PFASGroups 3.2.2__tar.gz → 3.3.1__tar.gz

{pfasgroups-3.2.2 → pfasgroups-3.3.1}/PFASGroups/HalogenGroupModel.py

@@ -578,13 +578,12 @@ class HalogenGroup():
         if len(augmented_matched_components) == 0:
             return 0, [], 0, []
 
-        # Get all component sizes from all path types
-        all_components = list(set([comp for comps in augmented_matched_components for comp in comps]))
-        component_sizes = [len(x) for x in all_components]
+        # Get all component sizes from matched component dicts
+        component_sizes = [comp.get('size', 0) for comp in augmented_matched_components]
 
         self.all_matches = []  # Clear matches after use
         self.component_specific_extra_atoms = []
-        return max([0] + component_sizes), component_sizes, len(all_components), augmented_matched_components
+        return max([0] + component_sizes), component_sizes, len(augmented_matched_components), augmented_matched_components
 
     def find_aryl_components(self,mol, component_solver=None, **kwargs):
         """Find aryl components in a molecule with comprehensive metrics."""

{pfasgroups-3.2.2 → pfasgroups-3.3.1}/PFASGroups/PFASEmbeddings.py

@@ -311,7 +311,7 @@ def _get_group_info() -> Dict[int, Dict[str, str]]:
 
 # Groups always excluded from the non-OECD category label (perhalogenated /
 # polyhalogenated alkyl catch-alls that add no structural specificity).
-_CLASSIFY_EXCLUDED_IDS: frozenset = frozenset({51, 52})
+_CLASSIFY_EXCLUDED_IDS: frozenset = frozenset({34,35})  # "Perhalogenated alkyl" and "Polyhalogenated alkyl"
 
 # Name subsumption for non-OECD classification: when a more-specific group
 # name (key) is present, every name in its list is suppressed.

{pfasgroups-3.2.2 → pfasgroups-3.3.1}/PFASGroups/cli.py

@@ -47,7 +47,7 @@ Examples:
   # List available path types
   PFASGroups list-paths
 
-Note: Use get_componentSMARTSs() and get_PFASGroups() in Python to extend defaults.
+Note: Use get_componentSMARTSs() and get_HalogenGroups() in Python to extend defaults.
         """
     )
 
@@ -186,7 +186,7 @@ Note: Use get_componentSMARTSs() and get_PFASGroups() in Python to extend defaul
     # List groups command
     list_parser = subparsers.add_parser(
         'list-groups',
-        help='List available PFAS groups (use in Python to extend with get_PFASGroups)'
+        help='List available PFAS groups (use in Python to extend with get_HalogenGroups)'
     )
     list_parser.add_argument(
         '-o', '--output',

{pfasgroups-3.2.2 → pfasgroups-3.3.1}/PFASGroups/data/Halogen_groups_smarts.json

@@ -4258,8 +4258,8 @@
   },
   {
     "id": 65,
-    "name": "thia keto propanoic acid",
-    "alias": "Thia keto propanoic acid",
+    "name": "thioester keto dicarboxylic acid",
+    "alias": "Thioester keto dicarboxylic acid",
     "componentSmarts": null,
     "linker_smarts": null,
     "max_dist_from_comp": 0,
@@ -4331,7 +4331,7 @@
     "sorting": {
       "category": "generic",
       "elements": "SCO",
-      "group": "thia keto propanoic acids",
+      "group": "thioester keto dicarboxylic acids",
       "saturation": null,
       "aggregated": false
     }
@@ -7871,8 +7871,8 @@
   },
   {
     "id": 106,
-    "name": "telomer thia keto propanoic acid",
-    "alias": "telomer thia keto propanoic acid",
+    "name": "telomer thioester keto dicarboxylic acid",
+    "alias": "telomer thioester keto dicarboxylic acid",
     "componentSmarts": null,
     "componentSaturation": "per",
     "linker_smarts": "[CH2X4]",
@@ -7960,7 +7960,7 @@
     "sorting": {
       "category": "telomer",
       "elements": "SCO",
-      "group": "thia keto propanoic acids",
+      "group": "thioester keto dicarboxylic acids",
       "saturation": null,
       "aggregated": false
     }

{pfasgroups-3.2.2 → pfasgroups-3.3.1}/PFASGroups/getter.py

@@ -36,7 +36,7 @@ def get_compiled_HalogenGroups(**kwargs):
 
     Examples
     --------
-    >>> from PFASgroups import get_compiled_HalogenGroups, HalogenGroup, parse_smiles
+    >>> from PFASGroups import get_compiled_HalogenGroups, HalogenGroup, parse_smiles
     >>> groups = get_compiled_HalogenGroups()
     >>> groups.append(HalogenGroup(
     ...     id=200, name="Perfluoroalkyl nitrates",

{pfasgroups-3.2.2 → pfasgroups-3.3.1/PFASGroups.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: PFASGroups
-Version: 3.2.2
+Version: 3.3.1
 Summary: A comprehensive cheminformatics package for automated detection, classification, and analysis of halogenated substances, with a focus on per- and polyfluoroalkyl substances (PFAS). Combines SMARTS pattern matching, molecular formula constraints, and graph-based pathfinding to identify 119 groups. Supports creations of embeddings with graph metrics for Machine Learning workflows.
 Author: Luc T. Miaz
 Author-email: "Luc T. Miaz" <luc@miaz.ch>
@@ -38,6 +38,13 @@ Requires-Dist: pytest; extra == "dev"
 Requires-Dist: pytest-cov; extra == "dev"
 Provides-Extra: database
 Requires-Dist: sqlalchemy>=1.4.0; extra == "database"
+Provides-Extra: gui
+Requires-Dist: PySide6>=6.4; extra == "gui"
+Requires-Dist: plotly>=5.0; extra == "gui"
+Requires-Dist: umap-learn>=0.5; extra == "gui"
+Requires-Dist: openpyxl>=3.0; extra == "gui"
+Requires-Dist: sqlalchemy>=1.4.0; extra == "gui"
+Requires-Dist: matplotlib>=3.5; extra == "gui"
 
 # PFASGroups
 
@@ -71,7 +78,15 @@ PFASGroups combines SMARTS pattern matching, molecular formula constraints, and
 
 ## Installation
 
-### From Pypi
+The recommended installation method is via pip.  RDKit must already be available in the target environment. It is recommended to use an environment manager (like Conda/Mamba, e.g [Miniforge](https://github.com/conda-forge/miniforge)) and install RDKit via 
+
+```sh
+mamba install -y -c rdkit rdkit
+```
+
+### From PyPI
+
+PFASGroups is available on PyPI:
 
 ```sh
 pip install PFASGroups
@@ -106,6 +121,20 @@ print(results)        # prints PFASEmbeddingSet summary (molecule count, matched
 print(results[0])     # prints PFASEmbedding summary for the first molecule
 ```
 
+## Graphical User Interface (GUI)
+
+A GUI is available to run the main commands from the module. To launch the GUI, call the following command from the PFASGroups folder:
+
+```python
+python -m gui
+```
+
+Note that pyQt6 must be installed as well as pyCSRML to benefit from all features.
+
+## Binary release 
+
+To be done...
+
 ## Repository Structure
 
 ```
@@ -465,7 +494,7 @@ print(arr.shape)
 ### Option B – import `PFASgroups` and specify `halogens` explicitly
 
 ```python
-from PFASgroups import parse_smiles
+from PFASGroups import parse_smiles
 
 smiles_list = [
     "C(C(F)(F)F)F",

{pfasgroups-3.2.2 → pfasgroups-3.3.1}/PFASGroups.egg-info/SOURCES.txt

@@ -30,6 +30,27 @@ PFASGroups/data/PFAS_definitions_smarts.json
 PFASGroups/data/component_smarts.json
 PFASGroups/data/component_smarts_halogens.json
 PFASGroups/data/diatomic_bonds_dict.json
+gui/__init__.py
+gui/__main__.py
+gui/app.py
+gui/style.py
+gui/workers.py
+gui/data/help.html
+gui/tabs/__init__.py
+gui/tabs/chemspace_tab.py
+gui/tabs/classify_tab.py
+gui/tabs/definition_tester_tab.py
+gui/tabs/help_tab.py
+gui/tabs/modelling_tab.py
+gui/tabs/prioritise_tab.py
+gui/tabs/results_tab.py
+gui/utils/__init__.py
+gui/utils/chemspace.py
+gui/utils/export_dialog.py
+gui/utils/fingerprints.py
+gui/utils/io_readers.py
+gui/utils/modelling.py
+gui/utils/mol_renderer.py
 tests/test_bde_resistance.py
 tests/test_component_fractions.py
 tests/test_component_ratios.py

{pfasgroups-3.2.2 → pfasgroups-3.3.1}/PFASGroups.egg-info/entry_points.txt

@@ -1,3 +1,4 @@
 [console_scripts]
 halogengroups = PFASGroups.cli:main_halogen
 pfasgroups = PFASGroups.cli:main
+pfasgroups-gui = gui.__main__:main

{pfasgroups-3.2.2 → pfasgroups-3.3.1}/PFASGroups.egg-info/requires.txt

@@ -13,3 +13,11 @@ sqlalchemy>=1.4.0
 [dev]
 pytest
 pytest-cov
+
+[gui]
+PySide6>=6.4
+plotly>=5.0
+umap-learn>=0.5
+openpyxl>=3.0
+sqlalchemy>=1.4.0
+matplotlib>=3.5

{pfasgroups-3.2.2 → pfasgroups-3.3.1}/PFASGroups.egg-info/top_level.txt

@@ -1,2 +1,3 @@
 HalogenGroups
 PFASGroups
+gui

{pfasgroups-3.2.2/PFASGroups.egg-info → pfasgroups-3.3.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: PFASGroups
-Version: 3.2.2
+Version: 3.3.1
 Summary: A comprehensive cheminformatics package for automated detection, classification, and analysis of halogenated substances, with a focus on per- and polyfluoroalkyl substances (PFAS). Combines SMARTS pattern matching, molecular formula constraints, and graph-based pathfinding to identify 119 groups. Supports creations of embeddings with graph metrics for Machine Learning workflows.
 Author: Luc T. Miaz
 Author-email: "Luc T. Miaz" <luc@miaz.ch>
@@ -38,6 +38,13 @@ Requires-Dist: pytest; extra == "dev"
 Requires-Dist: pytest-cov; extra == "dev"
 Provides-Extra: database
 Requires-Dist: sqlalchemy>=1.4.0; extra == "database"
+Provides-Extra: gui
+Requires-Dist: PySide6>=6.4; extra == "gui"
+Requires-Dist: plotly>=5.0; extra == "gui"
+Requires-Dist: umap-learn>=0.5; extra == "gui"
+Requires-Dist: openpyxl>=3.0; extra == "gui"
+Requires-Dist: sqlalchemy>=1.4.0; extra == "gui"
+Requires-Dist: matplotlib>=3.5; extra == "gui"
 
 # PFASGroups
 
@@ -71,7 +78,15 @@ PFASGroups combines SMARTS pattern matching, molecular formula constraints, and
 
 ## Installation
 
-### From Pypi
+The recommended installation method is via pip.  RDKit must already be available in the target environment. It is recommended to use an environment manager (like Conda/Mamba, e.g [Miniforge](https://github.com/conda-forge/miniforge)) and install RDKit via 
+
+```sh
+mamba install -y -c rdkit rdkit
+```
+
+### From PyPI
+
+PFASGroups is available on PyPI:
 
 ```sh
 pip install PFASGroups
@@ -106,6 +121,20 @@ print(results)        # prints PFASEmbeddingSet summary (molecule count, matched
 print(results[0])     # prints PFASEmbedding summary for the first molecule
 ```
 
+## Graphical User Interface (GUI)
+
+A GUI is available to run the main commands from the module. To launch the GUI, call the following command from the PFASGroups folder:
+
+```python
+python -m gui
+```
+
+Note that pyQt6 must be installed as well as pyCSRML to benefit from all features.
+
+## Binary release 
+
+To be done...
+
 ## Repository Structure
 
 ```
@@ -465,7 +494,7 @@ print(arr.shape)
 ### Option B – import `PFASgroups` and specify `halogens` explicitly
 
 ```python
-from PFASgroups import parse_smiles
+from PFASGroups import parse_smiles
 
 smiles_list = [
     "C(C(F)(F)F)F",

{pfasgroups-3.2.2 → pfasgroups-3.3.1}/README.md

@@ -30,7 +30,15 @@ PFASGroups combines SMARTS pattern matching, molecular formula constraints, and
 
 ## Installation
 
-### From Pypi
+The recommended installation method is via pip.  RDKit must already be available in the target environment. It is recommended to use an environment manager (like Conda/Mamba, e.g [Miniforge](https://github.com/conda-forge/miniforge)) and install RDKit via 
+
+```sh
+mamba install -y -c rdkit rdkit
+```
+
+### From PyPI
+
+PFASGroups is available on PyPI:
 
 ```sh
 pip install PFASGroups
@@ -65,6 +73,20 @@ print(results)        # prints PFASEmbeddingSet summary (molecule count, matched
 print(results[0])     # prints PFASEmbedding summary for the first molecule
 ```
 
+## Graphical User Interface (GUI)
+
+A GUI is available to run the main commands from the module. To launch the GUI, call the following command from the PFASGroups folder:
+
+```python
+python -m gui
+```
+
+Note that pyQt6 must be installed as well as pyCSRML to benefit from all features.
+
+## Binary release 
+
+To be done...
+
 ## Repository Structure
 
 ```
@@ -424,7 +446,7 @@ print(arr.shape)
 ### Option B – import `PFASgroups` and specify `halogens` explicitly
 
 ```python
-from PFASgroups import parse_smiles
+from PFASGroups import parse_smiles
 
 smiles_list = [
     "C(C(F)(F)F)F",

pfasgroups-3.3.1/gui/__init__.py

@@ -0,0 +1 @@
+"""PFASGroups desktop GUI (PyQt6)."""

pfasgroups-3.3.1/gui/__main__.py

@@ -0,0 +1,21 @@
+"""Entry point — run with:  python -m gui  or  pfasgroups-gui"""
+import sys
+import os
+
+# Ensure the PFASGroups package root is on the path when run from the
+# repository root (e.g. python -m gui).
+_repo_root = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+if _repo_root not in sys.path:
+    sys.path.insert(0, _repo_root)
+
+from gui.app import create_app
+
+
+def main():
+    app, window = create_app(sys.argv)
+    window.show()
+    sys.exit(app.exec())
+
+
+if __name__ == "__main__":
+    main()

pfasgroups-3.3.1/gui/app.py

@@ -0,0 +1,162 @@
+"""
+PFASGroups GUI — Main window.
+"""
+from __future__ import annotations
+
+import sys
+from pathlib import Path
+
+from PySide6.QtCore import Qt
+from PySide6.QtGui import QIcon, QPixmap, QFont
+from PySide6.QtWidgets import (
+    QApplication, QMainWindow, QTabWidget, QWidget,
+    QHBoxLayout, QVBoxLayout, QLabel, QStatusBar,
+    QSizePolicy,
+)
+
+from gui import style
+
+_LOGO_PATH = Path(__file__).parent.parent / "logo" / "PFASGroups_logo.png"
+_TASKBAR_LOGO_PATH = Path(__file__).parent.parent / "logo" / "taskbar_logo.svg"
+
+
+class HeaderWidget(QWidget):
+    """Dark purple header bar with logo and app title."""
+
+    def __init__(self, parent=None):
+        super().__init__(parent)
+        self.setObjectName("header_widget")
+        self.setSizePolicy(QSizePolicy.Policy.Expanding, QSizePolicy.Policy.Fixed)
+        self.setFixedHeight(70)
+
+        layout = QHBoxLayout(self)
+        layout.setContentsMargins(16, 8, 16, 8)
+        layout.setSpacing(14)
+
+        # Logo
+        if _LOGO_PATH.exists():
+            pix = QPixmap(str(_LOGO_PATH)).scaledToHeight(
+                48, Qt.TransformationMode.SmoothTransformation
+            )
+            logo_lbl = QLabel()
+            logo_lbl.setPixmap(pix)
+            logo_lbl.setFixedSize(pix.size())
+            layout.addWidget(logo_lbl)
+        else:
+            placeholder = QLabel("PFASGroups")
+            placeholder.setStyleSheet(
+                f"color: {style.C_ORANGE}; font-size: 22px; font-weight: 800;"
+            )
+            layout.addWidget(placeholder)
+
+        # Title block
+        title_block = QVBoxLayout()
+        title_block.setSpacing(2)
+
+        title_lbl = QLabel("PFASGroups")
+        title_lbl.setObjectName("header_title")
+        title_block.addWidget(title_lbl)
+
+        sub_lbl = QLabel(
+            "Classification · Screening · Chemical Space · Modelling"
+        )
+        sub_lbl.setObjectName("header_subtitle")
+        title_block.addWidget(sub_lbl)
+
+        layout.addLayout(title_block)
+        layout.addStretch()
+
+        version_lbl = QLabel("v3.2")
+        version_lbl.setStyleSheet(
+            f"color: {style.C_PURPLE_LIGHT}; font-size: 11px;"
+        )
+        layout.addWidget(version_lbl)
+
+
+class MainWindow(QMainWindow):
+    """Application main window."""
+
+    def __init__(self):
+        super().__init__()
+        self.setWindowTitle("PFASGroups")
+        self.resize(1280, 900)
+        self.setMinimumSize(900, 650)
+
+        if _TASKBAR_LOGO_PATH.exists():
+            self.setWindowIcon(QIcon(str(_TASKBAR_LOGO_PATH)))
+        elif _LOGO_PATH.exists():
+            self.setWindowIcon(QIcon(str(_LOGO_PATH)))
+
+        # ── Central widget ─────────────────────────────────────────────────
+        central = QWidget()
+        self.setCentralWidget(central)
+        root_layout = QVBoxLayout(central)
+        root_layout.setContentsMargins(0, 0, 0, 0)
+        root_layout.setSpacing(0)
+
+        # Header
+        root_layout.addWidget(HeaderWidget())
+
+        # Tab widget
+        self._tabs = QTabWidget()
+        self._tabs.setDocumentMode(True)
+        root_layout.addWidget(self._tabs)
+
+        # ── Status bar ─────────────────────────────────────────────────────
+        self._status_bar = QStatusBar()
+        self.setStatusBar(self._status_bar)
+        self._status_bar.showMessage("Ready.")
+
+        # ── Build tabs ─────────────────────────────────────────────────────
+        self._build_tabs()
+
+    # ------------------------------------------------------------------
+    def _build_tabs(self):
+        from gui.tabs.classify_tab import ClassifyTab
+        from gui.tabs.results_tab import ResultsTab
+        from gui.tabs.definition_tester_tab import DefinitionTesterTab
+        from gui.tabs.prioritise_tab import PrioritiseTab
+        from gui.tabs.chemspace_tab import ChemSpaceTab
+        from gui.tabs.modelling_tab import ModellingTab
+        from gui.tabs.help_tab import HelpTab
+
+        self._classify_tab = ClassifyTab()
+        self._results_tab = ResultsTab()
+        self._def_tab = DefinitionTesterTab()
+        self._prior_tab = PrioritiseTab()
+        self._chem_tab = ChemSpaceTab()
+        self._model_tab = ModellingTab()
+        self._help_tab = HelpTab()
+
+        # Wire classification → results
+        self._classify_tab.classification_done.connect(self._on_classification_done)
+
+        self._tabs.addTab(self._classify_tab,  "1 · Classification")
+        self._tabs.addTab(self._results_tab,   "2 · Results")
+        self._tabs.addTab(self._def_tab,       "3 · Definition Tester")
+        self._tabs.addTab(self._prior_tab,     "4 · Prioritisation")
+        self._tabs.addTab(self._chem_tab,      "5 · Chemical Space")
+        self._tabs.addTab(self._model_tab,     "6 · Modelling")
+        self._tabs.addTab(self._help_tab,      "7 · Help")
+
+    # ------------------------------------------------------------------
+    def _on_classification_done(self, embedding_set):
+        """Propagate classification results to all dependent tabs."""
+        self._results_tab.set_results(embedding_set)
+        self._prior_tab.set_results(embedding_set)
+        self._chem_tab.set_results(embedding_set)
+        self._model_tab.set_results(embedding_set)
+        self._tabs.setCurrentIndex(1)  # jump to Results tab
+
+    # ------------------------------------------------------------------
+    def show_status(self, msg: str):
+        self._status_bar.showMessage(msg)
+
+
+def create_app(argv=None) -> tuple[QApplication, MainWindow]:
+    app = QApplication(argv or sys.argv)
+    app.setApplicationName("PFASGroups")
+    app.setOrganizationName("Stockholm University")
+    style.apply(app)
+    window = MainWindow()
+    return app, window