posebench-fast 0.1.0__tar.gz

posebench_fast-0.1.0/PKG-INFO

@@ -0,0 +1,109 @@
+Metadata-Version: 2.4
+Name: posebench-fast
+Version: 0.1.0
+Summary: Fast docking evaluation metrics: symmetry-corrected RMSD and PoseBusters filters
+Keywords: docking,rmsd,posebusters,molecular-docking,benchmark
+Author: Nikolenko
+License-Expression: MIT
+Requires-Dist: numpy>=1.21
+Requires-Dist: pandas>=1.3
+Requires-Dist: spyrmsd>=0.6
+Requires-Dist: rdkit>=2022.3
+Requires-Dist: torch>=1.10
+Requires-Dist: posebusters>=0.2
+Requires-Dist: tqdm>=4.60
+Requires-Dist: pytest>=7.0 ; extra == 'dev'
+Requires-Dist: pytest-cov>=4.0 ; extra == 'dev'
+Requires-Dist: ruff>=0.4 ; extra == 'dev'
+Requires-Python: >=3.11
+Provides-Extra: dev
+Description-Content-Type: text/markdown
+
+# posebench-fast
+
+Fast docking evaluation metrics for molecular docking benchmarks.
+
+## Features
+
+- **Symmetry-corrected RMSD**: Compute RMSD accounting for molecular symmetry using graph isomorphisms
+- **Fast PoseBusters filters**: Quick physical validity checks without full energy evaluation
+  - Distance checks (not too far, no clashes)
+  - Volume overlap detection
+  - Internal geometry validation
+- **Metrics aggregation**: Success rates, averages, filtering by scores
+
+## Installation
+
+```bash
+uv add posebench-fast
+```
+
+Or with pip:
+```bash
+pip install posebench-fast
+```
+
+## Usage
+
+### Symmetry RMSD
+
+```python
+from posebench_fast import compute_all_isomorphisms, get_symmetry_rmsd_with_isomorphisms
+import numpy as np
+
+# Compute isomorphisms once per molecule
+isomorphisms = compute_all_isomorphisms(rdkit_mol)
+
+# Compute symmetry-corrected RMSD
+rmsd = get_symmetry_rmsd_with_isomorphisms(true_coords, pred_coords, isomorphisms)
+```
+
+### Fast PoseBusters Filters
+
+```python
+from posebench_fast import calc_posebusters
+
+results = calc_posebusters(
+    pos_pred=ligand_positions,      # (n_samples, n_atoms, 3)
+    pos_cond=protein_positions,     # (n_atoms, 3)
+    atom_ids_pred=ligand_atom_ids,
+    atom_names_cond=protein_atom_names,
+    names="sample_id",
+    lig_mol_for_posebusters=rdkit_mol
+)
+
+# Results contain:
+# - not_too_far_away: bool list
+# - no_clashes: bool list
+# - no_volume_clash: bool list
+# - no_internal_clash: bool list
+# - is_buried_fraction: float list
+```
+
+### Metrics Aggregation
+
+```python
+from posebench_fast import get_final_results_for_df
+
+rows, scored_results = get_final_results_for_df(
+    full_results,
+    score_names=['error_estimate_0'],
+    posebusters_filter=True,
+    fast_filter=True
+)
+# Returns DataFrame with: RMSD < 2A, RMSD < 5A, SymRMSD < 2A, etc.
+```
+
+## Dependencies
+
+- numpy
+- pandas
+- spyrmsd
+- rdkit
+- torch
+- posebusters
+- tqdm
+
+## License
+
+MIT

posebench_fast-0.1.0/README.md

@@ -0,0 +1,88 @@
+# posebench-fast
+
+Fast docking evaluation metrics for molecular docking benchmarks.
+
+## Features
+
+- **Symmetry-corrected RMSD**: Compute RMSD accounting for molecular symmetry using graph isomorphisms
+- **Fast PoseBusters filters**: Quick physical validity checks without full energy evaluation
+  - Distance checks (not too far, no clashes)
+  - Volume overlap detection
+  - Internal geometry validation
+- **Metrics aggregation**: Success rates, averages, filtering by scores
+
+## Installation
+
+```bash
+uv add posebench-fast
+```
+
+Or with pip:
+```bash
+pip install posebench-fast
+```
+
+## Usage
+
+### Symmetry RMSD
+
+```python
+from posebench_fast import compute_all_isomorphisms, get_symmetry_rmsd_with_isomorphisms
+import numpy as np
+
+# Compute isomorphisms once per molecule
+isomorphisms = compute_all_isomorphisms(rdkit_mol)
+
+# Compute symmetry-corrected RMSD
+rmsd = get_symmetry_rmsd_with_isomorphisms(true_coords, pred_coords, isomorphisms)
+```
+
+### Fast PoseBusters Filters
+
+```python
+from posebench_fast import calc_posebusters
+
+results = calc_posebusters(
+    pos_pred=ligand_positions,      # (n_samples, n_atoms, 3)
+    pos_cond=protein_positions,     # (n_atoms, 3)
+    atom_ids_pred=ligand_atom_ids,
+    atom_names_cond=protein_atom_names,
+    names="sample_id",
+    lig_mol_for_posebusters=rdkit_mol
+)
+
+# Results contain:
+# - not_too_far_away: bool list
+# - no_clashes: bool list
+# - no_volume_clash: bool list
+# - no_internal_clash: bool list
+# - is_buried_fraction: float list
+```
+
+### Metrics Aggregation
+
+```python
+from posebench_fast import get_final_results_for_df
+
+rows, scored_results = get_final_results_for_df(
+    full_results,
+    score_names=['error_estimate_0'],
+    posebusters_filter=True,
+    fast_filter=True
+)
+# Returns DataFrame with: RMSD < 2A, RMSD < 5A, SymRMSD < 2A, etc.
+```
+
+## Dependencies
+
+- numpy
+- pandas
+- spyrmsd
+- rdkit
+- torch
+- posebusters
+- tqdm
+
+## License
+
+MIT

posebench_fast-0.1.0/pyproject.toml

@@ -0,0 +1,94 @@
+[project]
+name = "posebench-fast"
+version = "0.1.0"
+description = "Fast docking evaluation metrics: symmetry-corrected RMSD and PoseBusters filters"
+readme = "README.md"
+requires-python = ">=3.11"
+license = "MIT"
+authors = [
+    { name = "Nikolenko" }
+]
+keywords = ["docking", "rmsd", "posebusters", "molecular-docking", "benchmark"]
+
+dependencies = [
+    "numpy>=1.21",
+    "pandas>=1.3",
+    "spyrmsd>=0.6",
+    "rdkit>=2022.03",
+    "torch>=1.10",
+    "posebusters>=0.2",
+    "tqdm>=4.60",
+]
+
+[project.optional-dependencies]
+dev = [
+    "pytest>=7.0",
+    "pytest-cov>=4.0",
+    "ruff>=0.4",
+]
+
+[build-system]
+requires = ["uv_build"]
+build-backend = "uv_build"
+
+[tool.uv.build-backend]
+module-name = "posebench_fast"
+
+[tool.uv]
+dev-dependencies = [
+    "pytest>=7.0",
+    "pytest-cov>=4.0",
+    "ruff>=0.4",
+]
+
+[tool.ruff]
+src = ["src/"]
+target-version = "py311"
+
+[tool.ruff.lint]
+select = [
+    "E",    # pycodestyle errors
+    "F",    # Pyflakes (undefined names, unused imports, etc.)
+    "W",    # pycodestyle warnings
+    "I",    # isort (import sorting)
+    "N",    # pep8-naming
+    "UP",   # pyupgrade (modernize Python code)
+    "B",    # flake8-bugbear (common bugs)
+    "C4",   # flake8-comprehensions
+    "PIE",  # flake8-pie (misc lints)
+    "RET",  # flake8-return (return statement improvements)
+    "SIM",  # flake8-simplify
+    "NPY",  # NumPy-specific rules
+]
+
+ignore = [
+    "E501",   # Line too long (handled by formatter)
+    "E741",   # Ambiguous variable name (allow 'l', 'O', etc. for math)
+    "N803",   # Argument name should be lowercase (allow CamelCase for classes)
+    "N806",   # Variable should be lowercase (allow uppercase in math/ML code)
+    "B008",   # Do not perform function call in argument defaults (common in ML)
+    "RET504", # Unnecessary variable assignment before return
+    "SIM108", # Use ternary operator (can reduce readability)
+    "SIM102", # Avoid assigning lambda expressions (OK for ML configs)
+    "N812",   # Lowercase imported as non-lowercase (allow for 'import torch.nn.functional as F')
+    "N802",   # Function name should be lowercase (allow for class-factory style)
+    "E731",   # Do not assign a lambda expression (allow in configs/utilities)
+    "C408",   # Unnecessary 'dict' call (allow for explicitness)
+    "SIM113", # Use 'any'/'all' for simple comparisons (allow explicit loops)
+]
+
+fixable = ["ALL"]
+unfixable = []
+
+[tool.ruff.lint.per-file-ignores]
+"__init__.py" = ["F401"]
+"tests/**/*.py" = ["S101"]
+
+[tool.ruff.format]
+quote-style = "double"
+indent-style = "space"
+skip-magic-trailing-comma = false
+line-ending = "auto"
+
+[tool.ruff.lint.pydocstyle]
+convention = "numpy"

posebench_fast-0.1.0/src/posebench_fast/__init__.py

@@ -0,0 +1,53 @@
+"""
+posebench-fast: Fast docking evaluation metrics
+
+Provides:
+- Symmetry-corrected RMSD computation
+- Fast PoseBusters filters (without full energy evaluation)
+- Docking metrics (success rates, averages)
+"""
+
+from posebench_fast.filters.fast_filters import (
+    calc_posebusters,
+    check_geometry,
+    check_intermolecular_distance,
+    check_volume_overlap,
+)
+from posebench_fast.metrics.aggregation import (
+    filter_results_by_fast,
+    filter_results_by_posebusters,
+    get_best_results_by_score,
+    get_final_results_for_df,
+    get_simple_metrics_df,
+)
+from posebench_fast.metrics.rmsd import (
+    TimeoutException,
+    compute_all_isomorphisms,
+    get_symmetry_rmsd,
+    get_symmetry_rmsd_with_isomorphisms,
+    symmrmsd,
+    time_limit,
+)
+
+__version__ = "0.1.0"
+
+__all__ = [
+    # RMSD
+    "compute_all_isomorphisms",
+    "get_symmetry_rmsd_with_isomorphisms",
+    "get_symmetry_rmsd",
+    "symmrmsd",
+    "TimeoutException",
+    "time_limit",
+    # Filters
+    "calc_posebusters",
+    "check_intermolecular_distance",
+    "check_volume_overlap",
+    "check_geometry",
+    # Metrics
+    "get_simple_metrics_df",
+    "get_final_results_for_df",
+    "filter_results_by_posebusters",
+    "filter_results_by_fast",
+    "get_best_results_by_score",
+]

posebench_fast-0.1.0/src/posebench_fast/datasets/__init__.py

@@ -0,0 +1 @@
+"""Dataset utilities for posebench-fast."""

posebench_fast-0.1.0/src/posebench_fast/filters/__init__.py

@@ -0,0 +1,15 @@
+"""Fast PoseBusters filters for docking evaluation."""
+
+from posebench_fast.filters.fast_filters import (
+    calc_posebusters,
+    check_geometry,
+    check_intermolecular_distance,
+    check_volume_overlap,
+)
+
+__all__ = [
+    "calc_posebusters",
+    "check_intermolecular_distance",
+    "check_volume_overlap",
+    "check_geometry",
+]