Skip to content

Refactor AlphaFold unit tests for clarity, robustness, and better coverage #1096

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
propulsive-guy wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Refactor AlphaFold unit tests for clarity, robustness, and better coverage #1096

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
271 changes: 161 additions & 110 deletions run_alphafold_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,9 +1,8 @@
# Copyright 2021 DeepMind Technologies Limited
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# You may not use this file except in compliance with the License.
# You may obtain a copy at:
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
Expand All @@ -12,126 +11,178 @@
# See the License for the specific language governing permissions and
# limitations under the License.



import json
import os
import tempfile
import glob
from unittest import mock

from absl.testing import absltest
from absl.testing import parameterized
import run_alphafold
from absl.testing import absltest, parameterized
import numpy as np
import run_alphafold

# Internal import (7716).

TEST_DATA_DIR = 'alphafold/common/testdata/'


class RunAlphafoldTest(parameterized.TestCase):

@parameterized.named_parameters(
('relax', run_alphafold.ModelsToRelax.ALL),
('no_relax', run_alphafold.ModelsToRelax.NONE),
)
def test_end_to_end(self, models_to_relax):

data_pipeline_mock = mock.Mock()
model_runner_mock = mock.Mock()
amber_relaxer_mock = mock.Mock()

data_pipeline_mock.process.return_value = {}
model_runner_mock.process_features.return_value = {
'aatype': np.zeros((12, 10), dtype=np.int32),
'residue_index': np.tile(np.arange(10, dtype=np.int32)[None], (12, 1)),
}
model_runner_mock.predict.return_value = {
'structure_module': {
'final_atom_positions': np.zeros((10, 37, 3)),
'final_atom_mask': np.ones((10, 37)),
},
'predicted_lddt': {
'logits': np.ones((10, 50)),
},
'plddt': np.ones(10) * 42,
'ranking_confidence': 90,
'ptm': np.array(0.0),
'aligned_confidence_probs': np.zeros((10, 10, 50)),
'predicted_aligned_error': np.zeros((10, 10)),
'max_predicted_aligned_error': np.array(0.0),
}
model_runner_mock.multimer_mode = False

with open(
os.path.join(
absltest.get_default_test_srcdir(), TEST_DATA_DIR, 'glucagon.pdb'
)
) as f:
pdb_string = f.read()
amber_relaxer_mock.process.return_value = (
pdb_string,
None,
[1.0, 0.0, 0.0],
"""End-to-end AlphaFold structure prediction test suite."""

def setUp(self):
"""Configure mocks and reusable test setup."""
super().setUp()
self.data_pipeline_mock = mock.Mock()
self.model_runner_mock = mock.Mock()
self.amber_relaxer_mock = mock.Mock()
self._configure_mocks()

def _configure_mocks(self):
"""Initialize default mock return values."""
self.data_pipeline_mock.process.return_value = {}
self.model_runner_mock.process_features.return_value = {
'aatype': np.zeros((12, 10), dtype=np.int32),
'residue_index': np.tile(np.arange(10, dtype=np.int32)[None], (12, 1)),
}
self.model_runner_mock.predict.return_value = {
'structure_module': {
'final_atom_positions': np.zeros((10, 37, 3)),
'final_atom_mask': np.ones((10, 37)),
},
'predicted_lddt': {'logits': np.ones((10, 50))},
'plddt': np.ones(10) * 42,
'ranking_confidence': 90,
'ptm': np.array(0.0),
'aligned_confidence_probs': np.zeros((10, 10, 50)),
'predicted_aligned_error': np.zeros((10, 10)),
'max_predicted_aligned_error': np.array(0.0),
}
self.model_runner_mock.multimer_mode = False

def _create_fasta(self, out_dir: str) -> str:
"""Helper to create a small FASTA file."""
fasta_path = os.path.join(out_dir, 'target.fasta')
with open(fasta_path, 'wt') as f:
f.write('>A\nAAAAAAAAAAAAA')
return fasta_path

def _read_test_pdb(self, pdb_filename: str) -> str:
"""Helper to load a test PDB structure."""
with open(
os.path.join(
absltest.get_default_test_srcdir(),
TEST_DATA_DIR,
pdb_filename
)
) as f:
return f.read()

@parameterized.named_parameters(
('relax_monomer', run_alphafold.ModelsToRelax.ALL, 'Monomer'),
('no_relax_monomer', run_alphafold.ModelsToRelax.NONE, 'Monomer'),
('relax_multimer', run_alphafold.ModelsToRelax.ALL, 'Multimer'),
)
def test_end_to_end(self, models_to_relax, model_type):
"""Ensures AlphaFold runs from FASTA → PDB end-to-end correctly."""
out_dir = tempfile.mkdtemp()
fasta_path = self._create_fasta(out_dir)
pdb_string = self._read_test_pdb('glucagon.pdb')

self.amber_relaxer_mock.process.return_value = (
pdb_string, None, [1.0, 0.0, 0.0]
)

out_dir = self.create_tempdir().full_path
fasta_path = os.path.join(out_dir, 'target.fasta')
with open(fasta_path, 'wt') as f:
f.write('>A\nAAAAAAAAAAAAA')
fasta_name = 'test'

run_alphafold.predict_structure(
fasta_path=fasta_path,
fasta_name=fasta_name,
output_dir_base=out_dir,
data_pipeline=data_pipeline_mock,
model_runners={'model1': model_runner_mock},
amber_relaxer=amber_relaxer_mock,
benchmark=False,
random_seed=0,
models_to_relax=models_to_relax,
model_type='Monomer',
)
fasta_name = 'test'

# Run AlphaFold prediction
run_alphafold.predict_structure(
fasta_path=fasta_path,
fasta_name=fasta_name,
output_dir_base=out_dir,
data_pipeline=self.data_pipeline_mock,
model_runners={'model1': self.model_runner_mock},
amber_relaxer=self.amber_relaxer_mock,
benchmark=False,
random_seed=0,
models_to_relax=models_to_relax,
model_type=model_type,
)

base_output_files = os.listdir(out_dir)
self.assertIn('target.fasta', base_output_files)
self.assertIn('test', base_output_files)

target_output_files = os.listdir(os.path.join(out_dir, 'test'))
expected_files = [
'confidence_model1.json',
'features.pkl',
'msas',
'pae_model1.json',
'ranked_0.cif',
'ranked_0.pdb',
'ranking_debug.json',
'result_model1.pkl',
'timings.json',
'unrelaxed_model1.cif',
'unrelaxed_model1.pdb',
]
if models_to_relax == run_alphafold.ModelsToRelax.ALL:
expected_files.extend(
['relaxed_model1.cif', 'relaxed_model1.pdb', 'relax_metrics.json']
)
with open(os.path.join(out_dir, 'test', 'relax_metrics.json')) as f:
relax_metrics = json.loads(f.read())
self.assertDictEqual(
{
'model1': {
'remaining_violations': [1.0, 0.0, 0.0],
'remaining_violations_count': 1.0,
}
},
relax_metrics,
)
self.assertCountEqual(expected_files, target_output_files)

# Check that pLDDT is set in the B-factor column.
with open(os.path.join(out_dir, 'test', 'unrelaxed_model1.pdb')) as f:
for line in f:
if line.startswith('ATOM'):
self.assertEqual(line[61:66], '42.00')
# Validate output directory structure
base_files = os.listdir(out_dir)
self.assertIn('target.fasta', base_files, msg="FASTA file missing in output.")
self.assertIn('test', base_files, msg="Model output folder missing.")

target_dir = os.path.join(out_dir, 'test')
output_files = os.listdir(target_dir)

expected_files = [
'confidence_model1.json',
'features.pkl',
'msas',
'pae_model1.json',
'ranked_0.cif',
'ranked_0.pdb',
'ranking_debug.json',
'result_model1.pkl',
'timings.json',
'unrelaxed_model1.cif',
'unrelaxed_model1.pdb',
]

if models_to_relax == run_alphafold.ModelsToRelax.ALL:
expected_files += [
'relaxed_model1.cif',
'relaxed_model1.pdb',
'relax_metrics.json',
]

with open(os.path.join(target_dir, 'relax_metrics.json')) as f:
relax_metrics = json.load(f)

self.assertDictEqual(
{
'model1': {
'remaining_violations': [1.0, 0.0, 0.0],
'remaining_violations_count': 1.0,
}
},
relax_metrics,
msg="Relaxation metrics content mismatch.",
)

# File validation
for f in expected_files:
file_path = os.path.join(target_dir, f)
self.assertTrue(os.path.exists(file_path), msg=f"Missing: {f}")
# Ensure non-empty files (except msas folder)
if not os.path.isdir(file_path):
self.assertGreater(os.path.getsize(file_path), 0, msg=f"Empty file: {f}")

# Validate B-factor correctness (pLDDT stored)
pdb_path = os.path.join(target_dir, 'unrelaxed_model1.pdb')
with open(pdb_path) as f:
atom_lines = [line for line in f if line.startswith('ATOM')]
self.assertTrue(atom_lines, msg="No ATOM records found in PDB file.")
for line in atom_lines:
self.assertEqual(line[61:66], '42.00', msg="Incorrect pLDDT value in B-factor.")

def test_invalid_fasta_path_raises_error(self):
"""Verify that an invalid FASTA path raises a FileNotFoundError."""
with self.assertRaises(FileNotFoundError):
run_alphafold.predict_structure(
fasta_path='invalid_path.fasta',
fasta_name='invalid',
output_dir_base='.',
data_pipeline=self.data_pipeline_mock,
model_runners={'model1': self.model_runner_mock},
amber_relaxer=self.amber_relaxer_mock,
benchmark=False,
random_seed=0,
models_to_relax=run_alphafold.ModelsToRelax.NONE,
model_type='Monomer',
)


if __name__ == '__main__':
absltest.main()
absltest.main()