spacr.timelapse

Module Contents

spacr.timelapse.link_by_iou(mask_prev, mask_next, iou_threshold=0.1)

spacr.timelapse.exponential_decay(x, a, b, c)

spacr.timelapse.preprocess_pathogen_data(pathogen_df)

spacr.timelapse.plot_data(measurement, group, ax, label, marker='o', linestyle='-')

spacr.timelapse.infected_vs_noninfected(result_df, measurement)

spacr.timelapse.save_figure(fig, src, figure_number)

spacr.timelapse.save_results_dataframe(df, src, results_name)

spacr.timelapse.summarize_per_well(peak_details_df)

spacr.timelapse.summarize_per_well_inf_non_inf(peak_details_df)

spacr.timelapse.analyze_calcium_oscillations(db_loc, measurement='cell_channel_1_mean_intensity', size_filter='cell_area', fluctuation_threshold=0.25, num_lines=None, peak_height=0.01, pathogen=None, cytoplasm=None, remove_transient=True, verbose=False, transience_threshold=0.9)

spacr.timelapse.create_results_figure()

Create a Figure with 3 subplots arranged as:

• PCA (top-left)

• XGBoost (top-right)

• Histogram (bottom spanning both columns)

Returns:

• fig (Figure)

• ax_pca, ax_xgb, ax_hist (matplotlib.axes.Axes)

spacr.timelapse.automated_motility_assay(settings)

End-to-end:

1. Read merged/*.npy (plate_well_field_time.npy)

2. Build intensity + cell/nucleus/pathogen masks, derive cytoplasm

3. Per cell & frame: metadata + cell regionprops

4. Aggregate child (nucleus/pathogen/cytoplasm) features per cell

5. Concatenate across all merged files

6. Clean impossible jumps + measurement glitches

7. Save per-cell measurements to SQLite DB (measurements/measurements.db, table=db_table_name) This table is always the original, pre-QC measurements.

8. Compute per-track velocities (after smoothing)

9. Save a well-level motility summary table in the same DB

10. Generate panel plots combining intensity + motility: - original (mask-based) infection labels - adjusted infection labels (if QC modifies labels)

11. Optional infection intensity QC based on pathogen channel.

New-relevant settings (all optional):

# Infection QC / strategy ‘infection_intensity_qc’: True/False ‘infection_intensity_strategy’: one of

{‘xgboost’, ‘histogram’, ‘pca’, ‘umap’, ‘tsne’}

‘infection_intensity_mode’: {‘relabel’, ‘remove’} # existing

# XGBoost ambiguous-band filtering (track-level) ‘infection_xgb_drop_ambiguous’: True/False (default True) ‘infection_xgb_ambiguous_low’: 0.25 (default) ‘infection_xgb_ambiguous_high’: 0.75 (default) ‘infection_xgb_proba_column’: ‘name_of_proba_col’ # optional override

# Histogram strategy ‘infection_hist_percentile’: 25 # used inside _apply_infection_intensity_qc

# Panel toggles ‘make_mask_panel’: True/False (default True) ‘make_adjusted_panel’: True/False (default True)

# Plot ranges (unchanged) - ‘motility_xlim’, ‘motility_ylim’ - ‘motility_origin_xlim’, ‘motility_origin_ylim’

# Measurements reuse ‘reuse_existing_measurements’: True/False (default True)