annotation validator

Author: matthiaskoenig

pyproject.toml

@@ -34,7 +34,7 @@ keywords = [
     "computational modeling",
 ]
 dependencies = [
-    "pymetadata>=0.5.6",
+    "pymetadata>=0.5.7",
 	"depinfo",
 	"rich",
 	"lxml",

release-notes/0.9.5.md

@@ -0,0 +1,10 @@
+# Release notes for sbmlutils 0.9.5
+![sbmlutils](https://github.com/matthiaskoenig/sbmlutils/raw/develop/docs_builder/images/sbmlutils-logo-60.png)
+
+We are pleased to release the next version of sbmlutils including the
+following changes:
+
+# Features
+- annotation validation functionality
+
+Your sbmlutils team

src/sbmlutils/metadata/validator.py

@@ -0,0 +1,65 @@
+from pathlib import Path
+from typing import Union
+
+import libsbml
+import pandas as pd
+from pymetadata.identifiers.miriam import BQB
+
+from sbmlutils.console import console
+from sbmlutils.io.sbml import read_sbml
+from sbmlutils.log import get_logger
+from pymetadata.core.annotation import RDFAnnotation
+
+
+logger = get_logger(__name__)
+
+
+def validate_sbml_annotations(source: Union[Path, str]) -> pd.DataFrame:
+    """Validate annotations in a given SBML file.
+
+    :param source: SBML to check
+    :return: DataFrame of invalid annotations
+    """
+    doc: libsbml.SBMLDocument = read_sbml(source=source)
+    console.rule(style="white")
+    console.print(f"Validate annotations: {source}", style="white bold")
+    console.rule(style="white")
+
+    elements = doc.getListOfAllElements()
+    element: libsbml.SBase
+    invalid_annotations: list = []
+    for element in elements:
+        if element.isSetAnnotation():
+            cvterm: libsbml.CVTerm
+            cvterms = element.getCVTerms()
+
+            # console.rule(f"id='{element.id}' | {type(element)} | '{element.name}'", align="left", style="bold white")
+            for cvterm in cvterms:
+                qualifier_type = cvterm.getQualifierType()
+                for k in range(cvterm.getNumResources()):
+
+                    resource_uri = cvterm.getResourceURI(k)
+                    # console.print(f"{qualifier_type} | {resource_uri}")
+                    annotation = RDFAnnotation(qualifier=BQB.IS, resource=resource_uri, validate=False)
+                    valid: bool = annotation.validate()
+                    if not valid:
+                        console.print(
+                            f"id='{element.id}' | {type(element).__name__} | '{element.name}' | {resource_uri}",
+                            style="warning"
+                        )
+                        invalid_annotations.append(
+                            {
+                                "id": element.id,
+                                "object": type(element).__name__,
+                                "resource": resource_uri,
+                            }
+                        )
+    df = pd.DataFrame(invalid_annotations)
+    if len(invalid_annotations) == 0:
+        console.print("All annotations valid", style="success")
+    else:
+        console.print(df.to_string())
+        console.print("Invalid annotations", style="error")
+    console.rule(style="white")
+    return df
+

src/sbmlutils/metadata/validator_example.py

@@ -0,0 +1,9 @@
+import pandas as pd
+
+from sbmlutils.console import console
+from sbmlutils.resources import MODELS_DIR
+from sbmlutils.metadata.validator import validate_sbml_annotations
+
+if __name__ == "__main__":
+    sbml_faure2006 = MODELS_DIR / "qual" / "Faure2006_MammalianCellCycle.sbml"
+    results: pd.DataFrame = validate_sbml_annotations(sbml_faure2006)