Develop strategy for handling errors in the input data

docs/user-guide/input-data.rst

@@ -77,8 +77,8 @@ Here we walk the file system, load all phenopacket JSON files, and transform the
   ...       pp_path = os.path.join(dirpath, filename)
   ...       with open(pp_path) as fh:
   ...         pp = Parse(fh.read(), Phenopacket())
-  ...       patient = patient_creator.create_patient(pp)
-  ...       patients.append(patient)
+  ...       output = patient_creator.process(pp)
+  ...       patients.append(output.outcome)
 
 
   >>> f'Loaded {len(patients)} phenopackets'

src/genophenocorr/data/_toy.py

@@ -77,132 +77,132 @@ def get_toy_cohort() -> Cohort:
         Genotypes.empty())  # Not used in the patients below, hence `empty()`.
 
     patients = (
-        Patient('A',
+        Patient(SampleLabels('A'),
                 phenotypes=(arachnodactyly_T, spasticity_F, seizure_T),
                 variants=[snv],
                 proteins=[prot]
                 ),
-        Patient('B',
+        Patient(SampleLabels('B'),
                 phenotypes=(arachnodactyly_T, seizure_T, spasticity_T),
                 variants=[snv],
                 proteins=[prot]
                 ),
-        Patient('C',
+        Patient(SampleLabels('C'),
                 phenotypes=(arachnodactyly_F, spasticity_T, seizure_T),
                 variants=[snv],
                 proteins=[prot]
                 ),
-        Patient('D',
+        Patient(SampleLabels('D'),
                 phenotypes=(arachnodactyly_T, spasticity_T, seizure_T),
                 variants=[snv, deletion],
                 proteins=[prot]
                 ),
-        Patient('E',
+        Patient(SampleLabels('E'),
                 phenotypes=(arachnodactyly_T, spasticity_T, seizure_F),
                 variants=[snv],
                 proteins=[prot]
                 ),
-        Patient('F',
+        Patient(SampleLabels('F'),
                 phenotypes=(arachnodactyly_F, spasticity_F, seizure_T),
                 variants=[deletion],
                 proteins=[prot]
                 ),
-        Patient('G',
+        Patient(SampleLabels('G'),
                 phenotypes=(arachnodactyly_T, seizure_T, spasticity_T),
                 variants=[snv, deletion],
                 proteins=[prot]
                 ),
-        Patient('H',
+        Patient(SampleLabels('H'),
                 phenotypes=(arachnodactyly_T, seizure_T, spasticity_F),
                 variants=[deletion],
                 proteins=[prot]
                 ),
-        Patient('I',
+        Patient(SampleLabels('I'),
                 phenotypes=(arachnodactyly_F, spasticity_F, seizure_T),
                 variants=[deletion],
                 proteins=[prot]
                 ),
-        Patient('J',
+        Patient(SampleLabels('J'),
                 phenotypes=(arachnodactyly_T, seizure_T, spasticity_T),
                 variants=[snv],
                 proteins=[prot]
                 ),
-        Patient('K',
+        Patient(SampleLabels('K'),
                 phenotypes=(arachnodactyly_F, spasticity_T, seizure_T),
                 variants=[snv],
                 proteins=[prot]
                 ),
-        Patient('L',
+        Patient(SampleLabels('L'),
                 phenotypes=(arachnodactyly_F, seizure_F, spasticity_F),
                 variants=[deletion],
                 proteins=[prot]
                 ),
-        Patient('M',
+        Patient(SampleLabels('M'),
                 phenotypes=(arachnodactyly_T, seizure_F, spasticity_T),
                 variants=[snv],
                 proteins=[prot]
                 ),
-        Patient('N',
+        Patient(SampleLabels('N'),
                 phenotypes=(arachnodactyly_F, seizure_T, spasticity_F),
                 variants=[snv],
                 proteins=[prot]
                 ),
-        Patient('O',
+        Patient(SampleLabels('O'),
                 phenotypes=(arachnodactyly_F, seizure_F, spasticity_T),
                 variants=[deletion],
                 proteins=[prot]
                 ),
-        Patient('P',
+        Patient(SampleLabels('P'),
                 phenotypes=(arachnodactyly_T, seizure_T, spasticity_F),
                 variants=[snv],
                 proteins=[prot]
                 ),
-        Patient('Q',
+        Patient(SampleLabels('Q'),
                 phenotypes=(arachnodactyly_T, seizure_F, spasticity_F),
                 variants=[snv],
                 proteins=[prot]
                 ),
-        Patient('R',
+        Patient(SampleLabels('R'),
                 phenotypes=(arachnodactyly_T, seizure_T, spasticity_F),
                 variants=[snv, deletion],
                 proteins=[prot]
                 ),
-        Patient('S',
+        Patient(SampleLabels('S'),
                 phenotypes=(arachnodactyly_F, seizure_T, spasticity_T),
                 variants=[deletion],
                 proteins=[prot]
                 ),
-        Patient('T',
+        Patient(SampleLabels('T'),
                 phenotypes=(arachnodactyly_T, seizure_F, spasticity_T),
                 variants=[snv],
                 proteins=[prot]
                 ),
-        Patient('U',
+        Patient(SampleLabels('U'),
                 phenotypes=(arachnodactyly_F, seizure_T, spasticity_T),
                 variants=[deletion],
                 proteins=[prot]
                 ),
-        Patient('V',
+        Patient(SampleLabels('V'),
                 phenotypes=(arachnodactyly_T, seizure_T, spasticity_T),
                 variants=[snv],
                 proteins=[prot]
                 ),
-        Patient('W',
+        Patient(SampleLabels('W'),
                 phenotypes=(arachnodactyly_F, seizure_T, spasticity_T),
                 variants=[deletion],
                 proteins=[prot]
                 ),
-        Patient('X',
+        Patient(SampleLabels('X'),
                 phenotypes=(arachnodactyly_F, seizure_T, spasticity_T),
                 variants=[deletion],
                 proteins=[prot]
                 ),
-        Patient('Y',
+        Patient(SampleLabels('Y'),
                 phenotypes=(arachnodactyly_T, seizure_T, spasticity_T),
                 variants=[snv],
                 proteins=[prot]
                 ),
-        Patient('Z',
+        Patient(SampleLabels('Z'),
                 phenotypes=(arachnodactyly_F, seizure_T, spasticity_T),
                 variants=[deletion],
                 proteins=[prot]

src/genophenocorr/model/__init__.py

@@ -4,7 +4,7 @@
 and protein info.
 """
 from . import genome
-
+from ._base import SampleLabels
 from ._cohort import Cohort, Patient
 from ._gt import Genotype, Genotypes, Genotyped
 from ._phenotype import Phenotype
@@ -14,7 +14,7 @@
 from ._variant_effects import VariantEffect
 
 __all__ = [
-    'Cohort', 'Patient',
+    'Cohort', 'Patient', 'SampleLabels',
     'Phenotype',
     'Variant', 'VariantCoordinates', 'Genotype', 'Genotypes', 'Genotyped',
     'TranscriptAnnotation', 'VariantEffect', 'TranscriptInfoAware', 'TranscriptCoordinates',

src/genophenocorr/model/_base.py

@@ -0,0 +1,60 @@
+import typing
+
+import hpotk
+
+
+class SampleLabels:
+    """
+    A data model for subject identifiers.
+
+    The subject has a mandatory :attr:`label` and an optional :attr:`meta_label`.
+
+    The identifiers support natural ordering, equality tests, and are hashable.
+    """
+
+    def __init__(self, label: str,
+                 meta_label: typing.Optional[str] = None):
+        self._label = hpotk.util.validate_instance(label, str, 'label')
+        self._meta_label = hpotk.util.validate_optional_instance(meta_label, str, 'meta_label')
+
+    @property
+    def label(self) -> str:
+        return self._label
+
+    @property
+    def meta_label(self) -> typing.Optional[str]:
+        return self._meta_label
+
+    def label_summary(self) -> str:
+        """
+        Summarize `label` and `meta_label` into a `str` where the sub-parts are inserted as ``<label>[<meta_label>]``.
+        """
+        return self._label if self._meta_label is None else f'{self._label}[{self._meta_label}]'
+
+    def __eq__(self, other):
+        return isinstance(other, SampleLabels) and self._label == other.label and self._meta_label == other._meta_label
+
+    def __lt__(self, other):
+        if isinstance(other, SampleLabels):
+            if self._label < other._label:
+                return True
+            elif self._label == other._label:
+                if self._meta_label is None or other._meta_label is None:
+                    if self._meta_label == other._meta_label:
+                        return False
+                    else:
+                        return True if self._meta_label is None else False  # `None` is less
+                else:
+                    return self._meta_label < other._meta_label
+            return False
+        else:
+            return NotImplemented
+
+    def __hash__(self):
+        return hash((self._label, self._meta_label))
+
+    def __str__(self):
+        return self.label_summary()
+
+    def __repr__(self):
+        return f'SampleLabels(label={self._label}, meta_label={self._meta_label})'

src/genophenocorr/model/_cohort.py

@@ -1,6 +1,7 @@
 import typing
 from collections import Counter
 
+from ._base import SampleLabels
 from ._phenotype import Phenotype
 from ._protein import ProteinMetadata
 from ._variant import Variant
@@ -10,25 +11,25 @@ class Patient:
     """A class that represents an individual patient
 
     Attributes:
-        patient_id (string): A string unique to this Patient object
+        patient_id (SampleLabels): The patient identifiers
         phenotypes (Sequence[Phenotype]): A list of Phenotype objects
         variants (Sequence[Variant]): A list of Variant objects
         proteins (Sequence[ProteinMetadata]): A list of ProteinMetadata objects
     """
 
-    def __init__(self, patient_id: str,
+    def __init__(self, labels: SampleLabels,
                  phenotypes: typing.Iterable[Phenotype],
                  variants: typing.Iterable[Variant],
                  proteins: typing.Iterable[ProteinMetadata]):
         """Constructs all necessary attributes for a Patient object
 
         Args:
-            patient_id (string): A string unique to this Patient object
+            labels (string): A string unique to this Patient object
             phenotypes (Iterable[Phenotype]): A list of Phenotype objects
             variants (Iterable[Variant]): A list of Variant objects
             proteins (Iterable[ProteinMetadata]): A list of ProteinMetadata objects
         """
-        self._id = patient_id
+        self._labels = labels
         self._phenotypes = tuple(phenotypes)
         self._variants = tuple(variants)
         self._proteins = tuple(proteins)
@@ -39,7 +40,14 @@ def patient_id(self) -> str:
         Returns:
             string: Patient ID unique to this Patient object
         """
-        return self._id
+        return self._labels.label_summary()
+
+    @property
+    def labels(self) -> SampleLabels:
+        """
+        Get the sample identifiers.
+        """
+        return self._labels
 
     @property
     def phenotypes(self) -> typing.Sequence[Phenotype]:
@@ -79,7 +87,7 @@ def excluded_phenotypes(self) -> typing.Iterator[Phenotype]:
 
     def __str__(self) -> str:
         return (f"Patient("
-                f"patient_id:{self.patient_id}, "
+                f"labels:{self._labels}, "
                 f"variants:{self.variants}, "
                 f"phenotypes:{[pheno.identifier for pheno in self.phenotypes]}, "
                 f"proteins:{[prot.protein_id for prot in self.proteins]})")