transmart-loader 1.4.1

Usage

Define a TranSMART data collection, using the classes in transmart_loader/transmart.py, e.g.,

# Create the dimension elements
age_concept = Concept('test:age', 'Age', '\\Test\\age', ValueType.Numeric)
concepts = [age_concept]
studies = [Study('test', 'Test study')]
trial_visits = [TrialVisit(studies[0], 'Week 1', 'Week', 1)]
patients = [Patient('SUBJ0', 'male', [])]
visits = [Visit(patients[0], 'visit1', None, None, None, None, None, None, [])]
# Create the observations
observations = [
    Observation(patients[0], age_concept, visits[0], trial_visits[0],
                date(2019, 3, 28), None, NumericalValue(28))]

Create a hierarchical ontology for the concepts, e.g., to create the following structure:

└ Ontology
  └ Age

# Create an ontology with one top node and a concept node
top_node = TreeNode('Ontology')
top_node.add_child(ConceptNode(concepts[0]))
ontology = [top_node]

Write the data collection to a format that can be loaded using transmart-copy:

collection = DataCollection(concepts, [], [], studies,
                            trial_visits, visits, ontology, patients, observations)

# Write collection to a temporary directory
# The generated files can be loaded into TranSMART with transmart-copy.
output_dir = mkdtemp()
copy_writer = TransmartCopyWriter(output_dir)
copy_writer.write_collection(collection)

Check examples/data_collection.py for a complete example.

Usage examples can be found in these projects:

• fhir2transmart: a tool that translates core HL7 FHIR resources to the TranSMART data model.

• claml2transmart: a tool that translates ontologies in ClaML format (e.g., ICD-10, available from DIMDI) to TranSMART ontologies.

• csr2transmart: a custom data transformation and loading pipeline for a Dutch center for pediatric oncology.

• transmart-hyper-dicer: a tool that reads a selection of data from a TranSMART instance using its REST API and loads it into another TranSMART instance.

Python versions

This packages is tested with Python versions >= 3.7.

Package management and dependencies

This project uses pip for installing dependencies and package management.

• Dependencies should be added to setup.py in the install_requires list.

Testing and code coverage

• Tests are in the tests folder.

• The tests folder contains:

  • A test if files for transmart-copy are generated for fake data (file: test_transmart_loader)

  • A test that checks whether your code conforms to the Python style guide (PEP 8) (file: test_lint.py)

• The testing framework used is PyTest

  • PyTest introduction

• Tests can be run with python setup.py test

Documentation

• Documentation should be put in the docs folder.

• To generate html documentation run python setup.py build_sphinx

Coding style conventions and code quality

• Check your code style with prospector

• You may need run pip install .[dev] first, to install the required dependencies