Skip to content

Commit

Permalink
adjustments for single timeframe
Browse files Browse the repository at this point in the history
  • Loading branch information
@jo-mueller
jo-mueller committed Sep 26, 2023
1 parent 65b01bf commit 7e5ea52
Showing 1 changed file with 57 additions and 32 deletions.
89 changes: 57 additions & 32 deletions src/napari_stress/_utils/_aggregate_measurements.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -119,7 +119,7 @@ def find_metadata_in_layers(layers: list, name: str
import pandas as pd
for layer in layers:
if 'metadata' in layer[1].keys():
if name in pd.DataFrame(layer[1]['metadata']).columns:
if name in layer[1]['metadata'].keys():
return pd.DataFrame(layer[1]['metadata'])
if 'features' in layer[1].keys():
if name in pd.DataFrame(layer[1]['features']).columns:
Expand Down Expand Up @@ -180,9 +180,8 @@ def flatten_dictionary(input_dict, parent_key_prefix='', separator='_'):
# Single values over time
_metadata = [layer[1]['metadata'] for layer in results_stress_analysis if 'metadata' in layer[1].keys()]
_metadata = [flatten_dictionary(d) for d in _metadata]
df_over_time = pd.DataFrame(_metadata[0])
for d in _metadata[1:]:
df_over_time = df_over_time.merge(pd.DataFrame(d), on='frame', how='inner')
df_over_time = pd.concat([pd.DataFrame(x) for x in _metadata], axis=1)
df_over_time = df_over_time.loc[:,~df_over_time.columns.duplicated()].copy()

# Find layer with stress_tissue in features
for layer in results_stress_analysis:
Expand All @@ -198,11 +197,11 @@ def flatten_dictionary(input_dict, parent_key_prefix='', separator='_'):

df_over_time['time'] = df_over_time['frame'] * time_step
df_over_time[_METADATAKEY_AUTOCORR_TEMPORAL_TOTAL] = temporal_autocorrelation(
df_total_stress, 'stress_total_radial', frame_column_name='time')
df_total_stress, 'stress_total_radial', frame_column_name='frame')
df_over_time[_METADATAKEY_AUTOCORR_TEMPORAL_CELL] = temporal_autocorrelation(
df_total_stress, 'stress_cell', frame_column_name='time')
df_total_stress, 'stress_cell', frame_column_name='frame')
df_over_time[_METADATAKEY_AUTOCORR_TEMPORAL_TISSUE] = temporal_autocorrelation(
df_tissue_stress, 'stress_tissue', frame_column_name='time')
df_tissue_stress, 'stress_tissue', frame_column_name='frame')

return df_over_time

Expand Down Expand Up @@ -259,14 +258,20 @@ def aggregate_extrema_results(results_stress_analysis: List[LayerDataTuple],

# stack keys of metadata into dataframe and add frame column
metadata = layer[1]['metadata']
frames = np.concatenate(
[[i] * len(metadata[_METADATAKEY_STRESS_CELL_NEAREST_PAIR_ANISO][i]
) for i in range(n_frames)]
) * time_step
min_max_pair_distances = np.concatenate(
metadata[_METADATAKEY_STRESS_CELL_NEAREST_PAIR_DIST])
min_max_pair_anisotropies = np.concatenate(
metadata[_METADATAKEY_STRESS_CELL_NEAREST_PAIR_ANISO])
if n_frames > 1:
frames = np.concatenate(
[[i] * len(metadata[_METADATAKEY_STRESS_CELL_NEAREST_PAIR_ANISO][i]
) for i in range(n_frames)]
) * time_step
min_max_pair_distances = np.concatenate(
metadata[_METADATAKEY_STRESS_CELL_NEAREST_PAIR_DIST])
min_max_pair_anisotropies = np.concatenate(
metadata[_METADATAKEY_STRESS_CELL_NEAREST_PAIR_ANISO])
else:
frames = np.array([0])
min_max_pair_distances = [metadata[_METADATAKEY_STRESS_CELL_NEAREST_PAIR_DIST]]
min_max_pair_anisotropies = [metadata[_METADATAKEY_STRESS_CELL_NEAREST_PAIR_ANISO]]


df_nearest_pair = pd.DataFrame(
{'frame': frames,
Expand All @@ -283,14 +288,19 @@ def aggregate_extrema_results(results_stress_analysis: List[LayerDataTuple],

# stack keys of metadata into dataframe and add frame column
metadata = layer[1]['metadata']
frames = np.concatenate(
[[i] * len(metadata[_METADATAKEY_STRESS_CELL_ALL_PAIR_ANISO][i]
) for i in range(n_frames)]
) * time_step
all_pair_distances = np.concatenate(
metadata[_METADATAKEY_STRESS_CELL_ALL_PAIR_DIST])
all_pair_anisotropies = np.concatenate(
metadata[_METADATAKEY_STRESS_CELL_ALL_PAIR_ANISO])
if n_frames > 1:
frames = np.concatenate(
[[i] * len(metadata[_METADATAKEY_STRESS_CELL_ALL_PAIR_ANISO][i]
) for i in range(n_frames)]
) * time_step
all_pair_distances = np.concatenate(
metadata[_METADATAKEY_STRESS_CELL_ALL_PAIR_DIST])
all_pair_anisotropies = np.concatenate(
metadata[_METADATAKEY_STRESS_CELL_ALL_PAIR_ANISO])
else:
frames = np.array([0])
all_pair_distances = [metadata[_METADATAKEY_STRESS_CELL_ALL_PAIR_DIST]]
all_pair_anisotropies = [metadata[_METADATAKEY_STRESS_CELL_ALL_PAIR_ANISO]]

df_all_pair = pd.DataFrame(
{'frame': frames,
Expand Down Expand Up @@ -347,9 +357,14 @@ def aggregate_spatial_autocorrelations_results(results_stress_analysis: List[Lay

# TOTAL STRESS
metadata = layer[1]['metadata'][_METADATAKEY_AUTOCORR_SPATIAL_TOTAL]
distances = [metadata[t]['auto_correlations_distances'] for t in range(n_frames)]
normalized_autocorrelation_total = [metadata[t]['auto_correlations_averaged_normalized'] for t in range(n_frames)]
frames = [[t] * len(metadata[t]['auto_correlations_averaged_normalized']) for t in range(n_frames)]
if n_frames > 1:
distances = [metadata[t]['auto_correlations_distances'] for t in range(n_frames)]
normalized_autocorrelation_total = [metadata[t]['auto_correlations_averaged_normalized'] for t in range(n_frames)]
frames = [[t] * len(metadata[t]['auto_correlations_averaged_normalized']) for t in range(n_frames)]
else:
distances = [metadata['auto_correlations_distances']]
normalized_autocorrelation_total = [metadata['auto_correlations_averaged_normalized']]
frames = [[0] * len(metadata['auto_correlations_averaged_normalized'])]

df_autocorrelations_total = pd.DataFrame(
{'time': np.concatenate(frames).squeeze() * time_step,
Expand All @@ -359,9 +374,14 @@ def aggregate_spatial_autocorrelations_results(results_stress_analysis: List[Lay

# CELL STRESS
metadata = layer[1]['metadata'][_METADATAKEY_AUTOCORR_SPATIAL_CELL]
distances = [metadata[t]['auto_correlations_distances'] for t in range(n_frames)]
normalized_autocorrelation_cell = [metadata[t]['auto_correlations_averaged_normalized'] for t in range(n_frames)]
frames = [[t] * len(metadata[t]['auto_correlations_averaged_normalized']) for t in range(n_frames)]
if n_frames > 1:
distances = [metadata[t]['auto_correlations_distances'] for t in range(n_frames)]
normalized_autocorrelation_cell = [metadata[t]['auto_correlations_averaged_normalized'] for t in range(n_frames)]
frames = [[t] * len(metadata[t]['auto_correlations_averaged_normalized']) for t in range(n_frames)]
else:
distances = [metadata['auto_correlations_distances']]
normalized_autocorrelation_cell = [metadata['auto_correlations_averaged_normalized']]
frames = [[0] * len(metadata['auto_correlations_averaged_normalized'])]

df_autocorrelations_cell = pd.DataFrame(
{'time': np.concatenate(frames).squeeze() * time_step,
Expand All @@ -371,9 +391,14 @@ def aggregate_spatial_autocorrelations_results(results_stress_analysis: List[Lay

# TISSUE STRESS
metadata = layer[1]['metadata'][_METADATAKEY_AUTOCORR_SPATIAL_TISSUE]
distances = [metadata[t]['auto_correlations_distances'] for t in range(n_frames)]
normalized_autocorrelation_tissue = [metadata[t]['auto_correlations_averaged_normalized'] for t in range(n_frames)]
frames = [[t] * len(metadata[t]['auto_correlations_averaged_normalized']) for t in range(n_frames)]
if n_frames > 1:
distances = [metadata[t]['auto_correlations_distances'] for t in range(n_frames)]
normalized_autocorrelation_tissue = [metadata[t]['auto_correlations_averaged_normalized'] for t in range(n_frames)]
frames = [[t] * len(metadata[t]['auto_correlations_averaged_normalized']) for t in range(n_frames)]
else:
distances = [metadata['auto_correlations_distances']]
normalized_autocorrelation_tissue = [metadata['auto_correlations_averaged_normalized']]
frames = [[0] * len(metadata['auto_correlations_averaged_normalized'])]

df_autocorrelations_tissue = pd.DataFrame(
{'time': np.concatenate(frames).squeeze() * time_step,
Expand Down

0 comments on commit 7e5ea52

Please sign in to comment.