Skip to content

Commit

Permalink
Line plotting on Plotting tab
Browse files Browse the repository at this point in the history
  • Loading branch information
@maikherbig
maikherbig authored Oct 12, 2021
1 parent 4d8ea14 commit c8fb685
Showing 1 changed file with 67 additions and 99 deletions.
166 changes: 67 additions & 99 deletions AIDeveloper/aid_backbone.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -2403,13 +2403,19 @@ def update_plottingTab(self):
for key in keys:
if type(rtdc_ds["events"][key])==h5py._hl.dataset.Dataset:
shape = rtdc_ds["events"][key].shape
if len(shape)==1: #images have special shape (2D arrays)
if len(shape)==1: #zero-dimensional info (single number per cell)
keys_0d.append(key)
elif len(shape)==2: #images have special shape (2D arrays)
elif len(shape)==2: #one-dimensional info (multiple numbers per cell)
keys_1d.append(key)
elif len(shape)==3: #images have special shape (2D arrays)
elif len(shape)==3: #two-dimensional info (images)
keys_2d.append(key)


#add the traces to the 1d features
if "trace" in keys:
for key_trace in list(rtdc_ds["events"]["trace"].keys()):
keys_1d.append(key_trace+" (RTFDC)")


#Sort keys_2d: "image" first; "mask" last
keys_2d.insert(0, keys_2d.pop(keys_2d.index("image")))
keys_2d.insert(len(keys_2d), keys_2d.pop(keys_2d.index("mask")))
Expand Down Expand Up @@ -2453,16 +2459,17 @@ def init_2d_options(self,keys_2d):
self.popup_2dOptions_ui = aid_frontend.Ui_2dOptions()
self.popup_2dOptions_ui.setupUi(self.popup_2dOptions,keys_2d) #open a popup

def init_1d_options(self,keys_2d):
def init_1d_options(self,keys_1d):
self.popup_1dOptions = MyPopup()
self.popup_1dOptions_ui = aid_frontend.Ui_2dOptions()
self.popup_1dOptions_ui.setupUi(self.popup_1dOptions,keys_2d) #open a popup
self.popup_1dOptions_ui = aid_frontend.Ui_1dOptions()
self.popup_1dOptions_ui.setupUi(self.popup_1dOptions,keys_1d) #open a popup

def show_contour_options():
self.contour_options_nr += 1
print("Work in progress")

def show_centroid_options(self):
print("Work in progress")
self.centroid_options_nr += 1
#self.popup_layercontrols_ui.pushButton_close.clicked.connect(self.visualization_settings)
if self.centroid_options_nr==1:
Expand Down Expand Up @@ -2490,17 +2497,12 @@ def show_1d_options(self):
self.show_1d_options_nr += 1
#self.popup_layercontrols_ui.pushButton_close.clicked.connect(self.visualization_settings)
if self.show_1d_options_nr==1:
for iterator in range(len(self.popup_2dOptions_ui.spinBox_minChX)):
slider = self.popup_2dOptions_ui.horizontalSlider_chX[iterator]
slider.startValueChanged.connect(lambda _, b=None: self.put_image(ind=b))
slider.endValueChanged.connect(lambda _, b=None: self.put_image(ind=b))
checkBox = self.popup_2dOptions_ui.checkBox_show_chX[iterator]
checkBox.stateChanged.connect(lambda _, b=None: self.put_image(ind=b))
comboBox = self.popup_2dOptions_ui.comboBox_cmap_chX[iterator]
comboBox.currentIndexChanged.connect(lambda _, b=None: self.put_image(ind=b))
checkBox = self.popup_2dOptions_ui.checkBox_auto_chX[iterator]
checkBox.stateChanged.connect(lambda _, b=None: self.put_image(ind=b))
self.popup_2dOptions.show()
for iterator in range(len(self.popup_1dOptions_ui.checkBox_show_chX)):
checkBox = self.popup_1dOptions_ui.checkBox_show_chX[iterator]
checkBox.stateChanged.connect(lambda _, b=None: self.put_line(index=b))
comboBox = self.popup_1dOptions_ui.comboBox_cmap_chX[iterator]
comboBox.clicked.connect(lambda _, b=None: self.put_line(index=b))
self.popup_1dOptions.show()


def activate_deactivate_spinbox(self,newstate):
Expand Down Expand Up @@ -2587,6 +2589,8 @@ def onClick(self,points,pointermethod):
self.changedbyuser = True

self.put_image(index)
self.put_line(index)


def put_image(self,ind):
#check that the user is looking at the plotting tab
Expand Down Expand Up @@ -2725,92 +2729,56 @@ def put_image(self,ind):
self.plot_contour = pg.PlotCurveItem(contour[:,0],contour[:,1],width=6,pen="r")
self.widget_showCell.getView().addItem(self.plot_contour)


def put_line(self,index):
curr_ind = self.tabWidget_Modelbuilder.currentIndex()
if curr_ind!=3:
return

#Fluorescence traces: clear first
try:
self.plot_fl_trace_.clear() #clear the plot
self.plot_fl_trace.clear() #clear the plot
except:
pass
if "trace" in list(rtdc_ds["events"].keys()):
#Show the flourescence traces
trace = rtdc_ds["events"]["trace"]
fl_keys = list(trace.keys())
feature_keys = list(rtdc_ds["events"].keys())
fl1_max,fl1_pos,fl2_max,fl2_pos,fl3_max,fl3_pos = 0,0,0,0,0,0
Traces_flx = []
for i in range(len(fl_keys)):
if "fl1_median" in fl_keys[i] and self.checkBox_fl1.isChecked():
trace_flx = trace[fl_keys[i]][index]
pencolor = "g"
Traces_flx.append(trace_flx)
self.plot_fl_trace_ = self.plot_fl_trace.plot(range(len(trace_flx)),trace_flx,width=6,pen=pencolor,clear=False)
if "fl1_max" in feature_keys and "fl1_pos" in feature_keys: #if also the maxima and position of the max are available: use it to put the region accordingly
fl1_max,fl1_pos = rtdc_ds["events"]["fl1_max"][index],rtdc_ds["events"]["fl1_pos"][index]
elif "fl2_median" in fl_keys[i] and self.checkBox_fl2.isChecked():
trace_flx = trace[fl_keys[i]][index]
Traces_flx.append(trace_flx)
pencolor = (255,128,0) #orange
self.plot_fl_trace_ = self.plot_fl_trace.plot(range(len(trace_flx)),trace_flx,width=6,pen=pencolor,clear=False)
if "fl2_max" in feature_keys and "fl2_pos" in feature_keys: #if also the maxima and position of the max are available: use it to put the region accordingly
fl2_max,fl2_pos = rtdc_ds["events"]["fl2_max"][index],rtdc_ds["events"]["fl2_pos"][index]
elif "fl3_median" in fl_keys[i] and self.checkBox_fl3.isChecked():
trace_flx = trace[fl_keys[i]][index]
Traces_flx.append(trace_flx)
pencolor = "r"
self.plot_fl_trace_ = self.plot_fl_trace.plot(range(len(trace_flx)),trace_flx,width=6,pen=pencolor,clear=False)
if "fl3_max" in feature_keys and "fl3_pos" in feature_keys: #if also the maxima and position of the max are available: use it to put the region accordingly
fl3_max,fl3_pos = rtdc_ds["events"]["fl3_max"][index],rtdc_ds["events"]["fl3_pos"][index]

#get the maximum of [fl1_max,fl2_max,fl3_max] and put the region to the corresponding fl-position
ind = np.argmax(np.array([fl1_max,fl2_max,fl3_max]))
region_pos = np.array([fl1_pos,fl2_pos,fl3_pos])[ind] #this region is already given in us. translate this back to range
peak_height = np.array([fl1_max,fl2_max,fl3_max])[ind]
sample_rate = rtdc_ds.attrs["fluorescence:sample rate"]
fl_pos_ind = float((sample_rate*region_pos))/1E6 #
#Indicate the used flx_max and flx_pos by a scatter dot
self.peak_dot = self.plot_fl_trace.plot([float(fl_pos_ind)], [float(peak_height)],pen=None,symbol='o',symbolPen='w',clear=False)

#Place a LinearRegionItem on the plot. FL_max is searched in this region
#Search, if a region_width was defined already
if not hasattr(self, 'region_width'): #if there was no region_width defined yet...
# self.region_width = 50 #width of the region in us
# self.region_width = float((sample_rate*self.region_width))/1E6 #
#to get a reasonable initial range, use 20% of the nr. of availeble samples
samples_per_event = self.rtdc_ds.attrs["fluorescence:samples per event"]
self.region_width = 0.2*samples_per_event #width of the region in samples
#Convert to SI unit:
sample_rate = self.rtdc_ds.attrs["fluorescence:sample rate"]
self.region_width = (float(self.region_width)/float(sample_rate))*1E6 #range[samples]*(1/sample_rate[1/s]) = range[s]; div by 1E6 to convert to us

region_width_samples = (self.region_width*float(sample_rate))/1E6
self.region = pg.LinearRegionItem([fl_pos_ind-region_width_samples/2.0, fl_pos_ind+region_width_samples/2.0], bounds=[-20,33000], movable=True)
self.plot_fl_trace.addItem(self.region)
def region_changed():
#delete the current maximum indicator
self.plot_fl_trace.removeItem(self.peak_dot)
#where did the user drag the region to?
new_region = self.region.getRegion()
#for each fl-trace, search for the maximum in that region
Fl_max_pos_,Fl_max_ = [],[]
for i in range(len(Traces_flx)):
trace_flx = Traces_flx[i]
trace_flx_ = trace_flx[int(new_region[0]):int(new_region[1])]
trace_flx_pos_ = range(len(trace_flx))[int(new_region[0]):int(new_region[1])]
ind = np.argmax(trace_flx_)
fl_max_ = trace_flx_[ind]
fl_max_pos_ = trace_flx_pos_[ind]
Fl_max_pos_.append(fl_max_pos_)
Fl_max_.append(fl_max_)
#Get the highest maximum across all the traces
ind = np.argmax(np.array(Fl_max_))
fl_max__ = Fl_max_[ind]
fl_max_pos__ = Fl_max_pos_[ind]
self.new_peak = {"fl_pos":fl_max_pos__,"fl_max":fl_max__,"pos_x":pos_x}
self.new_peaks.append(self.new_peak)
self.peak_dot = self.plot_fl_trace.plot([float(fl_max_pos__)], [float(fl_max__)],pen=None,symbol='o',symbolPen='w',clear=False)

self.region.sigRegionChangeFinished.connect(region_changed)

if index==None:
index = int(self.spinBox_cellInd.value())

rtdc_ds = self.rtdc_ds
feature_keys = list(rtdc_ds.keys())

#which features shouldbe displayed
features_nr = len(self.popup_1dOptions_ui.checkBox_show_chX)
keys_1d = [self.popup_1dOptions_ui.checkBox_show_chX[i].text() for i in range(features_nr)]
keys_1d_on = [self.popup_1dOptions_ui.checkBox_show_chX[i].isChecked() for i in range(features_nr)]
colors = [self.popup_1dOptions_ui.comboBox_cmap_chX[i].palette().button().color() for i in range(features_nr)]
colors = [list(c.getRgb()) for c in colors]
colors = [tuple(c) for c in colors]
ind = np.where(np.array(keys_1d_on)==True)[0]
keys_1d = list(np.array(keys_1d)[ind])
colors = list(np.array(colors)[ind])

for key_1d,color in zip(keys_1d,colors):
if key_1d.endswith(" (RTFDC)"):
key_1d = key_1d.split(" (RTFDC)")[0]
trace_flx = rtdc_ds["events"]["trace"][key_1d][index]
pencolor = pg.mkPen(color, width=2)
self.plot_fl_trace_ = self.plot_fl_trace.plot(range(len(trace_flx)),trace_flx,width=6,pen=pencolor,clear=False)
# if "fl1_max" in feature_keys and "fl1_pos" in feature_keys: #if also the maxima and position of the max are available: use it to put the region accordingly
# fl1_max,fl1_pos = rtdc_ds["events"]["fl1_max"][index],rtdc_ds["events"]["fl1_pos"][index]
else:
values = rtdc_ds["events"][key_1d][index]
pencolor = pg.mkPen(color, width=2)
self.plot_fl_trace_ = self.plot_fl_trace.plot(range(len(trace_flx)),trace_flx,width=6,pen=pencolor,clear=False)

#get the maximum of [fl1_max,fl2_max,fl3_max] and put the region to the corresponding fl-position
# ind = np.argmax(np.array([fl1_max,fl2_max,fl3_max]))
# region_pos = np.array([fl1_pos,fl2_pos,fl3_pos])[ind] #this region is already given in us. translate this back to range
# peak_height = np.array([fl1_max,fl2_max,fl3_max])[ind]
# sample_rate = rtdc_ds.attrs["fluorescence:sample rate"]
# fl_pos_ind = float((sample_rate*region_pos))/1E6 #
# #Indicate the used flx_max and flx_pos by a scatter dot
# self.peak_dot = self.plot_fl_trace.plot([float(fl_pos_ind)], [float(peak_height)],pen=None,symbol='o',symbolPen='w',clear=False)

def onScatterClick(self,event, points):
pointermethod = 'point'
Expand Down Expand Up @@ -2946,7 +2914,7 @@ def updateScatterPlot(self):
self.spinBox_cellInd.setMinimum(0)
self.spinBox_cellInd.setMaximum(len(self.feature_x)-1)


def selectPeakPos(self):
#Check if self.region exists
#If not, show a message and return:
Expand Down

0 comments on commit c8fb685

Please sign in to comment.