pxmeter 0.0.1__tar.gz → 0.1.1__tar.gz

pxmeter-0.1.1/PKG-INFO

@@ -0,0 +1,32 @@
+Metadata-Version: 2.4
+Name: pxmeter
+Version: 0.1.1
+Summary: PXMeter is a comprehensive toolkit for evaluating the quality of         structures generated by biomolecular structure prediction models.
+Author: Bytedance Inc.
+Author-email: ai4s-bio@bytedance.com
+License: Apache 2.0 License
+Platform: manylinux1
+Requires-Python: >=3.11
+License-File: LICENSE
+Requires-Dist: biotite>=1.2.0
+Requires-Dist: dockq==2.1.3
+Requires-Dist: gemmi==0.7.0
+Requires-Dist: joblib
+Requires-Dist: ml_collections
+Requires-Dist: numpy
+Requires-Dist: pandas
+Requires-Dist: pdbeccdutils==0.8.5
+Requires-Dist: posebusters==0.3.1
+Requires-Dist: rdkit>=2024.09.6
+Requires-Dist: scipy
+Requires-Dist: tabulate
+Requires-Dist: tqdm
+Requires-Dist: click
+Dynamic: author
+Dynamic: author-email
+Dynamic: license
+Dynamic: license-file
+Dynamic: platform
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary

{pxmeter-0.0.1 → pxmeter-0.1.1}/README.md

@@ -4,8 +4,8 @@
 [![Python 3.11+](https://img.shields.io/badge/python-3.11%2B-blue.svg)](https://www.python.org/downloads/)
 
 <div align="left" style="margin: 20px 0;">
-<span style="margin: 0 10px;">📄 <a href="URL">From Dataset Diagnostics to Evaluation: Revisiting
-Structure Prediction Benchmarks with PXMeter</a></span>
+<span style="margin: 0 10px;">📄 <a href="https://www.biorxiv.org/content/10.1101/2025.07.17.664878v1">From Dataset Curation to Unified Evaluation: Revisiting
+ Structure Prediction Benchmarks with PXMeter</a></span>
 </div>
 
 PXMeter is a comprehensive toolkit for evaluating the quality of structures generated by biomolecular structure prediction models, with support for proteins, nucleic acids, and small molecules.
@@ -86,7 +86,12 @@ Refer to [benchmark/README.md](./benchmark/README.md) for evaluation protocols o
 - PoseBusters V2
 
 The benchmark data is released under the CC0 license.
-We include code in the `benchmark` directory that evaluates various models using PXMeter and aggregates their metrics. This serves as an example of best practices for using the tool. For more details, please refer to the [Technical Report URL].
+We include code in the `benchmark` directory that evaluates various models using PXMeter and aggregates their metrics.
+This serves as an example of best practices for using the tool. For more details, please refer to our paper:
+
+📄 <a href="https://www.biorxiv.org/content/10.1101/2025.07.17.664878v1">From Dataset Curation to Unified Evaluation: Revisiting
+ Structure Prediction Benchmarks with PXMeter</a>
+
 
 
 ## 💪 Contributing to PXMeter
@@ -102,21 +107,27 @@ pre-commit install
 ```
 
 
-## 🚧 Limitations
-- PXMeter supports chain/atom permutations but not residue-level permutations. As a result, the accuracy of evaluation for branched chains, such as glycans, cannot be fully guaranteed.
-- It is recommended to use CIF files from the RCSB PDB as references, as they ensure content accuracy. All development and testing were conducted exclusively on CIF files from this source.
-
-
 ## ✍️ Citing PXMeter
+If you use PXMeter in your research, please cite the following:
 ```bibtex
-@software{pxmeter,
-  title = {PXMeter: Biomolecular Structure Quality Assessment Toolkit},
-  author = {Bytedance AI for Science Team},
-  year = {2024},
-  url = {https://github.com/your-org/pxmeter},
+@article {Ma2025.07.17.664878,
+	author = {Ma, Wenzhi and Liu, Zhenyu and Yang, Jincai and Lu, Chan and Zhang, Hanyu and Xiao, Wenzhi},
+	title = {From Dataset Curation to Unified Evaluation: Revisiting Structure Prediction Benchmarks with PXMeter},
+	year = {2025},
+	doi = {10.1101/2025.07.17.664878},
+	publisher = {Cold Spring Harbor Laboratory},
+	URL = {https://www.biorxiv.org/content/early/2025/07/22/2025.07.17.664878},
+	eprint = {https://www.biorxiv.org/content/early/2025/07/22/2025.07.17.664878.full.pdf},
+	journal = {bioRxiv}
 }
 ```
 
+
+## 🚧 Limitations
+- PXMeter supports chain/atom permutations but not residue-level permutations. As a result, the accuracy of evaluation for branched chains, such as glycans, cannot be fully guaranteed.
+- It is recommended to use CIF files from the RCSB PDB as references, as they ensure content accuracy. All development and testing were conducted exclusively on CIF files from this source.
+
+
 ## 🛡️ Security
 If you discover a potential security issue in this project, or think you may
 have discovered a security issue, we ask that you notify Bytedance Security via our [security center](https://security.bytedance.com/src) or [vulnerability reporting email](sec@bytedance.com).

{pxmeter-0.0.1 → pxmeter-0.1.1}/pxmeter/cli.py

@@ -12,13 +12,19 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import sys
 from pathlib import Path
 
 import click
 
-from pxmeter.configs.data_config import COMPONENTS_FILE, download_ccd_cif
+from pxmeter.configs.data_config import (
+    COMPONENTS_FILE,
+    download_ccd_cif,
+    make_one_letter_code_json_from_ccd,
+    ONE_LETTER_CODE_JSON,
+)
 from pxmeter.eval import evaluate, MetricResult
-from pxmeter.utils import none_or_str, read_chain_id_to_mol_from_json
+from pxmeter.utils import read_chain_id_to_mol_from_json
 
 
 def run_eval_cif(
@@ -29,14 +35,14 @@ def run_eval_cif(
     ref_assembly_id: str | None = None,
     ref_altloc: str = "first",
     interested_lig_label_asym_id: str | None = None,
-    chain_id_to_mol_json: str | None = None,
+    chain_id_to_mol_json: Path | None = None,
     output_mapped_cif: bool = False,
 ) -> MetricResult:
     """
-    Evaluate the performance of a model CIF file against a reference CIF file.
-    And save the result to a JSON file.
+    Evaluate the performance of a model CIF file by comparing it to a reference CIF file,
+    and save the results in a JSON file.
     """
-    if chain_id_to_mol_json:
+    if chain_id_to_mol_json is not None:
         chain_id_to_mol = read_chain_id_to_mol_from_json(chain_id_to_mol_json)
     else:
         chain_id_to_mol = None
@@ -68,37 +74,9 @@ def run_eval_cif(
     return metric_result
 
 
-@click.group()
-def ccd_cli():
-    """
-    CCD Options.
-    """
-    return
-
-
-@ccd_cli.command(name="update")
-def update():
-    """
-    Update the CCD database.
-    """
-    download_ccd_cif(output_path=COMPONENTS_FILE.parent)
-
-
-@click.group()
-def eval_cif():
-    """
-    Evaluate the performance of a model CIF file against a reference CIF file.
-    """
-    return
-
-
 @click.group(invoke_without_command=True)
-@click.option(
-    "-r", "--ref_cif", type=Path, required=True, help="Path to the reference CIF file."
-)
-@click.option(
-    "-m", "--model_cif", type=Path, required=True, help="Path to the model CIF file."
-)
+@click.option("-r", "--ref_cif", type=Path, help="Path to the reference CIF file.")
+@click.option("-m", "--model_cif", type=Path, help="Path to the model CIF file.")
 @click.option(
     "-o",
     "--output_json",
@@ -114,7 +92,7 @@ def eval_cif():
 )
 @click.option(
     "--ref_assembly_id",
-    type=none_or_str,
+    type=str,
     default=None,
     help="Assembly ID in the reference CIF file. Defaults to None.",
 )
@@ -127,15 +105,15 @@ def eval_cif():
 @click.option(
     "-l",
     "--interested_lig_label_asym_id",
-    type=none_or_str,
+    type=str,
     default=None,
     help="The label_asym_id of the ligand of interest in the reference structure (for ligand RMSD metrics). \
-            If multiple ligands are present, separate them by comma. Defaults to None.",
+        If multiple ligands are present, separate them by comma. Defaults to None.",
 )
 @click.option(
     "-c",
     "--chain_id_to_mol_json",
-    type=none_or_str,
+    type=Path,
     default=None,
     help="Path to a JSON file containing a mapping of chain IDs to molecular input (SMILES). \
         E.g. {'B': 'c1ccccc1', 'D':'CCCC'}",
@@ -145,7 +123,9 @@ def eval_cif():
     is_flag=True,
     help="Whether to output the mapped CIF file. Defaults to False.",
 )
-def run_eval_cif_cli(
+@click.pass_context
+def cli(
+    ctx,
     ref_cif: Path,
     model_cif: Path,
     output_json: Path,
@@ -153,21 +133,47 @@ def run_eval_cif_cli(
     ref_assembly_id: str | None = None,
     ref_altloc: str = "first",
     interested_lig_label_asym_id: str | None = None,
-    chain_id_to_mol_json: str | None = None,
+    chain_id_to_mol_json: Path | None = None,
     output_mapped_cif: bool = False,
 ):
     """
-    Evaluate the performance of a model CIF file against a reference CIF file.
-    And save the result to a JSON file.
+    Evaluate the performance of a model CIF file by comparing it to a reference CIF file,
+    and save the results in a JSON file.
     """
-    run_eval_cif(
-        ref_cif,
-        model_cif,
-        output_json,
-        ref_model,
-        ref_assembly_id,
-        ref_altloc,
-        interested_lig_label_asym_id,
-        chain_id_to_mol_json,
-        output_mapped_cif,
-    )
+    if ctx.invoked_subcommand is None:
+        if len(sys.argv) == 1:
+            click.echo(ctx.get_help())
+            ctx.exit()
+
+        if ref_cif is None or model_cif is None:
+            click.echo("Error: --ref_cif and --model_cif are required.")
+            ctx.exit()
+
+        run_eval_cif(
+            ref_cif,
+            model_cif,
+            output_json,
+            ref_model,
+            ref_assembly_id,
+            ref_altloc,
+            interested_lig_label_asym_id,
+            chain_id_to_mol_json,
+            output_mapped_cif,
+        )
+
+
+@cli.group(name="ccd")
+def ccd_cli():
+    """
+    CCD Options.
+    """
+    return
+
+
+@ccd_cli.command(name="update")
+def update():
+    """
+    Update the CCD database.
+    """
+    download_ccd_cif(output_path=COMPONENTS_FILE.parent)
+    make_one_letter_code_json_from_ccd(COMPONENTS_FILE, ONE_LETTER_CODE_JSON)

{pxmeter-0.0.1 → pxmeter-0.1.1}/pxmeter/configs/data_config.py

@@ -52,7 +52,7 @@ def download_ccd_cif(output_path: Path):
 
     sp.run(f"gunzip -d {output_cif_gz}", shell=True, check=True)
 
-    logging.info("Download CCD CIF file successfully.")
+    logging.info("Download CCD CIF file successfully: %s", output_cif)
 
 
 def make_one_letter_code_json_from_ccd(components_file: Path, output_json: Path):

{pxmeter-0.0.1 → pxmeter-0.1.1}/pxmeter/data/ccd.py

@@ -232,7 +232,7 @@ def get_ccd_perm_info(ccd_code: str) -> dict[str, Any]:
         for node in removed_nodes:
             mol_graph.remove_node(node)
 
-        matches = get_mol_graph_matches(mol_graph, mol_graph)[:1000]
+        matches = get_mol_graph_matches(mol_graph, mol_graph, max_matches=1000)
 
         # re-index after removing H
         reverted_old_atom_map = {v: k for k, v in mol.atom_map.items()}

{pxmeter-0.0.1 → pxmeter-0.1.1}/pxmeter/data/utils.py

@@ -234,13 +234,16 @@ def rdkit_mol_to_nx_graph(mol: Chem.Mol) -> nx.Graph:
     return G
 
 
-def get_mol_graph_matches(mol_graph1: nx.Graph, mol_graph2: nx.Graph) -> list[dict]:
+def get_mol_graph_matches(
+    mol_graph1: nx.Graph, mol_graph2: nx.Graph, max_matches: int = 1000
+) -> list[dict]:
     """
     Find all isomorphisms between subgraph of mol_graph1 and mol_graph2.
 
     Args:
         mol_graph1 (nx.Graph): Source molecular graph (typically larger subgraph)
         mol_graph2 (nx.Graph): Target molecular graph (typically smaller supergraph)
+        max_matches (int): Maximum number of matches to return. Default is 1000.
 
     Returns:
         list[dict]: List of node mapping dictionaries where each dictionary represents
@@ -254,4 +257,12 @@ def get_mol_graph_matches(mol_graph1: nx.Graph, mol_graph2: nx.Graph) -> list[di
         mol_graph2,
         node_match=lambda x, y: x["atomic_num"] == y["atomic_num"],
     )
-    return list(isomatcher.subgraph_isomorphisms_iter())
+
+    matches = []
+    num = 0
+    for i in isomatcher.subgraph_isomorphisms_iter():
+        matches.append(i)
+        num += 1
+        if num >= max_matches:
+            break
+    return matches

{pxmeter-0.0.1 → pxmeter-0.1.1}/pxmeter/mapping.py

@@ -584,7 +584,7 @@ class MappingCIF:
             ref_graph = rdkit_mol_to_nx_graph(ref_mol)
             model_graph = rdkit_mol_to_nx_graph(model_mol)
             try:
-                matches = get_mol_graph_matches(model_graph, ref_graph)
+                matches = get_mol_graph_matches(model_graph, ref_graph, max_matches=1)
                 assert len(matches) > 0
                 match = matches[0]
 

{pxmeter-0.0.1 → pxmeter-0.1.1}/pxmeter/utils.py

@@ -13,30 +13,18 @@
 # limitations under the License.
 
 import json
+from pathlib import Path
 
 from rdkit import Chem
 
 
-def none_or_str(value: str | None) -> str | None:
-    """
-    Converts the string "None" to a NoneType, otherwise returns the input string.
-    Args:
-        value (str, optional): The input string which may be "None".
-    Returns:
-        str or None: None if the input is the string "None", otherwise the input string.
-    """
-    if value == "None":
-        return None
-    return value
-
-
-def read_chain_id_to_mol_from_json(json_f: str) -> dict[str, Chem.Mol]:
+def read_chain_id_to_mol_from_json(json_f: Path | str) -> dict[str, Chem.Mol]:
     """
     Reads a JSON file containing chain IDs and their corresponding SMILES representations,
     and returns a dictionary mapping chain IDs to RDKit Mol objects.
 
     Args:
-        json_f (str): The path to the JSON file.
+        json_f (Path | str): The path to the JSON file.
 
     Returns:
         dict[str, Chem.Mol]: A dictionary mapping chain IDs to RDKit Mol objects.

pxmeter-0.1.1/pxmeter.egg-info/PKG-INFO

@@ -0,0 +1,32 @@
+Metadata-Version: 2.4
+Name: pxmeter
+Version: 0.1.1
+Summary: PXMeter is a comprehensive toolkit for evaluating the quality of         structures generated by biomolecular structure prediction models.
+Author: Bytedance Inc.
+Author-email: ai4s-bio@bytedance.com
+License: Apache 2.0 License
+Platform: manylinux1
+Requires-Python: >=3.11
+License-File: LICENSE
+Requires-Dist: biotite>=1.2.0
+Requires-Dist: dockq==2.1.3
+Requires-Dist: gemmi==0.7.0
+Requires-Dist: joblib
+Requires-Dist: ml_collections
+Requires-Dist: numpy
+Requires-Dist: pandas
+Requires-Dist: pdbeccdutils==0.8.5
+Requires-Dist: posebusters==0.3.1
+Requires-Dist: rdkit>=2024.09.6
+Requires-Dist: scipy
+Requires-Dist: tabulate
+Requires-Dist: tqdm
+Requires-Dist: click
+Dynamic: author
+Dynamic: author-email
+Dynamic: license
+Dynamic: license-file
+Dynamic: platform
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary

pxmeter-0.1.1/pxmeter.egg-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+pxm = pxmeter.cli:cli

{pxmeter-0.0.1 → pxmeter-0.1.1}/setup.py

@@ -20,8 +20,9 @@ with open("requirements.txt") as f:
 setup(
     name="pxmeter",
     python_requires=">=3.11",
-    version="0.0.1",
-    description="PXMeter is a tool crafted to assess the structural quality derived from Biomolecular Folding Models",
+    version="0.1.1",
+    description="PXMeter is a comprehensive toolkit for evaluating the quality of \
+        structures generated by biomolecular structure prediction models.",
     author="Bytedance Inc.",
     author_email="ai4s-bio@bytedance.com",
     packages=find_packages(
@@ -41,7 +42,7 @@ setup(
     platforms="manylinux1",
     entry_points={
         "console_scripts": [
-            "pxm = pxmeter.cli:run_eval_cif_cli",
+            "pxm = pxmeter.cli:cli",
         ],
     },
 )

pxmeter-0.0.1/PKG-INFO

@@ -1,10 +0,0 @@
-Metadata-Version: 2.1
-Name: pxmeter
-Version: 0.0.1
-Summary: PXMeter is a tool crafted to assess the structural quality derived from Biomolecular Folding Models
-Author: Bytedance Inc.
-Author-email: ai4s-bio@bytedance.com
-License: Apache 2.0 License
-Platform: manylinux1
-Requires-Python: >=3.11
-License-File: LICENSE

pxmeter-0.0.1/pxmeter.egg-info/PKG-INFO

@@ -1,10 +0,0 @@
-Metadata-Version: 2.1
-Name: pxmeter
-Version: 0.0.1
-Summary: PXMeter is a tool crafted to assess the structural quality derived from Biomolecular Folding Models
-Author: Bytedance Inc.
-Author-email: ai4s-bio@bytedance.com
-License: Apache 2.0 License
-Platform: manylinux1
-Requires-Python: >=3.11
-License-File: LICENSE

pxmeter-0.0.1/pxmeter.egg-info/entry_points.txt

@@ -1,2 +0,0 @@
-[console_scripts]
-pxm = pxmeter.cli:run_eval_cif_cli