lint

Author: dweindl

petab/v2/core.py

@@ -441,7 +441,8 @@ class ExperimentPeriod(BaseModel):
     """
 
     #: The start time of the period in time units as defined in the model.
-    start: float = Field(alias=C.TIME)
+    # TODO: Only finite times and -inf are allowed as start time
+    time: float = Field(alias=C.TIME)
     # TODO: decide if optional
     #: The ID of the condition to be applied at the start time.
     condition_id: str = Field(alias=C.CONDITION_ID)
@@ -519,7 +520,7 @@ def from_df(cls, df: pd.DataFrame) -> ExperimentsTable:
         for experiment_id, cur_exp_df in df.groupby(C.EXPERIMENT_ID):
             periods = [
                 ExperimentPeriod(
-                    start=row[C.TIME], condition_id=row[C.CONDITION_ID]
+                    time=row[C.TIME], condition_id=row[C.CONDITION_ID]
                 )
                 for _, row in cur_exp_df.iterrows()
             ]
@@ -567,6 +568,13 @@ def __iadd__(self, other: Experiment) -> ExperimentsTable:
         self.experiments.append(other)
         return self
 
+    def __getitem__(self, item):
+        """Get an experiment by ID."""
+        for experiment in self.experiments:
+            if experiment.id == item:
+                return experiment
+        raise KeyError(f"Experiment ID {item} not found")
+
 
 class Measurement(BaseModel):
     """A measurement.
@@ -770,6 +778,13 @@ def __iadd__(self, other: Mapping) -> MappingTable:
         self.mappings.append(other)
         return self
 
+    def __getitem__(self, petab_id: str) -> Mapping:
+        """Get a mapping by PEtab ID."""
+        for mapping in self.mappings:
+            if mapping.petab_id == petab_id:
+                return mapping
+        raise KeyError(f"PEtab ID {petab_id} not found")
+
 
 class Parameter(BaseModel):
     """Parameter definition."""
@@ -862,3 +877,10 @@ def __iadd__(self, other: Parameter) -> ParameterTable:
             raise TypeError("Can only add Parameter to ParameterTable")
         self.parameters.append(other)
         return self
+
+    def __getitem__(self, item) -> Parameter:
+        """Get a parameter by ID."""
+        for parameter in self.parameters:
+            if parameter.id == item:
+                return parameter
+        raise KeyError(f"Parameter ID {item} not found")

petab/v2/lint.py

@@ -4,7 +4,7 @@
 
 import logging
 from abc import ABC, abstractmethod
-from collections import OrderedDict
+from collections import Counter, OrderedDict
 from collections.abc import Set
 from dataclasses import dataclass, field
 from enum import IntEnum
@@ -496,62 +496,44 @@ class CheckExperimentTable(ValidationTask):
     """A task to validate the experiment table of a PEtab problem."""
 
     def run(self, problem: Problem) -> ValidationIssue | None:
-        if problem.experiment_df is None:
-            return
-
-        df = problem.experiment_df
-
-        try:
-            _check_df(df, EXPERIMENT_DF_REQUIRED_COLS, "experiment")
-        except AssertionError as e:
-            return ValidationError(str(e))
+        messages = []
+        for experiment in problem.experiments_table.experiments:
+            # Check that there are no duplicate timepoints
+            counter = Counter(period.time for period in experiment.periods)
+            duplicates = {time for time, count in counter.items() if count > 1}
+            if duplicates:
+                messages.append(
+                    f"Experiment {experiment.id} contains duplicate "
+                    f"timepoints: {duplicates}"
+                )
 
-        # valid timepoints
-        invalid = []
-        for time in df[TIME].values:
-            try:
-                time = float(time)
-                if not np.isfinite(time) and time != -np.inf:
-                    invalid.append(time)
-            except ValueError:
-                invalid.append(time)
-        if invalid:
-            return ValidationError(
-                f"Invalid timepoints in experiment table: {invalid}"
-            )
+        if messages:
+            return ValidationError("\n".join(messages))
 
 
 class CheckExperimentConditionsExist(ValidationTask):
     """A task to validate that all conditions in the experiment table exist
     in the condition table."""
 
     def run(self, problem: Problem) -> ValidationIssue | None:
-        if problem.experiment_df is None:
-            return
-
-        if (
-            problem.condition_df is None
-            and problem.experiment_df is not None
-            and not problem.experiment_df.empty
-        ):
-            return ValidationError(
-                "Experiment table is non-empty, "
-                "but condition table is missing."
-            )
-
-        required_conditions = problem.experiment_df[CONDITION_ID].unique()
-        existing_conditions = problem.condition_df[CONDITION_ID].unique()
+        messages = []
+        available_conditions = {
+            c.id
+            for c in problem.conditions_table.conditions
+            if not pd.isna(c.id)
+        }
+        for experiment in problem.experiments_table.experiments:
+            missing_conditions = {
+                period.condition for period in experiment.periods
+            } - available_conditions
+            if missing_conditions:
+                messages.append(
+                    f"Experiment {experiment.id} requires conditions that are "
+                    f"not present in the condition table: {missing_conditions}"
+                )
 
-        missing_conditions = set(required_conditions) - set(
-            existing_conditions
-        )
-        # TODO NA allowed?
-        missing_conditions = {x for x in missing_conditions if not pd.isna(x)}
-        if missing_conditions:
-            return ValidationError(
-                f"Experiment table contains conditions that are not present "
-                f"in the condition table: {missing_conditions}"
-            )
+        if messages:
+            return ValidationError("\n".join(messages))
 
 
 class CheckAllParametersPresentInParameterTable(ValidationTask):

petab/v2/problem.py

@@ -153,6 +153,32 @@ def __str__(self):
             f"{observables}, {measurements}, {parameters}"
         )
 
+    def __getitem__(self, key):
+        """Get PEtab entity by ID.
+
+        This allows accessing PEtab entities such as conditions, experiments,
+        observables, and parameters by their ID.
+
+        Accessing model entities is not currently not supported.
+        """
+        for table in (
+            self.conditions_table,
+            self.experiments_table,
+            self.observables_table,
+            self.measurement_table,
+            self.parameter_table,
+            self.mapping_table,
+        ):
+            if table is not None:
+                try:
+                    return table[key]
+                except KeyError:
+                    pass
+
+        raise KeyError(
+            f"Entity with ID '{key}' not found in the PEtab problem"
+        )
+
     @staticmethod
     def from_yaml(
         yaml_config: dict | Path | str, base_path: str | Path = None
@@ -1062,20 +1088,13 @@ def add_experiment(self, id_: str, *args):
                 "Arguments must be pairs of timepoints and condition IDs."
             )
 
-        records = []
-        for i in range(0, len(args), 2):
-            records.append(
-                {
-                    EXPERIMENT_ID: id_,
-                    TIME: args[i],
-                    CONDITION_ID: args[i + 1],
-                }
-            )
-        tmp_df = pd.DataFrame(records)
-        self.experiment_df = (
-            pd.concat([self.experiment_df, tmp_df])
-            if self.experiment_df is not None
-            else tmp_df
+        periods = [
+            core.ExperimentPeriod(time=args[i], condition_id=args[i + 1])
+            for i in range(0, len(args), 2)
+        ]
+
+        self.experiments_table.experiments.append(
+            core.Experiment(id=id_, periods=periods)
         )
 
     def __iadd__(self, other):

tests/v2/test_core.py

@@ -42,9 +42,9 @@ def test_experiment_add_periods():
     exp = Experiment(id="exp1")
     assert exp.periods == []
 
-    p1 = ExperimentPeriod(start=0, condition_id="p1")
-    p2 = ExperimentPeriod(start=1, condition_id="p2")
-    p3 = ExperimentPeriod(start=2, condition_id="p3")
+    p1 = ExperimentPeriod(time=0, condition_id="p1")
+    p2 = ExperimentPeriod(time=1, condition_id="p2")
+    p3 = ExperimentPeriod(time=2, condition_id="p3")
     exp += p1
     exp += p2
 

tests/v2/test_lint.py

@@ -10,23 +10,14 @@
 def test_check_experiments():
     """Test ``CheckExperimentTable``."""
     problem = Problem()
-    problem.add_experiment("e1", 0, "c1", 1, "c2")
-    problem.add_experiment("e2", "-inf", "c1", 1, "c2")
-    assert problem.experiment_df.shape == (4, 3)
 
     check = CheckExperimentTable()
     assert check.run(problem) is None
 
-    assert check.run(Problem()) is None
-
-    tmp_problem = deepcopy(problem)
-    tmp_problem.experiment_df.loc[0, TIME] = "invalid"
-    assert check.run(tmp_problem) is not None
-
-    tmp_problem = deepcopy(problem)
-    tmp_problem.experiment_df.loc[0, TIME] = "inf"
-    assert check.run(tmp_problem) is not None
+    problem.add_experiment("e1", 0, "c1", 1, "c2")
+    problem.add_experiment("e2", "-inf", "c1", 1, "c2")
+    assert check.run(problem) is None
 
     tmp_problem = deepcopy(problem)
-    tmp_problem.experiment_df.drop(columns=[TIME], inplace=True)
+    tmp_problem["e1"].periods[0].time = tmp_problem["e1"].periods[1].time
     assert check.run(tmp_problem) is not None