pd.DataFrame(
data=dict(label=[], Bottom=[], Top=[], Slope=[],
p50=[])).to_dict('list')
model.xaxis.axis_label = 'Concentration'
elif experiment_db['model'] == 'pEC50/pIC50':
ic_source.data = pd.DataFrame(
data=dict(label=['Fit Value', 'Std. Error'],
Bottom=model_result['Bottom'],
Top=model_result['Top'],
Slope=model_result['Slope'],
p50=model_result['p50']),
index=['value', 'error']).to_dict('list')
mm_source.data = pd.DataFrame(
data=dict(label=[], Km=[], Vmax=[])).to_dict('list')
model.xaxis.axis_label = 'Log10(Concentration)'
else:
mm_source.data = pd.DataFrame(
data=dict(label=[], Km=[], Vmax=[])).to_dict('list')
ic_source.data = pd.DataFrame(
data=dict(label=[], Bottom=[], Top=[], Slope=[],
p50=[])).to_dict('list')
model.xaxis.axis_label = 'Sample #'
def update():
# get selections
fit_routine = fit_button.active
scalex = len(scalex_box.active)
model_eq = model_select.value
subtract = subtract_select.value
sample = sample_select.value
transform = transform_input.value
offset = offset_input.value
top = top_fix.value
bottom = bottom_fix.value
slope = slope_fix.value
threshold = threshold_slider.value
start = start_time.value
end = end_time.value
# update database
experiment_db[sample+'_fit'] = fit_routine
experiment_db['model'] = model_eq
# progress curve analysis
pdf = experiment_df[[experiment_df.columns[0], sample]]
experiment_db[sample] = ck.progress_curve(pdf)
if fit_routine == 0:
progress_data = experiment_db[sample].spline(start, end)
experiment_db[sample].spline = progress_data
elif fit_routine == 1:
progress_data = experiment_db[sample].linear(start, end)
experiment_db[sample].linear = progress_data
else:
try:
offset = float(offset)
except:
pass
progress_data = experiment_db[sample].logarithmic(start, end, offset)
experiment_db[sample].logarithmic = progress_data
raw_source.data = pd.DataFrame(data=dict(x=progress_data['x'],
y=progress_data['y'],
yr=progress_data['resi'],
yfit=progress_data['yfit'])).to_dict('list')
# model analysis
if len(list(experiment_df)) > 2:
model_dict = ck.kinetic_model(experiment_db)
model_result = model_dict.model(subtract, transform, threshold, bottom, top, slope, scalex)
model_data_source.data = pd.DataFrame(data=dict(xt=model_result['xt'], yt=model_result['yt'], et=model_result['et'], n=model_result['n'],
ct=model_result['ct'])).to_dict('list')
model_plot_source.data = pd.DataFrame(data=dict(xp=model_result['xp'], yp=model_result['yp'], u=model_result['u'], l=model_result['l'],
ep=model_result['ep'], cp=model_result['cp'])).to_dict('list')
model_fit_source.data = pd.DataFrame(data=dict(x=model_result['xfit'],
y=model_result['yfit'])).to_dict('list')
if experiment_db['model'] == 'Michaelis-Menten':
mm_source.data = pd.DataFrame(data=dict(label=['Fit Value', 'Std. Error'],
Km=model_result['Km'],
Vmax=model_result['Vmax']), index=['value', 'error']).to_dict('list')
ic_source.data = pd.DataFrame(data=dict(label=[], Bottom=[], Top=[], Slope=[], p50=[])).to_dict('list')
model.xaxis.axis_label = 'Concentration'
elif experiment_db['model'] == 'pEC50/pIC50':
ic_source.data = pd.DataFrame(data=dict(label=['Fit Value', 'Std. Error'],
Bottom=model_result['Bottom'], Top=model_result['Top'],
Slope=model_result['Slope'],
p50=model_result['p50']), index=['value', 'error']).to_dict('list')
mm_source.data = pd.DataFrame(data=dict(label=[], Km=[], Vmax=[])).to_dict('list')
model.xaxis.axis_label = 'Log10(Concentration)'
else:
mm_source.data = pd.DataFrame(data=dict(label=[], Km=[], Vmax=[])).to_dict('list')
ic_source.data = pd.DataFrame(data=dict(label=[], Bottom=[], Top=[], Slope=[], p50=[])).to_dict('list')
model.xaxis.axis_label = 'Sample #'
# update database
global experiment_db
experiment_db = dict(model='Michaelis-Menten')
xmin = experiment_df[experiment_df.columns[0]].values[0]
xmax = experiment_df[experiment_df.columns[0]].values[-1]
for s in experiment_df.columns[1:]:
if np.max(experiment_df[s]) > 0.:
df = experiment_df[[experiment_df.columns[0], s]].dropna(axis=0)
experiment_db[s] = ck.progress_curve(df, xmin, xmax)
experiment_db[s + '_fit'] = 0
# reload page
load_page(experiment_df, experiment_db)
########## import sample data ##########
experiment_file = join(dirname(__file__), 'test.csv')
experiment_df = pd.read_csv(experiment_file)
experiment_df.columns = [str(i) for i in list(experiment_df)]
experiment_db = dict(model='Michaelis-Menten')
xmin = experiment_df[experiment_df.columns[0]].values[0]
xmax = experiment_df[experiment_df.columns[0]].values[-1]
for s in experiment_df.columns[1:]:
df = experiment_df[[experiment_df.columns[0], s]]
experiment_db[s] = ck.progress_curve(df, xmin, xmax)
experiment_db[s + '_fit'] = 0
# load page
load_page(experiment_df, experiment_db)
global experiment_db
experiment_db = {'model' : 'Michaelis-Menten'}
xmin = experiment_df[experiment_df.columns[0]].values[0]
xmax = experiment_df[experiment_df.columns[0]].values[-1]
for s in experiment_df.columns[1:]:
if np.max(experiment_df[s]) > 0.:
df = experiment_df[[experiment_df.columns[0], s]].dropna(axis=0)
experiment_db[s] = ck.progress_curve(df)
experiment_db[s+'_fit'] = 0
experiment_db[s].spline(xmin, xmax)
# reload page
load_page(experiment_df, experiment_db)
########## import sample data ##########
experiment_file = join(dirname(__file__), 'test.csv')
experiment_df = pd.read_csv(experiment_file)
experiment_df.columns = [str(i) for i in list(experiment_df)]
experiment_db = {'model' : 'Michaelis-Menten'}
xmin = experiment_df[experiment_df.columns[0]].values[0]
xmax = experiment_df[experiment_df.columns[0]].values[-1]
for s in experiment_df.columns[1:]:
df = experiment_df[[experiment_df.columns[0], s]]
experiment_db[s] = ck.progress_curve(df)
experiment_db[s+'_fit'] = 0
experiment_db[s].spline(xmin, xmax)
# load page
load_page(experiment_df, experiment_db)