Overview

​

High-level description

This directory contains unit tests for the data module of the Cassiopeia project, which is likely a framework for phylogenetic analysis and lineage tracing. The tests cover various functionalities related to character matrices, tree topologies, bootstrapping, and utility functions used in phylogenetic analysis.

​

What does it do?

These tests ensure that the core data handling and manipulation functions of Cassiopeia work correctly. They verify:

1. Character matrix operations: Creating, modifying, and analyzing matrices that represent genetic or cellular states.
2. Tree topology handling: Converting trees to Newick format, reconstructing trees from character matrices, and manipulating tree structures.
3. Bootstrapping: Sampling techniques for character matrices and allele tables to assess statistical confidence.
4. Phylogenetic calculations: Computing least common ancestor characters, phylogenetic weight matrices, and inter-cluster distances.
5. Layer operations: Adding and managing different layers of data in the phylogenetic analysis.

These tests help maintain the reliability and accuracy of the Cassiopeia framework, ensuring that researchers can trust the results of their lineage tracing and phylogenetic analyses.

​

Key Files

1. cassiopeia_tree_test.py

   • Tests the CassiopeiaTree class, which is likely the core data structure for representing phylogenetic trees in the framework.
   • Covers operations like bootstrapping character matrices and allele tables, converting trees to Newick format, and calculating phylogenetic weights.

2. data_utilities_test.py

   • Tests various utility functions in the data module, including character matrix operations, bootstrap sampling, and dissimilarity map computations.
   • Ensures the correct functioning of helper functions that support the main phylogenetic analysis tasks.

3. layers_test.py

   • Tests the Layers class, which appears to handle multiple layers of character matrices or other data associated with phylogenetic trees.
   • Verifies operations like adding new layers, reconstructing trees with different layers, and handling various types of character matrices.

​

Dependencies

The test files rely on several external libraries and modules:

• unittest: The standard Python testing framework.
• networkx: Used for graph operations and tree topology manipulation.
• numpy and pandas: For data manipulation and analysis.
• ete3: A Python framework for phylogenetic tree manipulation.
• cassiopeia: The main package being tested, including its various modules like data, mixins, and preprocess.

​

Configuration

The tests use sample data and configurations set up within each test class. For example:

def setUp(self):
    self.character_matrix = pd.DataFrame({
        "r1": ["0", "1", "0"],
        "r2": ["1", "0", "1"]
    }, index=["cell1", "cell2", "cell3"])

    self.tree = nx.DiGraph()
    self.tree.add_edges_from([("root", "cell1"), ("root", "cell2"), ("root", "cell3")])

    self.ctree = CassiopeiaTree(character_matrix=self.character_matrix, tree=self.tree)

This setup creates a basic character matrix and tree topology used across multiple test cases in the TestLayers class.

The tests cover a wide range of scenarios and edge cases, ensuring robust functionality of the Cassiopeia framework. They include checks for error handling, such as raising appropriate exceptions when invalid inputs are provided:

def test_add_layer_with_incorrect_number_of_cells(self):
    modified_cm = self.character_matrix.copy()
    modified_cm.loc["cell4"] = ["0", "1"]
    with self.assertRaises(ValueError):
        self.ctree.add_layer(modified_cm, "test_layer")

Overall, these tests play a crucial role in maintaining the reliability and accuracy of the Cassiopeia framework for phylogenetic analysis and lineage tracing.