def test_filter_and_concat_datasets():
cortex_dataset_1 = CortexDataset(save_path='tests/data/')
cortex_dataset_1.subsample_genes(subset_genes=np.arange(0, 3))
cortex_dataset_1.filter_cell_types(["microglia", "oligodendrocytes"])
cortex_dataset_2 = CortexDataset(save_path='tests/data/')
cortex_dataset_2.subsample_genes(subset_genes=np.arange(1, 4))
cortex_dataset_2.filter_cell_types(["endothelial-mural", "interneurons", "microglia", "oligodendrocytes"])
cortex_dataset_2.filter_cell_types([2, 0])
cortex_dataset_merged = GeneExpressionDataset.concat_datasets(cortex_dataset_1, cortex_dataset_2)
assert cortex_dataset_merged.nb_genes == 2
synthetic_dataset_1 = SyntheticDataset(n_batches=2, n_labels=5)
synthetic_dataset_2 = SyntheticDataset(n_batches=3, n_labels=3)
synthetic_merged_1 = GeneExpressionDataset.concat_datasets(synthetic_dataset_1, synthetic_dataset_2)
assert synthetic_merged_1.n_batches == 5
assert synthetic_merged_1.n_labels == 5
synthetic_merged_2 = GeneExpressionDataset.concat_datasets(synthetic_dataset_1, synthetic_dataset_2,
shared_labels=False)
assert synthetic_merged_2.n_batches == 5
assert synthetic_merged_2.n_labels == 8
synthetic_dataset_1.filter_cell_types([0, 1, 2, 3])
assert synthetic_dataset_1.n_labels == 4
synthetic_dataset_1.subsample_cells(50)
assert len(synthetic_dataset_1) == 50
synthetic_dataset_3 = SyntheticDataset(n_labels=6)
synthetic_dataset_3.cell_types = np.arange(6).astype(np.str)
synthetic_dataset_3.map_cell_types({"2": "9", ("4", "3"): "8"})
def test_synthetic_1():
synthetic_dataset = SyntheticDataset()
synthetic_dataset.cell_types = np.array(['A', 'B', 'C'])
svaec = SCANVI(synthetic_dataset.nb_genes, synthetic_dataset.n_batches,
synthetic_dataset.n_labels)
trainer_synthetic_svaec = JointSemiSupervisedTrainer(svaec,
synthetic_dataset,
use_cuda=use_cuda)
trainer_synthetic_svaec.train(n_epochs=1)
trainer_synthetic_svaec.labelled_set.entropy_batch_mixing()
trainer_synthetic_svaec.full_dataset.knn_purity(verbose=True)
trainer_synthetic_svaec.labelled_set.show_t_sne(n_samples=5)
trainer_synthetic_svaec.unlabelled_set.show_t_sne(n_samples=5,
color_by='labels')
trainer_synthetic_svaec.labelled_set.show_t_sne(
n_samples=5, color_by='batches and labels')
trainer_synthetic_svaec.labelled_set.clustering_scores()
trainer_synthetic_svaec.labelled_set.clustering_scores(
prediction_algorithm='gmm')
trainer_synthetic_svaec.unlabelled_set.unsupervised_classification_accuracy(
)
trainer_synthetic_svaec.unlabelled_set.differential_expression_score(
'B', 'C', genes=['2', '4'], M_sampling=2, M_permutation=10)
trainer_synthetic_svaec.unlabelled_set.differential_expression_table(
M_sampling=2, M_permutation=10)
def test_synthetic_1():
synthetic_dataset = SyntheticDataset()
synthetic_dataset.cell_types = np.array(["A", "B", "C"])
svaec = SCANVI(
synthetic_dataset.nb_genes,
synthetic_dataset.n_batches,
synthetic_dataset.n_labels,
)
trainer_synthetic_svaec = JointSemiSupervisedTrainer(
svaec, synthetic_dataset, use_cuda=use_cuda
)
trainer_synthetic_svaec.train(n_epochs=1)
trainer_synthetic_svaec.labelled_set.entropy_batch_mixing()
trainer_synthetic_svaec.full_dataset.knn_purity()
trainer_synthetic_svaec.labelled_set.show_t_sne(n_samples=5)
trainer_synthetic_svaec.unlabelled_set.show_t_sne(n_samples=5, color_by="labels")
trainer_synthetic_svaec.labelled_set.show_t_sne(
n_samples=5, color_by="batches and labels"
)
trainer_synthetic_svaec.labelled_set.clustering_scores()
trainer_synthetic_svaec.labelled_set.clustering_scores(prediction_algorithm="gmm")
trainer_synthetic_svaec.unlabelled_set.unsupervised_classification_accuracy()
trainer_synthetic_svaec.unlabelled_set.differential_expression_score(
synthetic_dataset.labels.ravel() == 1,
synthetic_dataset.labels.ravel() == 2,
n_samples=2,
M_permutation=10,
)
trainer_synthetic_svaec.unlabelled_set.one_vs_all_degenes(
n_samples=2, M_permutation=10
)
def test_synthetic_1():
synthetic_dataset = SyntheticDataset()
synthetic_dataset.cell_types = np.array(["A", "B", "C"])
svaec = SCANVI(
synthetic_dataset.nb_genes,
synthetic_dataset.n_batches,
synthetic_dataset.n_labels,
)
trainer_synthetic_svaec = JointSemiSupervisedTrainer(svaec,
synthetic_dataset,
use_cuda=use_cuda)
trainer_synthetic_svaec.train(n_epochs=1)
trainer_synthetic_svaec.labelled_set.entropy_batch_mixing()
with tempfile.TemporaryDirectory() as temp_dir:
posterior_save_path = os.path.join(temp_dir, "posterior_data")
original_post = trainer_synthetic_svaec.labelled_set.sequential()
original_post.save_posterior(posterior_save_path)
new_svaec = SCANVI(
synthetic_dataset.nb_genes,
synthetic_dataset.n_batches,
synthetic_dataset.n_labels,
)
new_post = load_posterior(posterior_save_path,
model=new_svaec,
use_cuda=False)
assert np.array_equal(new_post.indices, original_post.indices)
assert np.array_equal(new_post.gene_dataset.X,
original_post.gene_dataset.X)
assert np.array_equal(new_post.gene_dataset.labels,
original_post.gene_dataset.labels)
trainer_synthetic_svaec.full_dataset.knn_purity()
trainer_synthetic_svaec.labelled_set.show_t_sne(n_samples=5)
trainer_synthetic_svaec.unlabelled_set.show_t_sne(n_samples=5,
color_by="labels")
trainer_synthetic_svaec.labelled_set.show_t_sne(
n_samples=5, color_by="batches and labels")
trainer_synthetic_svaec.labelled_set.clustering_scores()
trainer_synthetic_svaec.labelled_set.clustering_scores(
prediction_algorithm="gmm")
trainer_synthetic_svaec.unlabelled_set.unsupervised_classification_accuracy(
)
trainer_synthetic_svaec.unlabelled_set.differential_expression_score(
synthetic_dataset.labels.ravel() == 1,
synthetic_dataset.labels.ravel() == 2,
n_samples=2,
M_permutation=10,
)
trainer_synthetic_svaec.unlabelled_set.one_vs_all_degenes(n_samples=2,
M_permutation=10)
