diff --git a/.gitignore b/.gitignore
index ba16070..c28577a 100644
--- a/.gitignore
+++ b/.gitignore
@@ -2,7 +2,7 @@
.vscode/
.idea/
.trunk/
-local_tests/
+.local_tests/
*.pyc
*.db
.gitignore
diff --git a/Docs/ChangeLog.md b/Docs/ChangeLog.md
index 5a239dc..521a444 100644
--- a/Docs/ChangeLog.md
+++ b/Docs/ChangeLog.md
@@ -1,3 +1,29 @@
+# Version: 1.118.2
+## Date: 2024-11-8
+### Changes:
+- Change: changed the "Sum normalization" to "Percentages Scaling" in the data preprossing part to avoid the confusion with the normalization method.
+
+
+# Version: 1.118.1
+## Date: 2024-11-7
+### Changes:
+- New: added 'half_same_trends' optional to extract the table of each group has the same trend (all positive or all negative non-NA values) and at east 50% of the values are non-NA of the result of group-control in condition.
+- Change: refined the layout of the setting window.
+
+
+# Version: 1.118.0
+## Date: 2024-11-7
+### Changes:
+- New: added a debug console to run any python code in the MetaX to modify the object (FOR DEVELOPER ONLY).
+- Change: Load the some Setting from last profile when open the MetaX so that the user doesn't need to set the setting every time.
+- Fix: Fixed the bug of when plot mean of heatmap while selcect the sub meta, the mean calculation was not correct.
+
+# Version: 1.117.2
+## Date: 2024-11-5
+### Changes:
+- Change: add a column of focus list to the taxa-function network table.
+
+
# Version: 1.117.1
## Date: 2024-11-5
### Changes:
diff --git a/metax/gui/main_gui.py b/metax/gui/main_gui.py
index bf0bdf5..cf0a19f 100644
--- a/metax/gui/main_gui.py
+++ b/metax/gui/main_gui.py
@@ -86,6 +86,7 @@
from metax.gui.metax_gui.extended_combo_box import ExtendedComboBox
from metax.gui.metax_gui.show_plt import ExportablePlotDialog
from metax.gui.metax_gui.input_window import InputWindow
+ from metax.gui.metax_gui.command_window import CommandWindow
from metax.gui.metax_gui.user_agreement_dialog import UserAgreementDialog
from metax.gui.metax_gui.settings_widget import SettingsWidget
from metax.gui.metax_gui.cmap_combo_box import CmapComboBox
@@ -129,6 +130,7 @@
from .metax_gui.extended_combo_box import ExtendedComboBox
from .metax_gui.show_plt import ExportablePlotDialog
from .metax_gui.input_window import InputWindow
+ from .metax_gui.command_window import CommandWindow
from .metax_gui.user_agreement_dialog import UserAgreementDialog
from .metax_gui.settings_widget import SettingsWidget
from .metax_gui.cmap_combo_box import CmapComboBox
@@ -234,6 +236,7 @@ def __init__(self, MainWindow):
self.actionSave_As.setIcon(qta.icon('mdi.content-save'))
self.actionExport_Log_File.setIcon(qta.icon('mdi.export'))
self.actionHide_Show_Console.setIcon(qta.icon('mdi.console'))
+ self.actionDebug_Console.setIcon(qta.icon('fa5b.dev'))
self.actionAny_Table_Mode.setIcon(qta.icon('mdi.table'))
self.actionCheck_Update.setIcon(qta.icon('mdi.update'))
self.actionSettings.setIcon(qta.icon('mdi.cog'))
@@ -251,6 +254,7 @@ def __init__(self, MainWindow):
self.actionExport_Log_File.triggered.connect(self.export_log_file)
self.console_visible = False
self.actionHide_Show_Console.triggered.connect(self.show_hide_console)
+ self.actionDebug_Console.triggered.connect(self.show_command_line_window)
self.actionAny_Table_Mode.triggered.connect(self.set_any_table_mode)
self.actionCheck_Update.triggered.connect(lambda: self.check_update(show_message=True, manual_check_trigger=True))
self.actionSettings.triggered.connect(self.show_settings_window)
@@ -713,6 +717,14 @@ def change_event_comboBox_condition_group(comboBox, group_name):
def show_settings_window(self):
+ def get_stat_mean_by_zero_dominant():
+ if hasattr(self, 'tfa.stat_mean_by_zero_dominant'):
+ return self.tfa.stat_mean_by_zero_dominant
+ elif self.settings.contains("stat_mean_by_zero_dominant") and self.settings.value("stat_mean_by_zero_dominant", type=bool):
+ return True
+ else:
+ return False
+
if self.settings_dialog is None:
self.settings_dialog = QDialog(self.MainWindow)
self.settings_dialog.setWindowTitle("Settings")
@@ -725,6 +737,7 @@ def show_settings_window(self):
parent=self.settings_dialog,
update_branch=self.update_branch,
auto_check_update=self.auto_check_update,
+ stat_mean_by_zero_dominant = get_stat_mean_by_zero_dominant(),
QSettings=self.settings,
)
settings_widget.update_mode_changed.connect(self.on_update_mode_changed)
@@ -741,7 +754,10 @@ def show_settings_window(self):
self.settings_dialog.setLayout(layout)
self.settings_dialog.show()
-
+
+ def show_command_line_window(self):
+ self.command_window = CommandWindow(self.MainWindow, main_gui=self)
+ self.command_window.show()
# handle the update mode changed from settings window
def on_update_mode_changed(self, mode):
@@ -768,11 +784,11 @@ def on_html_theme_changed(self, theme):
def on_stat_mean_by_zero_dominant_changed(self, mode):
# chcek if self.tfa exists
- if not hasattr(self, 'tfa'):
+ if not hasattr(self.tfa, 'stat_mean_by_zero_dominant'):
print("Please load the data first.")
return
-
- self.tfa.stat_mean_by_zero_dominant = mode
+ self.tfa.stat_mean_by_zero_dominant = mode
+ self.settings.setValue("stat_mean_by_zero_dominant", mode)
print(f"Stat mean by zero dominant changed to: {mode}")
def on_protein_infer_method_changed(self, method):
@@ -1798,8 +1814,11 @@ def run_after_set_multi_tables(self):
# add "protein" "Custom" to comboBoxs to plot
self.add_or_remove_protein_custom_label()
-
+ #set stat_mean_by_zero_dominant mode by QSettings
+ if self.settings.contains("stat_mean_by_zero_dominant"):
+ self.tfa.stat_mean_by_zero_dominant = self.settings.value("stat_mean_by_zero_dominant", type=bool)
+
# add tables to table dict
if self.table_dict == {}:
if self.tfa.any_df_mode:
@@ -2655,7 +2674,7 @@ def set_multi_table(self, restore_taxafunc=False, saved_obj=None):
"Standard Scaling (Z-Score)": "zscore",
"Min-Max Scaling": "minmax",
"Pareto Scaling": "pareto",
- "Normalization by sum": "sum",
+ "Percentages Scaling": "percentage",
}
normalize_method = normalize_dict[normalize_method]
transform_method = transform_dict[transform_method]
@@ -3616,10 +3635,13 @@ def plot_basic_list(self, plot_type='heatmap'):
df = dft
else:
df = dft.loc[self.basic_heatmap_list]
-
+ # Done for creating the dataframe for the heatmap #
try:
if plot_type == 'heatmap':
+ df, sample_to_group_dict = self.tfa.BasicStats.get_df_by_mean_and_submeta(df = df,
+ sub_meta = sub_meta,
+ plot_mean = plot_mean)
if row_cluster or (scale =='row'):
df = self.delete_zero_rows(df)
if col_cluster or (scale =='col'):
@@ -3634,14 +3656,14 @@ def plot_basic_list(self, plot_type='heatmap'):
return
else:
pass
-
+
# plot heatmap
self.show_message(f'Plotting {plot_type}...')
HeatmapPlot(self.tfa, **self.heatmap_params_dict).plot_basic_heatmap(df=df, title=title, fig_size=(int(width), int(height)),
scale=scale, row_cluster=row_cluster, col_cluster=col_cluster,
cmap=cmap, rename_taxa=rename_taxa, font_size=font_size,
- show_all_labels=show_all_labels, rename_sample=rename_sample,
- plot_mean = plot_mean, sub_meta = sub_meta, return_type = 'fig')
+ show_all_labels=show_all_labels, return_type = 'fig',
+ sample_to_group_dict = sample_to_group_dict)
elif plot_type == 'bar':
@@ -5586,7 +5608,9 @@ def plot_tflink_heatmap(self, return_type = 'fig'):
QMessageBox.warning(self.MainWindow, 'Warning', 'No data!, please reselect!')
return None
-
+ df, sample_to_group_dict = self.tfa.BasicStats.get_df_by_mean_and_submeta(df = df,
+ sub_meta = sub_meta,
+ plot_mean = plot_mean)
if row_cluster or (scale == 'row'):
df = self.delete_zero_rows(df)
@@ -5598,8 +5622,7 @@ def plot_tflink_heatmap(self, return_type = 'fig'):
fig_res = HeatmapPlot(self.tfa, **self.heatmap_params_dict).plot_basic_heatmap(df=df, title=title, fig_size=(int(width), int(height)),
scale=scale, row_cluster=row_cluster, col_cluster=col_cluster,
cmap=cmap, rename_taxa=rename_taxa, font_size=font_size, show_all_labels=show_all_labels,
- rename_sample=rename_sample, sub_meta=sub_meta,
- plot_mean=plot_mean, return_type = return_type
+ return_type = return_type, sample_to_group_dict = sample_to_group_dict
)
if return_type == 'table':
@@ -5632,8 +5655,11 @@ def delete_zero_columns(self, dataframe):
zero_columns = dataframe.columns[(dataframe == 0).all(axis=0)]
if not zero_columns.empty:
dataframe = dataframe.drop(zero_columns, axis=1)
- # add group name to zero_columns
- zero_columns = [f'{i} ({self.tfa.get_group_of_a_sample(i)})' for i in zero_columns]
+ # show the message with group name
+ try: # add group name to zero_columns if possible
+ zero_columns = [f'{i} ({self.tfa.get_group_of_a_sample(i)})' for i in zero_columns]
+ except Exception:
+ print('The column name is not a sample name, Skip adding group name to the column name!')
col_str = '\n'.join(zero_columns)
if len(zero_columns) > 10:
# use InputWindow to show the deleted rows
diff --git a/metax/gui/metax_gui/command_window.py b/metax/gui/metax_gui/command_window.py
new file mode 100644
index 0000000..a114c3b
--- /dev/null
+++ b/metax/gui/metax_gui/command_window.py
@@ -0,0 +1,105 @@
+import sys
+import io
+from contextlib import redirect_stdout
+from PyQt5.QtWidgets import QApplication, QMainWindow, QTextEdit, QVBoxLayout, QWidget, QPushButton
+from PyQt5.QtCore import Qt, QEvent
+from PyQt5.QtGui import QTextCursor
+
+class PlainTextEditor(QTextEdit):
+ def insertFromMimeData(self, source):
+ if source.hasText():
+ self.insertPlainText(source.text())
+
+class OutputRedirector(io.StringIO):
+ def __init__(self, output_widget):
+ super().__init__()
+ self.output_widget = output_widget
+
+ def write(self, string):
+ super().write(string)
+ self.output_widget.append(string) # Append output to QTextEdit
+
+ def flush(self):
+ pass
+
+class CommandWindow(QMainWindow):
+ def __init__(self, parent=None, main_gui=None):
+ super(CommandWindow, self).__init__(parent)
+ self.main_gui = main_gui # Ensure main_gui is properly handled if None
+ self.initUI()
+ self.local_context = {'metax': main_gui} if main_gui else {}
+ self.history = []
+ self.history_index = 0
+
+ def initUI(self):
+ self.setWindowTitle('Debug Console')
+ self.resize(900, 600)
+ self.central_widget = QWidget()
+ self.setCentralWidget(self.central_widget)
+ layout = QVBoxLayout(self.central_widget)
+
+ self.output = QTextEdit()
+ self.output.setReadOnly(True)
+ layout.addWidget(self.output)
+
+ self.input = PlainTextEditor()
+ self.input.setFixedHeight(100)
+ layout.addWidget(self.input)
+
+ self.sendButton = QPushButton("Send")
+ self.sendButton.clicked.connect(self.process_command)
+ layout.addWidget(self.sendButton)
+
+ self.input.installEventFilter(self)
+
+ def process_command(self):
+ command = self.input.toPlainText().strip()
+ if command:
+ self.output.append(f"> {command}")
+ self.input.clear()
+ self.history.append(command)
+ self.history_index = len(self.history)
+
+ redirector = OutputRedirector(self.output) # 创建输出重定向器
+ with redirect_stdout(redirector): # 使用 redirect_stdout
+ try:
+ # 尝试作为表达式执行
+ result = eval(command, globals(), self.local_context)
+ if result is not None: # 如果有结果，显示它
+ print(result)
+ except SyntaxError:
+ # 如果表达式执行失败，尝试作为语句执行
+ try:
+ exec(command, globals(), self.local_context)
+ except Exception as e:
+ self.output.append(f"Error: {str(e)}")
+ except Exception as e:
+ self.output.append(f"Error: {str(e)}")
+
+ def eventFilter(self, source, event):
+ if source == self.input and event.type() == QEvent.KeyPress:
+ if event.key() == Qt.Key_Return:
+ if event.modifiers() & Qt.ShiftModifier:
+ self.input.insertPlainText('\n')
+ return True
+ else:
+ self.process_command()
+ return True
+ elif event.key() == Qt.Key_Up or event.key() == Qt.Key_Down:
+ if self.history_index > 0 and event.key() == Qt.Key_Up:
+ self.history_index -= 1
+ self.input.setText(self.history[self.history_index])
+ self.input.moveCursor(QTextCursor.End)
+ return True
+ elif self.history_index < len(self.history) - 1 and event.key() == Qt.Key_Down:
+ self.history_index += 1
+ self.input.setText(self.history[self.history_index])
+ self.input.moveCursor(QTextCursor.End)
+ return True
+ return super(CommandWindow, self).eventFilter(source, event)
+
+if __name__ == "__main__":
+ app = QApplication(sys.argv)
+ win = CommandWindow()
+ win.show()
+ sys.exit(app.exec_())
diff --git a/metax/gui/metax_gui/main_window.ui b/metax/gui/metax_gui/main_window.ui
index c6e3adb..68359a1 100644
--- a/metax/gui/metax_gui/main_window.ui
+++ b/metax/gui/metax_gui/main_window.ui
@@ -46,7 +46,7 @@
Qt::LeftToRight
- 3
+ 2
false
@@ -245,8 +245,8 @@
0
0
- 391
- 80
+ 528
+ 595
@@ -976,7 +976,7 @@
-
- Normalization by sum
+ Percentages Scaling
@@ -2789,8 +2789,8 @@
0
0
- 1016
- 184
+ 621
+ 152
@@ -3806,7 +3806,7 @@
0
0
- 878
+ 1020
128
@@ -4495,10 +4495,10 @@
- 3
+ 1
- 30.000000000000000
+ 99.000000000000000
@@ -4602,6 +4602,13 @@
16777215
+
+ - 'all_sig': DataFrame containing all significant rows across all groups, Non-significant values are replaced with NA.
+- 'half_same_trends': DataFrame containing rows where each group has the same trend (all positive or all negative non-NA values) and at least 50% of the values are non-NA.
+- 'no_na': DataFrame containing rows with no NA values in each group.
+- 'same_trends': DataFrame containing rows with no NA values, and all values in each group follow the same trend (all positive or all negative).
+
+
-
all_sig
@@ -4612,6 +4619,11 @@
no_na
+ -
+
+ half_same_trends
+
+
-
same_trends
@@ -4828,7 +4840,7 @@
QTabWidget::Triangular
- 2
+ 3
@@ -5700,8 +5712,8 @@
0
0
- 535
- 94
+ 996
+ 164
@@ -5906,7 +5918,7 @@
1
- 20.000000000000000
+ 99.000000000000000
@@ -7481,8 +7493,8 @@
0
0
- 620
- 65
+ 1016
+ 166
@@ -9302,8 +9314,8 @@
0
0
- 383
- 68
+ 1016
+ 179
@@ -10470,8 +10482,12 @@
Dev
+
+
+
+
@@ -10549,6 +10565,11 @@
Tutorial
+
+
+ Debug Console
+
+
comboBox_taxa_level_to_stast
diff --git a/metax/gui/metax_gui/setting_window.ui b/metax/gui/metax_gui/setting_window.ui
index 3851027..28ae700 100644
--- a/metax/gui/metax_gui/setting_window.ui
+++ b/metax/gui/metax_gui/setting_window.ui
@@ -17,61 +17,15 @@
-
- 1
+ 0
-
-
-
- 0
- 0
- 748
- 340
-
-
-
- General
-
-
-
-
-
-
-
-
-
- Auto Check Update
-
-
- true
-
-
-
- -
-
-
- Stable
-
-
- true
-
-
-
- -
-
-
- Beta
-
-
-
-
-
-
-
0
0
748
- 340
+ 367
@@ -739,7 +693,7 @@
0
0
748
- 340
+ 367
@@ -778,6 +732,37 @@
+ -
+
+
-
+
+
+ Auto Check Update
+
+
+ true
+
+
+
+ -
+
+
+ Stable
+
+
+ true
+
+
+
+ -
+
+
+ Beta
+
+
+
+
+
diff --git a/metax/gui/metax_gui/settings_widget.py b/metax/gui/metax_gui/settings_widget.py
index 53ebfb7..ceb9146 100644
--- a/metax/gui/metax_gui/settings_widget.py
+++ b/metax/gui/metax_gui/settings_widget.py
@@ -11,7 +11,7 @@ class SettingsWidget(QWidget):
protein_infer_method_changed = pyqtSignal(str)
stat_mean_by_zero_dominant_changed = pyqtSignal(bool)
- def __init__(self, parent=None, update_branch="main", auto_check_update=True, QSettings=None):
+ def __init__(self, parent=None, update_branch="main", auto_check_update=True, stat_mean_by_zero_dominant=False, QSettings=None):
super().__init__(parent)
self.update_mode = update_branch
self.auto_check_update = auto_check_update
@@ -22,7 +22,7 @@ def __init__(self, parent=None, update_branch="main", auto_check_update=True, QS
self.ui = Ui_Settings()
self.ui.setupUi(self)
- self.init_ui(self.update_mode, self.auto_check_update, QSettings)
+ self.init_ui(self.update_mode, self.auto_check_update, stat_mean_by_zero_dominant, QSettings)
# resize the window, 800 as default
self.resize(800, 400)
@@ -70,13 +70,16 @@ def __init__(self, parent=None, update_branch="main", auto_check_update=True, QS
self.ui.checkBox_stat_mean_by_zero_dominant.stateChanged.connect(self.handle_stat_mean_by_zero_dominant_changed)
- def init_ui(self, update_mode, auto_check_update, QSettings=None):
+ def init_ui(self, update_mode, auto_check_update, stat_mean_by_zero_dominant, QSettings=None,):
if update_mode == "main":
self.ui.radioButton_update_stable.setChecked(True)
elif update_mode == "dev":
self.ui.radioButton_update_beta.setChecked(True)
self.ui.checkBox_auto_check_update.setChecked(auto_check_update)
+ # set the default values for stat_mean_by_zero_dominant
+ self.ui.checkBox_stat_mean_by_zero_dominant.setChecked(stat_mean_by_zero_dominant)
+
if QSettings:
method = QSettings.value('protein_infer_greedy_mode', 'fast')
selected_method = 'normal' if method == 'greedy' else 'fast'
diff --git a/metax/gui/metax_gui/ui_main_window.py b/metax/gui/metax_gui/ui_main_window.py
index 0e713a3..f4a0cfc 100644
--- a/metax/gui/metax_gui/ui_main_window.py
+++ b/metax/gui/metax_gui/ui_main_window.py
@@ -147,7 +147,7 @@ def setupUi(self, metaX_main):
self.toolBox_2.setMaximumSize(QtCore.QSize(1677, 16777215))
self.toolBox_2.setObjectName("toolBox_2")
self.page_2 = QtWidgets.QWidget()
- self.page_2.setGeometry(QtCore.QRect(0, 0, 391, 80))
+ self.page_2.setGeometry(QtCore.QRect(0, 0, 528, 595))
self.page_2.setObjectName("page_2")
self.gridLayout_27 = QtWidgets.QGridLayout(self.page_2)
self.gridLayout_27.setObjectName("gridLayout_27")
@@ -1411,7 +1411,7 @@ def setupUi(self, metaX_main):
self.scrollArea_2.setWidgetResizable(True)
self.scrollArea_2.setObjectName("scrollArea_2")
self.scrollAreaWidgetContents_2 = QtWidgets.QWidget()
- self.scrollAreaWidgetContents_2.setGeometry(QtCore.QRect(0, 0, 1016, 184))
+ self.scrollAreaWidgetContents_2.setGeometry(QtCore.QRect(0, 0, 621, 152))
self.scrollAreaWidgetContents_2.setObjectName("scrollAreaWidgetContents_2")
self.gridLayout_50 = QtWidgets.QGridLayout(self.scrollAreaWidgetContents_2)
self.gridLayout_50.setObjectName("gridLayout_50")
@@ -1967,7 +1967,7 @@ def setupUi(self, metaX_main):
self.scrollArea_cross_heatmap_settings.setWidgetResizable(True)
self.scrollArea_cross_heatmap_settings.setObjectName("scrollArea_cross_heatmap_settings")
self.scrollAreaWidgetContents_3 = QtWidgets.QWidget()
- self.scrollAreaWidgetContents_3.setGeometry(QtCore.QRect(0, 0, 878, 128))
+ self.scrollAreaWidgetContents_3.setGeometry(QtCore.QRect(0, 0, 1020, 128))
self.scrollAreaWidgetContents_3.setObjectName("scrollAreaWidgetContents_3")
self.gridLayout_38 = QtWidgets.QGridLayout(self.scrollAreaWidgetContents_3)
self.gridLayout_38.setObjectName("gridLayout_38")
@@ -2263,8 +2263,8 @@ def setupUi(self, metaX_main):
self.doubleSpinBox_max_log2fc_heatmap.setEnabled(False)
self.doubleSpinBox_max_log2fc_heatmap.setMinimumSize(QtCore.QSize(0, 0))
self.doubleSpinBox_max_log2fc_heatmap.setMaximumSize(QtCore.QSize(16777215, 16777215))
- self.doubleSpinBox_max_log2fc_heatmap.setDecimals(3)
- self.doubleSpinBox_max_log2fc_heatmap.setProperty("value", 30.0)
+ self.doubleSpinBox_max_log2fc_heatmap.setDecimals(1)
+ self.doubleSpinBox_max_log2fc_heatmap.setProperty("value", 99.0)
self.doubleSpinBox_max_log2fc_heatmap.setObjectName("doubleSpinBox_max_log2fc_heatmap")
self.horizontalLayout_12.addWidget(self.doubleSpinBox_max_log2fc_heatmap)
self.gridLayout_51.addLayout(self.horizontalLayout_12, 3, 2, 1, 1)
@@ -2309,6 +2309,7 @@ def setupUi(self, metaX_main):
self.comboBox_cross_3_level_plot_df_type.addItem("")
self.comboBox_cross_3_level_plot_df_type.addItem("")
self.comboBox_cross_3_level_plot_df_type.addItem("")
+ self.comboBox_cross_3_level_plot_df_type.addItem("")
self.gridLayout_51.addWidget(self.comboBox_cross_3_level_plot_df_type, 3, 4, 1, 1)
self.checkBox_cross_3_level_plot_remove_zero_col = QtWidgets.QCheckBox(self.scrollAreaWidgetContents_3)
sizePolicy = QtWidgets.QSizePolicy(QtWidgets.QSizePolicy.Preferred, QtWidgets.QSizePolicy.Fixed)
@@ -2901,7 +2902,7 @@ def setupUi(self, metaX_main):
self.scrollArea_3.setWidgetResizable(True)
self.scrollArea_3.setObjectName("scrollArea_3")
self.scrollAreaWidgetContents_4 = QtWidgets.QWidget()
- self.scrollAreaWidgetContents_4.setGeometry(QtCore.QRect(0, 0, 535, 94))
+ self.scrollAreaWidgetContents_4.setGeometry(QtCore.QRect(0, 0, 996, 164))
self.scrollAreaWidgetContents_4.setObjectName("scrollAreaWidgetContents_4")
self.gridLayout_68 = QtWidgets.QGridLayout(self.scrollAreaWidgetContents_4)
self.gridLayout_68.setObjectName("gridLayout_68")
@@ -3014,7 +3015,7 @@ def setupUi(self, metaX_main):
sizePolicy.setHeightForWidth(self.doubleSpinBox_deseq2_log2fc_max.sizePolicy().hasHeightForWidth())
self.doubleSpinBox_deseq2_log2fc_max.setSizePolicy(sizePolicy)
self.doubleSpinBox_deseq2_log2fc_max.setDecimals(1)
- self.doubleSpinBox_deseq2_log2fc_max.setProperty("value", 20.0)
+ self.doubleSpinBox_deseq2_log2fc_max.setProperty("value", 99.0)
self.doubleSpinBox_deseq2_log2fc_max.setObjectName("doubleSpinBox_deseq2_log2fc_max")
self.gridLayout_53.addWidget(self.doubleSpinBox_deseq2_log2fc_max, 0, 7, 1, 1)
self.label_156 = QtWidgets.QLabel(self.scrollAreaWidgetContents_4)
@@ -3847,7 +3848,7 @@ def setupUi(self, metaX_main):
self.scrollArea_5.setWidgetResizable(True)
self.scrollArea_5.setObjectName("scrollArea_5")
self.scrollAreaWidgetContents_6 = QtWidgets.QWidget()
- self.scrollAreaWidgetContents_6.setGeometry(QtCore.QRect(0, 0, 620, 65))
+ self.scrollAreaWidgetContents_6.setGeometry(QtCore.QRect(0, 0, 1016, 166))
self.scrollAreaWidgetContents_6.setObjectName("scrollAreaWidgetContents_6")
self.gridLayout_57 = QtWidgets.QGridLayout(self.scrollAreaWidgetContents_6)
self.gridLayout_57.setObjectName("gridLayout_57")
@@ -4856,7 +4857,7 @@ def setupUi(self, metaX_main):
self.scrollArea_7.setWidgetResizable(True)
self.scrollArea_7.setObjectName("scrollArea_7")
self.scrollAreaWidgetContents_8 = QtWidgets.QWidget()
- self.scrollAreaWidgetContents_8.setGeometry(QtCore.QRect(0, 0, 383, 68))
+ self.scrollAreaWidgetContents_8.setGeometry(QtCore.QRect(0, 0, 1016, 179))
self.scrollAreaWidgetContents_8.setObjectName("scrollAreaWidgetContents_8")
self.gridLayout_66 = QtWidgets.QGridLayout(self.scrollAreaWidgetContents_8)
self.gridLayout_66.setObjectName("gridLayout_66")
@@ -5464,6 +5465,8 @@ def setupUi(self, metaX_main):
self.actionSettings.setObjectName("actionSettings")
self.actionTutorial = QtWidgets.QAction(metaX_main)
self.actionTutorial.setObjectName("actionTutorial")
+ self.actionDebug_Console = QtWidgets.QAction(metaX_main)
+ self.actionDebug_Console.setObjectName("actionDebug_Console")
self.menuTools.addAction(self.actionTaxaFuncAnalyzer)
self.menuTools.addAction(self.actionPeptide_to_TaxaFunc)
self.menuTools.addAction(self.actionDatabase_Builder)
@@ -5475,8 +5478,12 @@ def setupUi(self, metaX_main):
self.menuOthers.addAction(self.actionRestore_From)
self.menuOthers.addAction(self.actionSave_As)
self.menuDev.addAction(self.actionExport_Log_File)
+ self.menuDev.addSeparator()
self.menuDev.addAction(self.actionHide_Show_Console)
+ self.menuDev.addAction(self.actionDebug_Console)
+ self.menuDev.addSeparator()
self.menuDev.addAction(self.actionAny_Table_Mode)
+ self.menuDev.addSeparator()
self.menuDev.addAction(self.actionSettings)
self.menuBar.addAction(self.menuTools.menuAction())
self.menuBar.addAction(self.menuOthers.menuAction())
@@ -5485,10 +5492,10 @@ def setupUi(self, metaX_main):
self.retranslateUi(metaX_main)
self.stackedWidget.setCurrentIndex(0)
- self.tabWidget_TaxaFuncAnalyzer.setCurrentIndex(3)
+ self.tabWidget_TaxaFuncAnalyzer.setCurrentIndex(2)
self.toolBox_2.setCurrentIndex(0)
self.tabWidget_4.setCurrentIndex(0)
- self.tabWidget_3.setCurrentIndex(2)
+ self.tabWidget_3.setCurrentIndex(3)
self.tabWidget.setCurrentIndex(1)
self.tabWidget_2.setCurrentIndex(1)
self.tabWidget_6.setCurrentIndex(0)
@@ -5632,7 +5639,7 @@ def retranslateUi(self, metaX_main):
self.comboBox_set_data_normalization.setItemText(3, _translate("metaX_main", "Min-Max Scaling"))
self.comboBox_set_data_normalization.setItemText(4, _translate("metaX_main", "Pareto Scaling"))
self.comboBox_set_data_normalization.setItemText(5, _translate("metaX_main", "Mean centering"))
- self.comboBox_set_data_normalization.setItemText(6, _translate("metaX_main", "Normalization by sum"))
+ self.comboBox_set_data_normalization.setItemText(6, _translate("metaX_main", "Percentages Scaling"))
self.comboBox_remove_batch_effect.setItemText(0, _translate("metaX_main", "None"))
self.label_41.setText(_translate("metaX_main", "Data Normalization"))
self.label_43.setText(_translate("metaX_main", "Batch Effect Correction"))
@@ -5865,9 +5872,15 @@ def retranslateUi(self, metaX_main):
self.label_139.setText(_translate("metaX_main", "To"))
self.label_58.setText(_translate("metaX_main", "Top Number"))
self.label_141.setText(_translate("metaX_main", "Plot Type"))
+ self.comboBox_cross_3_level_plot_df_type.setToolTip(_translate("metaX_main", "- \'all_sig\': DataFrame containing all significant rows across all groups, Non-significant values are replaced with NA.\n"
+"- \'half_same_trends\': DataFrame containing rows where each group has the same trend (all positive or all negative non-NA values) and at least 50% of the values are non-NA.\n"
+"- \'no_na\': DataFrame containing rows with no NA values in each group.\n"
+"- \'same_trends\': DataFrame containing rows with no NA values, and all values in each group follow the same trend (all positive or all negative).\n"
+""))
self.comboBox_cross_3_level_plot_df_type.setItemText(0, _translate("metaX_main", "all_sig"))
self.comboBox_cross_3_level_plot_df_type.setItemText(1, _translate("metaX_main", "no_na"))
- self.comboBox_cross_3_level_plot_df_type.setItemText(2, _translate("metaX_main", "same_trends"))
+ self.comboBox_cross_3_level_plot_df_type.setItemText(2, _translate("metaX_main", "half_same_trends"))
+ self.comboBox_cross_3_level_plot_df_type.setItemText(3, _translate("metaX_main", "same_trends"))
self.checkBox_cross_3_level_plot_remove_zero_col.setText(_translate("metaX_main", "Remove Zero Col"))
self.label_30.setText(_translate("metaX_main", "with"))
self.comboBox_top_heatmap_scale_method.setItemText(0, _translate("metaX_main", "maxmin"))
@@ -6295,3 +6308,4 @@ def retranslateUi(self, metaX_main):
self.actionAny_Table_Mode.setText(_translate("metaX_main", "Any Table Mode"))
self.actionSettings.setText(_translate("metaX_main", "Settings"))
self.actionTutorial.setText(_translate("metaX_main", "Tutorial"))
+ self.actionDebug_Console.setText(_translate("metaX_main", "Debug Console"))
diff --git a/metax/gui/metax_gui/ui_setting_window.py b/metax/gui/metax_gui/ui_setting_window.py
index a9108fb..83c9c45 100644
--- a/metax/gui/metax_gui/ui_setting_window.py
+++ b/metax/gui/metax_gui/ui_setting_window.py
@@ -1,6 +1,6 @@
# -*- coding: utf-8 -*-
-# Form implementation generated from reading ui file 'c:\Users\Qing\OneDrive - University of Ottawa\code\TaxaFunc\MetaX\metax\gui\metax_gui\setting_window.ui'
+# Form implementation generated from reading ui file 'c:\Users\max\OneDrive - University of Ottawa\code\TaxaFunc\MetaX\metax\gui\metax_gui\setting_window.ui'
#
# Created by: PyQt5 UI code generator 5.15.9
#
@@ -19,28 +19,8 @@ def setupUi(self, Settings):
self.gridLayout.setObjectName("gridLayout")
self.toolBox = QtWidgets.QToolBox(Settings)
self.toolBox.setObjectName("toolBox")
- self.page = QtWidgets.QWidget()
- self.page.setGeometry(QtCore.QRect(0, 0, 748, 340))
- self.page.setObjectName("page")
- self.gridLayout_3 = QtWidgets.QGridLayout(self.page)
- self.gridLayout_3.setObjectName("gridLayout_3")
- self.gridLayout_2 = QtWidgets.QGridLayout()
- self.gridLayout_2.setObjectName("gridLayout_2")
- self.checkBox_auto_check_update = QtWidgets.QCheckBox(self.page)
- self.checkBox_auto_check_update.setChecked(True)
- self.checkBox_auto_check_update.setObjectName("checkBox_auto_check_update")
- self.gridLayout_2.addWidget(self.checkBox_auto_check_update, 0, 0, 1, 1)
- self.radioButton_update_stable = QtWidgets.QRadioButton(self.page)
- self.radioButton_update_stable.setChecked(True)
- self.radioButton_update_stable.setObjectName("radioButton_update_stable")
- self.gridLayout_2.addWidget(self.radioButton_update_stable, 0, 1, 1, 1)
- self.radioButton_update_beta = QtWidgets.QRadioButton(self.page)
- self.radioButton_update_beta.setObjectName("radioButton_update_beta")
- self.gridLayout_2.addWidget(self.radioButton_update_beta, 0, 2, 1, 1)
- self.gridLayout_3.addLayout(self.gridLayout_2, 0, 0, 1, 1)
- self.toolBox.addItem(self.page, "")
self.page_2 = QtWidgets.QWidget()
- self.page_2.setGeometry(QtCore.QRect(0, 0, 748, 340))
+ self.page_2.setGeometry(QtCore.QRect(0, 0, 748, 367))
self.page_2.setObjectName("page_2")
self.gridLayout_4 = QtWidgets.QGridLayout(self.page_2)
self.gridLayout_4.setObjectName("gridLayout_4")
@@ -303,7 +283,7 @@ def setupUi(self, Settings):
self.gridLayout_4.addLayout(self.gridLayout_12, 2, 3, 1, 1)
self.toolBox.addItem(self.page_2, "")
self.page_3 = QtWidgets.QWidget()
- self.page_3.setGeometry(QtCore.QRect(0, 0, 748, 340))
+ self.page_3.setGeometry(QtCore.QRect(0, 0, 748, 367))
self.page_3.setObjectName("page_3")
self.gridLayout_11 = QtWidgets.QGridLayout(self.page_3)
self.gridLayout_11.setObjectName("gridLayout_11")
@@ -321,20 +301,30 @@ def setupUi(self, Settings):
self.label_24.setObjectName("label_24")
self.gridLayout_10.addWidget(self.label_24, 0, 0, 1, 1)
self.gridLayout_11.addLayout(self.gridLayout_10, 1, 1, 1, 1)
+ self.gridLayout_2 = QtWidgets.QGridLayout()
+ self.gridLayout_2.setObjectName("gridLayout_2")
+ self.checkBox_auto_check_update = QtWidgets.QCheckBox(self.page_3)
+ self.checkBox_auto_check_update.setChecked(True)
+ self.checkBox_auto_check_update.setObjectName("checkBox_auto_check_update")
+ self.gridLayout_2.addWidget(self.checkBox_auto_check_update, 0, 0, 1, 1)
+ self.radioButton_update_stable = QtWidgets.QRadioButton(self.page_3)
+ self.radioButton_update_stable.setChecked(True)
+ self.radioButton_update_stable.setObjectName("radioButton_update_stable")
+ self.gridLayout_2.addWidget(self.radioButton_update_stable, 0, 1, 1, 1)
+ self.radioButton_update_beta = QtWidgets.QRadioButton(self.page_3)
+ self.radioButton_update_beta.setObjectName("radioButton_update_beta")
+ self.gridLayout_2.addWidget(self.radioButton_update_beta, 0, 2, 1, 1)
+ self.gridLayout_11.addLayout(self.gridLayout_2, 0, 0, 1, 2)
self.toolBox.addItem(self.page_3, "")
self.gridLayout.addWidget(self.toolBox, 0, 0, 1, 1)
self.retranslateUi(Settings)
- self.toolBox.setCurrentIndex(1)
+ self.toolBox.setCurrentIndex(0)
QtCore.QMetaObject.connectSlotsByName(Settings)
def retranslateUi(self, Settings):
_translate = QtCore.QCoreApplication.translate
Settings.setWindowTitle(_translate("Settings", "Settings"))
- self.checkBox_auto_check_update.setText(_translate("Settings", "Auto Check Update"))
- self.radioButton_update_stable.setText(_translate("Settings", "Stable"))
- self.radioButton_update_beta.setText(_translate("Settings", "Beta"))
- self.toolBox.setItemText(self.toolBox.indexOf(self.page), _translate("Settings", "General"))
self.label_4.setText(_translate("Settings", "Taxa-Functions Link Network"))
self.label_11.setText(_translate("Settings", "Line Width"))
self.label_15.setToolTip(_translate("Settings", "The larger the value the greater the repulsion"))
@@ -418,4 +408,7 @@ def retranslateUi(self, Settings):
self.comboBox_protein_infer_greedy_mode.setItemText(0, _translate("Settings", "fast"))
self.comboBox_protein_infer_greedy_mode.setItemText(1, _translate("Settings", "normal"))
self.label_24.setText(_translate("Settings", "Greedy Mode in Razor Method"))
+ self.checkBox_auto_check_update.setText(_translate("Settings", "Auto Check Update"))
+ self.radioButton_update_stable.setText(_translate("Settings", "Stable"))
+ self.radioButton_update_beta.setText(_translate("Settings", "Beta"))
self.toolBox.setItemText(self.toolBox.indexOf(self.page_3), _translate("Settings", "Others"))
diff --git a/metax/taxafunc_analyzer/analyzer_utils/basic_stats.py b/metax/taxafunc_analyzer/analyzer_utils/basic_stats.py
index 5772024..b0fda3f 100644
--- a/metax/taxafunc_analyzer/analyzer_utils/basic_stats.py
+++ b/metax/taxafunc_analyzer/analyzer_utils/basic_stats.py
@@ -6,6 +6,26 @@ class BasicStats:
def __init__(self, tfa):
self.tfa = tfa
+ def _get_mean_by_zero_dominant(self, df: pd.DataFrame) -> pd.Series:
+ """
+ Optimized function to calculate the mean of non-zero values in each row if the number of zero values
+ is less than half of the total values; otherwise, return 0.
+
+ Args:
+ df (pd.DataFrame): Input DataFrame.
+
+ Returns:
+ pd.Series: A Series with mean values based on the zero-dominant condition.
+ """
+ # 计算每行的零值数量
+ zero_counts = (df == 0).sum(axis=1)
+ # 判断每行零值是否超过一半，超过的行直接设为0
+ mean_series = pd.Series(0, index=df.index)
+ non_zero_rows = zero_counts <= (df.shape[1] / 2)
+ # 对非零主导的行计算非零均值
+ mean_series[non_zero_rows] = df[non_zero_rows].replace(0, pd.NA).mean(axis=1, skipna=True)
+ return mean_series
+
# get a mean df by group
def get_stats_mean_df_by_group(self, df: pd.DataFrame, condition: list|None = None, zero_dominant: bool|None = None) -> pd.DataFrame:
"""
@@ -20,31 +40,12 @@ def get_stats_mean_df_by_group(self, df: pd.DataFrame, condition: list|None = No
pd.DataFrame: A DataFrame containing the mean values of the groups.
"""
- def get_mean_by_zero_dominant(df: pd.DataFrame) -> pd.Series:
- """
- Optimized function to calculate the mean of non-zero values in each row if the number of zero values
- is less than half of the total values; otherwise, return 0.
-
- Args:
- df (pd.DataFrame): Input DataFrame.
-
- Returns:
- pd.Series: A Series with mean values based on the zero-dominant condition.
- """
- # 计算每行的零值数量
- zero_counts = (df == 0).sum(axis=1)
- # 判断每行零值是否超过一半，超过的行直接设为0
- mean_series = pd.Series(0, index=df.index)
- non_zero_rows = zero_counts <= (df.shape[1] / 2)
- # 对非零主导的行计算非零均值
- mean_series[non_zero_rows] = df[non_zero_rows].replace(0, pd.NA).mean(axis=1, skipna=True)
- return mean_series
if zero_dominant is None:
zero_dominant = self.tfa.stat_mean_by_zero_dominant
print(f"Caculating mean by zero_dominant: [{zero_dominant}]")
- mean_method = get_mean_by_zero_dominant if zero_dominant else lambda x: x.mean(axis=1)
+ mean_method = self._get_mean_by_zero_dominant if zero_dominant else lambda x: x.mean(axis=1)
data = df.copy()
@@ -187,7 +188,8 @@ def get_combined_sub_meta_df(
sub_meta: str,
rename_sample: bool = False,
plot_mean: bool = False,
- ) -> tuple[pd.DataFrame, list[str]]:
+ zero_dominant: bool|None = None
+ ) -> tuple[pd.DataFrame, dict[str, str]]:
"""
Combines the sub-meta information with the main meta information in the given DataFrame and returns the combined DataFrame and a list of sub-meta groups.
@@ -196,9 +198,10 @@ def get_combined_sub_meta_df(
sub_meta (str): The sub-meta information to be combined with the main meta information.
rename_sample (bool, optional): Whether to rename the samples in the DataFrame. Defaults to False.
plot_mean (bool, optional): Whether to plot the mean values. Defaults to False.
+ zero_dominant (bool, optional): Whether to calculate the mean of non-zero values in each group(return 0 if the >50% values are zero). Defaults to None.
Returns:
- tuple[pd.DataFrame, list[str]]: A tuple containing the combined DataFrame and a list of sub-meta groups.
+ tuple[pd.DataFrame, Dict[str, str]]: A tuple containing the combined DataFrame and a dictionary with the sample names as keys and the group names as values.
"""
if sub_meta != 'None':
@@ -206,7 +209,14 @@ def get_combined_sub_meta_df(
sub_groups = {sample: self.tfa.get_group_of_a_sample(sample, sub_meta) for sample in df.columns}
# Combine samples with the same meta and sub-meta, and calculate the mean value
- grouped_data = df.T.groupby([sample_groups, sub_groups]).mean().T
+ if zero_dominant is None:
+ zero_dominant = self.tfa.stat_mean_by_zero_dominant
+ print(f"Caculating mean by zero_dominant: [{zero_dominant}]")
+
+ if zero_dominant:
+ grouped_data = df.T.groupby([sample_groups, sub_groups]).apply(lambda x: self._get_mean_by_zero_dominant(x.T)).T
+ else:
+ grouped_data = df.T.groupby([sample_groups, sub_groups]).mean().T
# group_list is the sub-meta group
group_list = [i[1] for i in grouped_data.columns] if not plot_mean else grouped_data.columns.tolist()
@@ -222,7 +232,10 @@ def get_combined_sub_meta_df(
else:
group_list = [self.tfa.get_group_of_a_sample(i) for i in df.columns] if not plot_mean else df.columns.tolist()
- return df, group_list
+ # create a group_dict, key is column name, value is the group name bsed on the group_list
+ sample_to_group_dict = {col: group_list[i] for i, col in enumerate(df.columns)}
+
+ return df, sample_to_group_dict
# Shapiro-Wilk Test
def shapiro_test(self, df: pd.DataFrame, alpha=0.05) :
@@ -246,3 +259,24 @@ def shapiro_test(self, df: pd.DataFrame, alpha=0.05) :
shapiro_results[sample] = {'p_value': p, 'is_normal': p > alpha}
return shapiro_results
+
+
+
+ def get_df_by_mean_and_submeta(self, df, sub_meta:str = 'None', rename_sample:bool = True, plot_mean:bool = False):
+ """
+ Prepares a DataFrame for baisc heatmap plotting.
+ Parameters:
+ df (pd.DataFrame): The input DataFrame containing the sample data.
+ sub_meta (str): The sub-metadata to be used for grouping. Default is 'None'.
+ rename_sample (bool): Whether to rename the samples. Default is True.
+ plot_mean (bool): Whether to plot the mean values. Default is False.
+ Returns:
+ tuple: A tuple containing the processed DataFrame and the group list for each column.
+ """
+ if plot_mean and sub_meta == 'None': #! if sub_meta is not None, plot_mean will be set to False
+ df = self.tfa.BasicStats.get_stats_mean_df_by_group(df)
+ sample_to_group_dict = {col: col for col in df.columns}
+ else:
+ df, sample_to_group_dict = self.tfa.BasicStats.get_combined_sub_meta_df(df=df, sub_meta=sub_meta, rename_sample=rename_sample, plot_mean=plot_mean)
+
+ return df, sample_to_group_dict
\ No newline at end of file
diff --git a/metax/taxafunc_analyzer/analyzer_utils/cross_test.py b/metax/taxafunc_analyzer/analyzer_utils/cross_test.py
index ecc9b9b..e042a0d 100644
--- a/metax/taxafunc_analyzer/analyzer_utils/cross_test.py
+++ b/metax/taxafunc_analyzer/analyzer_utils/cross_test.py
@@ -1,5 +1,6 @@
# T-Test , ANOVA, Tukey HSD, Deseq2
import pandas as pd
+import numpy as np
from statsmodels.stats.multicomp import pairwise_tukeyhsd
from scipy.stats import f_oneway
from scipy.stats import ttest_ind
@@ -726,43 +727,94 @@ def extrcat_significant_fc_from_deseq2all(self, df, p_value=0.05, log2fc_min=1,
return dft
# return a dict of 3 dataframe: df_all, df_no_na, df_same_trends
- def extrcat_significant_fc_from_all_3_levels(self, df, p_value=0.05, log2fc_min=1, log2fc_max=30, p_type='padj', df_type:str='deseq2') -> dict:
- def filter_rows(group):
- # 保留所有值都为正或者都为负的行
- return group[(group > 0).all(axis=1) | (group < 0).all(axis=1)]
-
- res_df_dict = {}
-
- first_level_values = df.columns.get_level_values(0).unique()
- res_dict = {}
- for value in first_level_values: # iterate over first level values
- sub_df = df[value]
- print(f"\nExtracting significant Stats from '{value}':")
- if df_type == 'dunnett':
- dft = self.extrcat_significant_stat_from_dunnett(sub_df, p_value=p_value, p_type=p_type)
- elif df_type == 'deseq2':
- dft = self.extrcat_significant_fc_from_deseq2all(sub_df, p_value=p_value, log2fc_min=log2fc_min, log2fc_max=log2fc_max, p_type=p_type)
- else:
- raise ValueError("df_type must be in ['dunnett', 'deseq2']")
-
- res_dict[value] = dft
- df = pd.concat(res_dict, axis=1)
- df_swapped = df.swaplevel(axis=1)
- df_swapped = df_swapped.sort_index(axis=1)
- print(f"\nTotal number of all_siginificant: [{df_swapped.shape[0]}]")
- res_df_dict['all_sig'] = df_swapped
-
- df_no_na = df_swapped.groupby(level=0, axis=1).apply(lambda x: x.dropna())
- df_no_na = df_no_na.droplevel(1, axis=1)
- print(f"Total number of no_na_in_one_group: [{df_no_na.shape[0]}]")
- res_df_dict['no_na'] = df_no_na
-
- # Only keep rows that have all values positive or all values negative
- df_same_trends = df_no_na.groupby(level=0, axis=1).apply(filter_rows)
- # dropna level 0 index
- df_same_trends.columns = df_same_trends.columns.droplevel(1)
- print(f"Total number of same_trends_in_one_group: [{df_same_trends.shape[0]}]")
- res_df_dict['same_trends'] = df_same_trends
+ def extrcat_significant_fc_from_all_3_levels(self, df, p_value=0.05, log2fc_min=1, log2fc_max=99,
+ p_type='padj', df_type:str='deseq2') -> dict:
+ """
+ Extracts significant fold change data from a multi-level DataFrame and categorizes it based on different filtering criteria.
+
+ Parameters:
+ -----------
+ df : pd.DataFrame
+ A multi-level DataFrame containing statistical data for different groups.
+
+ p_value : float, optional, default=0.05
+ The threshold for significance based on p-values. Only rows with p-values below this threshold will be considered significant.
+
+ log2fc_min : float, optional, default=1
+ The minimum log2 fold change to consider a row significant.
+
+ log2fc_max : float, optional, default=99
+ The maximum log2 fold change to consider a row significant.
+
+ p_type : str, optional, default='padj'
+ The type of p-value to use for filtering. Typically 'padj' or 'pvalue'.
+
+ df_type : str, optional, default='deseq2'
+ Specifies the type of statistical method used. Must be either 'dunnett' or 'deseq2'.
+
+ Returns:
+ --------
+ dict
+ A dictionary containing three DataFrames:
+ - 'all_sig': DataFrame containing all significant rows across all groups, Non-significant values are replaced with NA.
+ - 'half_same_trends': DataFrame containing rows where each group has the same trend (all positive or all negative non-NA values)
+ and at least 50% of the values are non-NA.
+ - 'no_na': DataFrame containing rows with no NA values in each group.
+ - 'same_trends': DataFrame containing rows with no NA values, and all values in each group follow the same trend (all positive or all negative).
+ """
+ def filter_rows_with_same_trends_and_half_na(group):
+ # 筛选出所有非NA值都为正或都为负的行
+ filtered = group[group.apply(lambda row: (row.dropna() > 0).all() or (row.dropna() < 0).all(), axis=1)]
+ # 检查每行非NA的值是否超过一半，若非NA值少于一半则将该行置为NA
+ filtered = filtered.apply(lambda x: x if x.notna().sum() > len(x) / 2 else pd.Series([np.nan] * len(x), index=x.index), axis=1)
+ return filtered
+ def filter_rows(group):
+ # 保留所有值都为正或者都为负的行, 且不包含NA
+ return group[(group > 0).all(axis=1) | (group < 0).all(axis=1)]
+
+ res_df_dict = {}
+
+ first_level_values = df.columns.get_level_values(0).unique()
+ res_dict = {}
+ for value in first_level_values: # iterate over first level values
+ sub_df = df[value]
+ print(f"\nExtracting significant Stats from '{value}':")
+ if df_type == 'dunnett':
+ dft = self.extrcat_significant_stat_from_dunnett(sub_df, p_value=p_value, p_type=p_type)
+ elif df_type == 'deseq2':
+ dft = self.extrcat_significant_fc_from_deseq2all(sub_df, p_value=p_value, log2fc_min=log2fc_min, log2fc_max=log2fc_max, p_type=p_type)
+ else:
+ raise ValueError("df_type must be in ['dunnett', 'deseq2']")
- return res_df_dict
\ No newline at end of file
+ res_dict[value] = dft
+ df = pd.concat(res_dict, axis=1)
+ df_swapped = df.swaplevel(axis=1)
+ df_swapped = df_swapped.sort_index(axis=1)
+ print(f"\nTotal number of all_siginificant: [{df_swapped.shape[0]}]")
+ res_df_dict['all_sig'] = df_swapped
+ #TODO extract half of the columns in each group has no na and same trends
+
+
+ # 按groupby(level=0)分组，过滤每组符合条件的行
+ df_half = pd.concat([filter_rows_with_same_trends_and_half_na(group) for _, group in df_swapped.groupby(level=0, axis=1)], axis=1)
+ # df_half.columns = df_half.columns.droplevel(1) # 删除多余的层级
+ df_half = df_half.dropna(how='all') # 删除所有值都为NA的行
+ print(f"Total number of half_same_trends: [{df_half.shape[0]}]")
+ res_df_dict['half_same_trends'] = df_half
+
+
+ df_no_na = df_swapped.groupby(level=0, axis=1).apply(lambda x: x.dropna())
+ df_no_na = df_no_na.droplevel(1, axis=1)
+ print(f"Total number of no_na: [{df_no_na.shape[0]}]")
+ res_df_dict['no_na'] = df_no_na
+
+ # Only keep rows that have all values positive or all values negative
+ df_same_trends = df_no_na.groupby(level=0, axis=1).apply(filter_rows)
+ # dropna level 0 index
+ df_same_trends.columns = df_same_trends.columns.droplevel(1)
+ print(f"Total number of same_trends: [{df_same_trends.shape[0]}]")
+ res_df_dict['same_trends'] = df_same_trends
+
+
+ return res_df_dict
\ No newline at end of file
diff --git a/metax/taxafunc_analyzer/analyzer_utils/data_preprocessing.py b/metax/taxafunc_analyzer/analyzer_utils/data_preprocessing.py
index 210e013..0efc2fc 100644
--- a/metax/taxafunc_analyzer/analyzer_utils/data_preprocessing.py
+++ b/metax/taxafunc_analyzer/analyzer_utils/data_preprocessing.py
@@ -118,7 +118,7 @@ def trace_shift(x):
normalize_operations = {
'None': lambda x: x,
'mean': lambda x: x - x.mean(),
- 'sum': lambda x: x / (x.sum() + epsilon),
+ 'percentage': lambda x: x / (x.sum() + epsilon) * 100,
'minmax': lambda x: (x - x.min()) / (x.max() - x.min()),
'zscore': lambda x: (x - x.mean()) / (x.std() + epsilon),
'pareto': lambda x: (x - x.mean()) / (np.sqrt(x.std() + epsilon)),
@@ -143,7 +143,7 @@ def trace_shift(x):
print(f'Data normalized by [{normalize_method}]')
else:
- raise ValueError('normalize_method must be in [None, mean, sum, minmax, zscore, pareto]')
+ raise ValueError(f'normalize_method must be in {list(normalize_operations.keys())}')
# move the data to positive
df_mat = df_mat - df_mat.min()
@@ -637,7 +637,7 @@ def data_preprocess(self, df: pd.DataFrame, normalize_method: str|None = None,
- `None`: No normalization.
- `trace_shift`: Trace shift normalization inspired by DirectLFQ.
- `mean`: Mean normalization.
- - `sum`: Sum normalization.
+ - `percentage`: Percentage normalization, then *100.
- `minmax`: Min-max normalization.
- `zscore`: Z-score normalization.
- `pareto`: Pareto scaling.
diff --git a/metax/taxafunc_ploter/heatmap_plot.py b/metax/taxafunc_ploter/heatmap_plot.py
index 0cdbd21..6a46d5f 100644
--- a/metax/taxafunc_ploter/heatmap_plot.py
+++ b/metax/taxafunc_ploter/heatmap_plot.py
@@ -349,18 +349,11 @@ def plot_basic_heatmap_of_test_res(self, df, top_number:int = 100, value_type:st
def plot_basic_heatmap(self, df, title = 'Heatmap',fig_size:tuple|None = None,
scale = None, col_cluster:bool = True, row_cluster:bool = True,
cmap:str|None = None, rename_taxa:bool = True, font_size:int = 10,
- show_all_labels:tuple = (False, False), rename_sample:bool = True, plot_mean:bool = False,
- sub_meta: str = "None", scale_method:str = 'maxmin', return_type:str = 'fig'
- ):
- '''
- sub_meta is higher plot_mean, if sub_meta provided, plot_mean is False
- '''
-
- if plot_mean and sub_meta == 'None': # if sub_meta is not None, plot_mean is False
- print('Plot the mean of the data, set rename_sample to False')
- rename_sample = False
-
- df = self.tfa.BasicStats.get_stats_mean_df_by_group(df)
+ show_all_labels:tuple = (False, False), scale_method:str = 'maxmin', return_type:str = 'fig',
+ sample_to_group_dict:dict|None = None):
+
+ # check if any row or column is all 0
+ if (df == 0).all().any():
# remove all 0 rows
row_num = len(df)
df = df.loc[~(df==0).all(axis=1)] if row_cluster else df
@@ -375,19 +368,12 @@ def plot_basic_heatmap(self, df, title = 'Heatmap',fig_size:tuple|None = None,
col_cluster = False
- mat = df.copy()
- mat = self.scale_data(df = mat, scale_by = scale, method = scale_method)
+ df = self.scale_data(df = df, scale_by = scale, method = scale_method)
# if index is Taxon, rename index
if rename_taxa:
- mat = self.rename_taxa(mat)
+ df = self.rename_taxa(df)
- mat, group_list = self.tfa.BasicStats.get_combined_sub_meta_df(df=mat, sub_meta=sub_meta, rename_sample=rename_sample, plot_mean=plot_mean)
-
- # if only one column, remove col_cluster, set scale to None
- if len(mat.columns) < 2:
- col_cluster = False
- # scale = None
if return_type == 'table':
sns_params = {
@@ -396,16 +382,16 @@ def plot_basic_heatmap(self, df, title = 'Heatmap',fig_size:tuple|None = None,
"method": self.linkage_method,
"metric": self.distance_metric,
}
- fig = sns.clustermap(mat, **sns_params)
+ fig = sns.clustermap(df, **sns_params)
# get the sorted dataframe
if row_cluster and not col_cluster:
- sorted_df = mat.iloc[fig.dendrogram_row.reordered_ind, :]
+ sorted_df = df.iloc[fig.dendrogram_row.reordered_ind, :]
elif col_cluster and not row_cluster:
- sorted_df = mat.iloc[:, fig.dendrogram_col.reordered_ind]
+ sorted_df = df.iloc[:, fig.dendrogram_col.reordered_ind]
elif row_cluster and col_cluster:
- sorted_df = mat.iloc[fig.dendrogram_row.reordered_ind, fig.dendrogram_col.reordered_ind]
+ sorted_df = df.iloc[fig.dendrogram_row.reordered_ind, fig.dendrogram_col.reordered_ind]
else:
- sorted_df = mat
+ sorted_df = df
plt.close(fig.figure)
return sorted_df
@@ -414,8 +400,12 @@ def plot_basic_heatmap(self, df, title = 'Heatmap',fig_size:tuple|None = None,
cmap = 'YlOrRd'
if fig_size is None:
fig_size = (30,30)
+ if sample_to_group_dict is not None:
+ group_list = [sample_to_group_dict.get(i, i) for i in df.columns]
+ color_list = self.assign_colors(group_list)
+ else:
+ color_list = None
- color_list = self.assign_colors(group_list)
sns_params = {
# "center": 0,
"cmap": cmap,
@@ -428,11 +418,11 @@ def plot_basic_heatmap(self, df, title = 'Heatmap',fig_size:tuple|None = None,
"row_cluster": row_cluster,
"method": self.linkage_method,
"metric": self.distance_metric,
- "col_colors": color_list if not plot_mean else None,
+ "col_colors": color_list,
"xticklabels": True if show_all_labels[0] else "auto",
"yticklabels": True if show_all_labels[1] else "auto",
}
- fig = sns.clustermap(mat, **sns_params)
+ fig = sns.clustermap(df, **sns_params)
fig.ax_heatmap.set_xticklabels(
fig.ax_heatmap.get_xmajorticklabels(),
@@ -490,7 +480,7 @@ def plot_heatmap_of_all_condition_res(self, df, pvalue:float = 0.05,scale:str|N
- return_type (str): The type of the return value. Default is 'fig'. options: 'fig', 'table'
- res_df_type (str): The type of the result DataFrame. Default is 'deseq2'.
- p_type (str): The type of pvalue. Default is 'padj'. options: 'pvalue', 'padj'
- - three_levels_df_type (str): The type of the three levels DataFrame. Default is 'same_trends'. options: 'all_sig', 'no_na', 'same_trends'
+ - three_levels_df_type (str): The type of the three levels DataFrame. Default is 'same_trends'. options: 'all_sig', 'no_na', 'half_same_trnds', 'same_trends'
- show_col_colors (bool): Whether to show column colors. Default is True.
- remove_zero_col (bool): Whether to remove zero columns. Default is True.
diff --git a/metax/taxafunc_ploter/network_plot.py b/metax/taxafunc_ploter/network_plot.py
index f857861..9c24810 100644
--- a/metax/taxafunc_ploter/network_plot.py
+++ b/metax/taxafunc_ploter/network_plot.py
@@ -113,6 +113,12 @@ def create_nodes_links(
- categories (list): Categories for nodes, used for coloring in the graph.
- cytoscape_df (DataFrame): DataFrame containing nodes and links for Cytoscape export.
"""
+ def update_focus(row):
+ focus_taxa = 'Y' if row['taxa'] in focus_list else 'N'
+ focus_func = 'Y' if row['function'] in focus_list else 'N'
+ focus = 'Y' if focus_taxa == 'Y' or focus_func == 'Y' else 'N'
+ return pd.Series([focus_taxa, focus_func, focus])
+
df = self.tfa.taxa_func_df.copy()
if self.rename_taxa:
print("Renaming taxa to last level")
@@ -134,8 +140,10 @@ def create_nodes_links(
network_df.columns = ['taxa', 'function'] + network_df.columns.tolist()[2:]
taxa_dict = network_df.drop('function', axis=1).groupby('taxa').sum().to_dict()
func_dict = network_df.drop('taxa', axis=1).groupby('function').sum().to_dict()
- network_df['focus_taxa'] = network_df['taxa'].apply(lambda x: 'Y' if x in focus_list else 'N')
- network_df['focus_func'] = network_df['function'].apply(lambda x: 'Y' if x in focus_list else 'N')
+
+ network_df[['focus_taxa', 'focus_func', 'focus']] = network_df.apply(update_focus, axis=1)
+
+
# cerate attributes_df
attributes_taxa_df = pd.DataFrame(network_df[['taxa']])
attributes_taxa_df.drop_duplicates(inplace=True)
@@ -157,7 +165,7 @@ def create_nodes_links(
# concatenate the taxa and function attributes_df
attributes_df = pd.concat([attributes_taxa_df, attributes_func_df])
- attributes_df['mean'] = attributes_df.drop(['node', 'focus', 'type'], axis=1).mean(axis=1)
+ attributes_df['mean_value'] = attributes_df.drop(['node', 'focus', 'type'], axis=1).mean(axis=1)
# Done creating network_df and attributes_df for export to cytoscape
df['mean'] = df.mean(axis=1)
diff --git a/metax/taxafunc_ploter/sankey_plot.py b/metax/taxafunc_ploter/sankey_plot.py
index 9913ca9..286db0f 100644
--- a/metax/taxafunc_ploter/sankey_plot.py
+++ b/metax/taxafunc_ploter/sankey_plot.py
@@ -44,6 +44,9 @@ def convert_df_by_group_for_sankey(self,df, sub_meta, plot_mean) -> dict:
df_dict = {}
# add all samples to the dict
df['sum'] = df.sum(axis=1)
+ df = df[df['sum'] != 0]
+ if len(df) == 0:
+ raise ValueError('\n\nAll values are 0!\n\nDo you calculate mean values for each group by Zero-Domainat methods?\n\nPlease check the settings.')
df_dict['All'] = self.df_to_sankey_df(df, value_col='sum')
# add samples for each group to the dict
for group, samples in group_dict.items():
@@ -55,9 +58,10 @@ def convert_df_by_group_for_sankey(self,df, sub_meta, plot_mean) -> dict:
df_temp['sum'] = df_temp.sum(axis=1)
# remove values that are 0
df_temp = df_temp[df_temp['sum'] != 0]
+ if len(df_temp) == 0:
+ continue
df_temp = self.df_to_sankey_df(df_temp, value_col='sum')
df_dict[group] = df_temp
-
return df_dict
diff --git a/metax/utils/version.py b/metax/utils/version.py
index f987022..d467f5e 100644
--- a/metax/utils/version.py
+++ b/metax/utils/version.py
@@ -1,2 +1,2 @@
-__version__ = '1.117.1'
+__version__ = '1.118.2'
API_version = '3'
\ No newline at end of file
diff --git a/pyproject.toml b/pyproject.toml
index b89d208..fd170fb 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
[project]
name = "MetaXTools"
-version = "1.117.1"
+version = "1.118.2"
description = "MetaXTools is a novel tool for linking peptide sequences with taxonomic and functional information in Metaproteomics."
readme = "README_PyPi.md"
license = { text = "NorthOmics" }