Skip to content
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

Jump to bottom

Update uns to precomp stats to take a list as parameter #21

Open
wants to merge 4 commits into
base: rc/v1.4.0
Choose a base branch
from
Open

Update uns to precomp stats to take a list as parameter #21

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
7cdd378
modified uns_to_precomputed_stats to take a list of uns as parameter
Jul 14, 2024
a837b0e
removed unnecessary import
Jul 14, 2024
1082846
int casting for cluster to row lookup
Jul 20, 2024
f58532e
float to int casting for gt0, gt1, ge1
Jul 21, 2024
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
33 changes: 28 additions & 5 deletions src/cell_type_mapper/diff_exp/score_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -187,8 +187,10 @@ def read_raw_precomputed_stats(
precomputed_stats['gene_names'] = json.loads(
in_file['col_names'][()].decode('utf-8'))

row_lookup = json.loads(
in_file['cluster_to_row'][()].decode('utf-8'))
row_lookup = _read_cluster_to_row(
json.loads(
in_file['cluster_to_row'][()].decode('utf-8'))
)

all_keys = set(['n_cells', 'sum', 'sumsq', 'gt0', 'gt1', 'ge1'])
all_keys = list(all_keys.intersection(set(in_file.keys())))
Expand Down Expand Up @@ -286,13 +288,13 @@ def aggregate_stats(
sumsq_arr += these_stats['sumsq']

if 'gt0' in these_stats:
gt0 += these_stats['gt0']
gt0 += list(map(int, these_stats['gt0']))

if 'gt1' in these_stats:
gt1 += these_stats['gt1']
gt1 += list(map(int, these_stats['gt1']))

if 'ge1' in these_stats:
ge1 += these_stats['ge1']
ge1 += list(map(int, these_stats['ge1']))
else:
has_ge1 = False

Expand All @@ -317,3 +319,24 @@ def aggregate_stats(
result[k] = result[k].astype(new_dtype)

return result


def _read_cluster_to_row(cluster_row_lookup):
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please add a unit test for this function. I know it's trivial, I just want to make sure that this is actually solving the problem we want it to solve.

"""
Take cluster row lookup dict.
Return cluster row lookup dict with int casted values.

Parameters
----------
cluster_row_lookup:
A dictionary mapping cluster names to index values
for looking up n_cells for a given cluster.

Returns
-------
cluster_to_row
"""
cluster_to_row = {}
for cluster, row in cluster_row_lookup.items():
cluster_to_row[cluster] = int(row)
return cluster_to_row
7 changes: 5 additions & 2 deletions src/cell_type_mapper/utils/output_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -555,7 +555,7 @@ def precomputed_stats_to_uns(

def uns_to_precomputed_stats(
h5ad_path,
uns_key,
uns_keys_list,
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please update the docstring to reflect the new argument.

tmp_dir=None):
"""
Read a serialized precomputed stats file from the uns element
Expand Down Expand Up @@ -593,7 +593,10 @@ def uns_to_precomputed_stats(
['n_cells', 'sum', 'sumsq', 'ge1', 'gt1', 'gt0']
)

serialized_data = read_uns_from_h5ad(h5ad_path)[uns_key]
serialized_data = read_uns_from_h5ad(h5ad_path)
for uns_key in uns_keys_list:
serialized_data = serialized_data[uns_key]

with h5py.File(h5_path, 'w') as dst:
for dataset_name in serialized_data:
data = serialized_data[dataset_name]
Expand Down
23 changes: 18 additions & 5 deletions tests/utils/test_output_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -468,9 +468,11 @@ def test_re_order_blob(tmp_dir_fixture):
assert len(new_blob) == len(results_lookup)


@pytest.mark.parametrize('nested_uns', [True, False])
def test_precomputed_stats_to_uns(
tmp_dir_fixture,
precomputed_stats_fixture):
precomputed_stats_fixture,
nested_uns):
"""
Test utility functions to move precomputed stats data from HDF5 to
the uns element of an h5ad file and back.
Expand Down Expand Up @@ -507,16 +509,26 @@ def test_precomputed_stats_to_uns(
a_data.write_h5ad(h5ad_path)

uns_key = 'serialization_test'
uns_key_nested = 'serialization_test_nested'

precomputed_stats_to_uns(
precomputed_stats_path=precomputed_stats_fixture,
h5ad_path=h5ad_path,
uns_key=uns_key)

roundtrip_path = uns_to_precomputed_stats(
uns_key=uns_key,
h5ad_path=h5ad_path,
tmp_dir=tmp_dir_fixture)
if not nested_uns:
roundtrip_path = uns_to_precomputed_stats(
uns_keys_list=[uns_key],
h5ad_path=h5ad_path,
tmp_dir=tmp_dir_fixture)
else:
a_data = anndata.read_h5ad(h5ad_path)
a_data.uns[uns_key] = {uns_key_nested: a_data.uns[uns_key]}
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Just being paranoid:

The way this code is constructed, the file at h5ad_path still has the expected data at [uns_key] which means that, even if uns_to_precomputed_stats were not modified, the test would still pass, I think.

Can you write a new anndata object at a different location that only has the expected data at your nested location. I am imagining something like this:

new_h5ad_path = mkstemp_clean(dir=tmp_dir_fixture, suffix='.h5ad')
src = anndata.read_h5ad(h5ad_path)
new_data = anndata.AnnData(
    obs=src.obs,
    var=src.var,
    X=src.X[()],
    uns = {uns_key_nested: {uns_key: {src.uns[uns_key]}}}
)
new_data.write_h5ad(new_h5ad_path)
h5ad_path = new_h5ad_path

a_data.write_h5ad(h5ad_path)
roundtrip_path = uns_to_precomputed_stats(
uns_keys_list=[uns_key, uns_key_nested],
h5ad_path=h5ad_path,
tmp_dir=tmp_dir_fixture)

with h5py.File(precomputed_stats_fixture, 'r') as expected_src:
with h5py.File(roundtrip_path, 'r') as actual_src:
Expand All @@ -540,3 +552,4 @@ def test_precomputed_stats_to_uns(
original_uns['maybe'],
roundtrip_h5ad.uns['maybe']
)