Extend validator to check if all required parameters are present in p…

…arameter table (Closes #43)

petab/core.py

@@ -11,7 +11,7 @@
 from . import lint
 from . import sbml
 from . import parameter_mapping
-from typing import Optional, List, Union, Iterable
+from typing import Optional, List, Union, Iterable, Set
 import warnings
 
 
@@ -668,6 +668,43 @@ def create_parameter_df(sbml_model: libsbml.Model,
         lower_bound: lower bound for parameter value
         upper_bound: upper bound for parameter value
     """
+    parameter_ids = list(get_required_parameters_for_parameter_table(
+        sbml_model, condition_df, measurement_df))
+
+    df = pd.DataFrame(
+        data={
+            'parameterId': parameter_ids,
+            'parameterName': parameter_ids,
+            'parameterScale': parameter_scale,
+            'lowerBound': lower_bound,
+            'upperBound': upper_bound,
+            'nominalValue': np.nan,
+            'estimate': 1,
+            'priorType': '',
+            'priorParameters': ''
+        })
+    df.set_index(['parameterId'], inplace=True)
+
+    # For SBML model parameters set nominal values as defined in the model
+    for parameter_id in df.index:
+        try:
+            parameter = sbml_model.getParameter(parameter_id)
+            if parameter:
+                df.loc[parameter_id, 'nominalValue'] = parameter.getValue()
+        except ValueError:
+            # parameter was introduced as condition-specific override and
+            # is potentially not present in the model
+            pass
+    return df
+
+
+def get_required_parameters_for_parameter_table(
+        sbml_model: libsbml.Model,
+        condition_df: pd.DataFrame,
+        measurement_df: pd.DataFrame) -> Set[str]:
+    """
+    Get set of parameters which need to go into the parameter table
+    """
 
     observables = get_observables(sbml_model)
     sigmas = get_sigmas(sbml_model)
@@ -727,33 +764,7 @@ def append_overrides(overrides):
         for overrider in overrides:
             parameter_ids[overrider] = None
 
-    parameter_ids = list(parameter_ids.keys())
-
-    df = pd.DataFrame(
-        data={
-            'parameterId': parameter_ids,
-            'parameterName': parameter_ids,
-            'parameterScale': parameter_scale,
-            'lowerBound': lower_bound,
-            'upperBound': upper_bound,
-            'nominalValue': np.nan,
-            'estimate': 1,
-            'priorType': '',
-            'priorParameters': ''
-        })
-    df.set_index(['parameterId'], inplace=True)
-
-    # For SBML model parameters set nominal values as defined in the model
-    for parameter_id in df.index:
-        try:
-            parameter = sbml_model.getParameter(parameter_id)
-            if parameter:
-                df.loc[parameter_id, 'nominalValue'] = parameter.getValue()
-        except ValueError:
-            # parameter was introduced as condition-specific override and
-            # is potentially not present in the model
-            pass
-    return df
+    return parameter_ids.keys()
 
 
 def get_priors_from_df(parameter_df: pd.DataFrame):

petab/lint.py

@@ -9,6 +9,7 @@
 import logging
 import libsbml
 import pandas as pd
+from typing import Optional
 
 logger = logging.getLogger(__name__)
 
@@ -68,7 +69,11 @@ def check_measurement_df(df):
     _check_df(df, req_cols, "measurement")
 
 
-def check_parameter_df(df):
+def check_parameter_df(
+        df: pd.DataFrame,
+        sbml_model: Optional[libsbml.Model],
+        measurement_df: Optional[pd.DataFrame],
+        condition_df: Optional[pd.DataFrame]):
     req_cols = [
         "parameterName", "parameterScale",
         "lowerBound", "upperBound", "nominalValue", "estimate"
@@ -94,6 +99,37 @@ def check_parameter_df(df):
     assert_parameter_id_is_unique(df)
     check_parameter_bounds(df)
 
+    if sbml_model and measurement_df is not None \
+            and condition_df is not None:
+        assert_all_parameters_present_in_parameter_df(
+            df, sbml_model, measurement_df, condition_df)
+
+
+def assert_all_parameters_present_in_parameter_df(
+        parameter_df: pd.DataFrame,
+        sbml_model: libsbml.Model,
+        measurement_df: pd.DataFrame,
+        condition_df: pd.DataFrame):
+    """Ensure all required parameters are contained in the parameter table
+    with no additional ones"""
+
+    expected = core.get_required_parameters_for_parameter_table(
+        sbml_model=sbml_model, condition_df=condition_df,
+        measurement_df=measurement_df)
+
+    actual = set(parameter_df.index)
+
+    missing = expected - actual
+    extraneous = actual - expected
+
+    if missing:
+        raise AssertionError('Missing parameter(s) in parameter table: '
+                             + str(missing))
+
+    if extraneous:
+        raise AssertionError('Extraneous parameter(s) in parameter table: '
+                             + str(extraneous))
+
 
 def assert_measured_observables_present_in_model(measurement_df, sbml_model):
     """Check if all observables in measurement files have been specified in
@@ -400,7 +436,8 @@ def lint_problem(problem: 'core.Problem'):
     if problem.parameter_df is not None:
         logger.info("Checking parameter table...")
         try:
-            check_parameter_df(problem.parameter_df)
+            check_parameter_df(problem.parameter_df, problem.sbml_model,
+                               problem.measurement_df, problem.condition_df)
         except AssertionError as e:
             logger.error(e)
             errors_occurred = True