feat: add assays to existing investigations

altamisa/isatab/__init__.py

@@ -16,3 +16,4 @@
 from .validate_investigation import InvestigationValidator  # noqa: F401
 from .write_assay_study import AssayWriter, StudyWriter, RefTableBuilder  # noqa: F401
 from .write_investigation import InvestigationWriter  # noqa: F401
+from .modify_investigation import InvestigationForge  # noqa: F401

altamisa/isatab/modify_investigation.py

@@ -0,0 +1,88 @@
+from pathlib import Path
+from typing import Union
+
+import attr
+
+from altamisa.isatab import InvestigationReader
+
+__author__ = "Thomas Sell <[email protected]>"
+
+
+class InvestigationForge:
+    """
+    Provides methods to add assays to an existing investigation.
+
+    :type input_path: str | Path
+    :param input_path: Location of investigation to modify.
+    """
+
+    def __init__(self, input_path: Union[str, Path]):
+        i_file = Path(input_path).expanduser().resolve()
+
+        with i_file.open("rt") as f:
+            self.investigation = InvestigationReader.from_stream(f).read()
+
+        if len(self.investigation.studies) != 1:
+            # TODO: add support for multiple studies
+            raise IndexError("Only single study investigations are supported.")
+
+    @staticmethod
+    def _join_protocols(x: dict, y: dict) -> dict:
+        """
+        Join two dicts of ISA protocols.
+        For duplicate protocols, join parameter dicts.
+        """
+        for protocol in y:
+            if protocol not in x:
+                # no conflict > add protocol
+                x.update({protocol: y[protocol]})
+            else:
+                # same protocol > check parameters
+                old_parameters = x[protocol].parameters
+                new_parameters = y[protocol].parameters
+                for parameter in new_parameters:
+                    if parameter not in old_parameters:
+                        # no conflict > add parameter
+                        old_parameters.update({parameter: new_parameters[parameter]})
+                    else:
+                        # existing parameter: raise for ontology mismatch
+                        if new_parameters[parameter] != old_parameters[parameter]:
+                            raise ValueError(
+                                f'Ontology term mismatch for parameter "{parameter}" of protocol "{protocol}".'
+                            )
+        return x
+
+    def add_assay(self, input_path: Union[str, Path]):
+        """
+        Add assay to investigation file.
+
+        :type input_path: str | Path
+        :param input_path: Location of bare assay definition to add.
+        :rtype: models.InvestigationInfo
+        :returns: Investigation model including all information from the investigation file.
+        """
+        new_assay = Path(input_path).expanduser().resolve()
+        with new_assay.open("rt") as f:
+            reader = InvestigationReader(f)
+            new_assays = tuple(a for a in reader._read_study_assays())
+            new_protocols = {p.name: p for p in reader._read_study_protocols()}
+
+        old_protocols = self.investigation.studies[0].protocols
+        joined_protocols = self._join_protocols(
+            old_protocols,
+            new_protocols,
+        )
+
+        assays = self.investigation.studies[0].assays
+        assays += new_assays
+
+        updated_study = attr.evolve(
+            self.investigation.studies[0], protocols=joined_protocols, assays=assays
+        )
+        self.investigation = attr.evolve(self.investigation, studies=(updated_study,))
+
+    def add_protocol(self):
+        raise NotImplementedError
+
+    def remove_protocol(self):
+        raise NotImplementedError

tests/conftest.py

@@ -112,6 +112,14 @@ def small2_investigation_file():
         yield file
 
 
+@pytest.fixture
+def small2o_investigation_file():
+    """This file contains an ontology reference for one of the protocol parameters."""
+    path = os.path.join(os.path.dirname(__file__), "data/i_small2/i_small2_o.txt")
+    with open(path, "rt") as file:
+        yield file
+
+
 @pytest.fixture
 def small2_study_file():
     path = os.path.join(os.path.dirname(__file__), "data/i_small2/s_small2.txt")

tests/data/i_small2/i_small2.txt

@@ -72,9 +72,9 @@ Study Protocol Type Term Source REF
 Study Protocol Description						
 Study Protocol URI						
 Study Protocol Version						
-Study Protocol Parameters Name						
-Study Protocol Parameters Name Term Accession Number						
-Study Protocol Parameters Name Term Source REF						
+Study Protocol Parameters Name	method;instrument					
+Study Protocol Parameters Name Term Accession Number	;					
+Study Protocol Parameters Name Term Source REF	;					
 Study Protocol Components Name						
 Study Protocol Components Type						
 Study Protocol Components Type Term Accession Number						

tests/data/i_small2/i_small2_o.txt

@@ -0,0 +1,93 @@
+ONTOLOGY SOURCE REFERENCE
+Term Source Name	OBI	NCBITAXON	ROLEO
+Term Source File	http://data.bioontology.org/ontologies/OBI	http://data.bioontology.org/ontologies/NCBITAXON	http://data.bioontology.org/ontologies/ROLEO
+Term Source Version	31	8	1
+Term Source Description	Ontology for Biomedical Investigations	National Center for Biotechnology Information (NCBI) Organismal Classification	Role Ontology
+INVESTIGATION
+Investigation Identifier	i_small2
+Investigation Title	Small Investigation
+Investigation Description
+Investigation Submission Date
+Investigation Public Release Date
+INVESTIGATION PUBLICATIONS
+Investigation PubMed ID
+Investigation Publication DOI
+Investigation Publication Author List
+Investigation Publication Title
+Investigation Publication Status
+Investigation Publication Status Term Accession Number
+Investigation Publication Status Term Source REF
+INVESTIGATION CONTACTS
+Investigation Person Last Name
+Investigation Person First Name
+Investigation Person Mid Initials
+Investigation Person Email
+Investigation Person Phone
+Investigation Person Fax
+Investigation Person Address
+Investigation Person Affiliation
+Investigation Person Roles
+Investigation Person Roles Term Accession Number
+Investigation Person Roles Term Source REF
+STUDY
+Study Identifier	s_small2
+Study Title	Small Germline Study
+Study Description
+Comment[Study Grant Number]
+Comment[Study Funding Agency]
+Study Submission Date
+Study Public Release Date
+Study File Name	s_small2.txt
+STUDY DESIGN DESCRIPTORS
+Study Design Type
+Study Design Type Term Accession Number
+Study Design Type Term Source REF
+STUDY PUBLICATIONS
+Study PubMed ID
+Study Publication DOI
+Study Publication Author List
+Study Publication Title
+Study Publication Status
+Study Publication Status Term Accession Number
+Study Publication Status Term Source REF
+STUDY FACTORS
+Study Factor Name
+Study Factor Type
+Study Factor Type Term Accession Number
+Study Factor Type Term Source REF
+STUDY ASSAYS
+Study Assay File Name	a_small2.txt
+Study Assay Measurement Type	protein expression profiling
+Study Assay Measurement Type Term Accession Number	http://purl.obolibrary.org/obo/OBI_0000615
+Study Assay Measurement Type Term Source REF	OBI
+Study Assay Technology Type	mass spectrometry
+Study Assay Technology Type Term Accession Number	http://purl.obolibrary.org/obo/OBI_0000470
+Study Assay Technology Type Term Source REF	OBI
+Study Assay Technology Platform	LC-MS/MS
+STUDY PROTOCOLS
+Study Protocol Name	sample collection	extraction	labeling	chromatography	mass spectrometry	data transformation
+Study Protocol Type	sample collection	extraction	labeling	chromatography	mass spectrometry	data transformation
+Study Protocol Type Term Accession Number						
+Study Protocol Type Term Source REF						
+Study Protocol Description						
+Study Protocol URI						
+Study Protocol Version						
+Study Protocol Parameters Name	method;instrument					
+Study Protocol Parameters Name Term Accession Number	;123					
+Study Protocol Parameters Name Term Source REF	;OBI					
+Study Protocol Components Name						
+Study Protocol Components Type						
+Study Protocol Components Type Term Accession Number						
+Study Protocol Components Type Term Source REF						
+STUDY CONTACTS
+Study Person Last Name	Doe
+Study Person First Name	John
+Study Person Mid Initials	
+Study Person Email	
+Study Person Phone	
+Study Person Fax	
+Study Person Address	
+Study Person Affiliation	
+Study Person Roles	author
+Study Person Roles Term Accession Number	http://purl.obolibrary.org/obo/RoleO_0000061
+Study Person Roles Term Source REF	ROLEO

tests/test_modify_investigation.py

@@ -0,0 +1,30 @@
+import pytest
+
+from altamisa.isatab import InvestigationForge
+
+
+def test_forge(assays_investigation_file):
+    # multiple studies
+    with pytest.raises(IndexError):
+        InvestigationForge(assays_investigation_file.name)
+
+
+def test_add_assay(
+    small_investigation_file,
+    small2_investigation_file,
+    assays_investigation_file,
+    small2o_investigation_file,
+):
+    forge = InvestigationForge(small_investigation_file.name)
+    forge.add_assay(small2_investigation_file.name)
+    output = forge.investigation
+    assert len(output.studies[0].assays) == 2
+    assert len(output.studies[0].protocols) == 8
+
+    # multiple studies
+    with pytest.raises(IndexError):
+        forge.add_assay(assays_investigation_file.name)
+
+    # ontology reference mismatch
+    with pytest.raises(ValueError):
+        forge.add_assay(small2o_investigation_file.name)