Legacy Module

legacy.py

Deprecated functions retained for backward compatibility only.

Do not use these in new code; they may be removed in a future version. Use the corresponding functions from the main exo2micro modules instead.

exo2micro.legacy.compile_ims_old(target_ch, ref_ch, dirname='./', name_str='')[source]

Deprecated since version Use: load_image_pair() from exo2micro.utils instead.

Given a directory of .tif image files, read them into arrays. Assumes a target channel and variable reference channels.

Parameters:

  • target_ch (str) – Single 2-digit string for the target channel, e.g. ‘00’.

  • ref_ch (list of str) – List of 2-digit strings for reference channels, e.g. [‘01’, ‘02’].

  • dirname (str) – Directory containing the images (default ‘./’).

  • name_str (str) – Partial filename string to select a specific set (default ‘’).

Return type:

target_fname, ref_fnames, target_im, ref_ims

exo2micro.legacy.mask_em(red)[source]

Deprecated since version Use: exo2micro.utils.build_tissue_mask() instead.

Create a binary mask: 1.0 where red > 0, NaN elsewhere.

Parameters:

red (ndarray)

Return type:

ndarray

exo2micro.legacy.maskandsave(blue_img, red_img, green_img, sub00, sub02)[source]

Deprecated since version Use: the SampleDye pipeline for output management instead.

Apply mask and save processed images to Analyzed_Images/.

Parameters:

  • blue_img (str) – Filename strings for channel TIFFs.

  • red_img (str) – Filename strings for channel TIFFs.

  • green_img (str) – Filename strings for channel TIFFs.

  • sub00 (ndarray) – Processed channel images.

  • sub02 (ndarray) – Processed channel images.

exo2micro.legacy.optimize_subtraction(post_im, pre_im, method='least_squares', percentile=None, n_hist_bins=100, noise_floor_percentile=5, boundary_erosion=50, mask=None, signal_percentile=50, plot_ratio_hist=False, sample='', dye='', save_prefix=None)[source]

Deprecated since version 2.3: Moved from exo2micro.scaling into exo2micro.legacy and no longer used by the main pipeline. Use SampleDye(scale_percentile=...) for the equivalent percentile-based workflow in the current pipeline.

Compute a pre-stain scale factor for subtraction.

Parameters:

  • post_im (ndarray) – Post-stain and registered pre-stain images (2-D).

  • pre_im (ndarray) – Post-stain and registered pre-stain images (2-D).

  • method (str) – 'least_squares' or 'robust_percentile' (default 'least_squares').

  • percentile (int or None) – For robust_percentile: None = histogram mode, int = use that percentile of the log-ratio distribution.

  • n_hist_bins (int) – Histogram bins for the log-ratio distribution (default 100).

  • noise_floor_percentile (float) – Pixels below this percentile of post_im are excluded (default 5).

  • boundary_erosion (int) – Erode the signal mask by this many pixels (default 50).

  • mask (ndarray or None) – Optional boolean base mask.

  • signal_percentile (float) – Bright-in-both percentile (default 50).

  • plot_ratio_hist (bool) – If True, also return a data dict for plotting (default False).

  • sample (str) – Labels for plots.

  • dye (str) – Labels for plots.

  • save_prefix (str or None) – If set, save diagnostic figures to disk.

Returns:

  • opt_scale (float)

  • scale_sig (float or None)

  • tissue_mask_out (ndarray of bool)

  • plot_data (dict or None)

exo2micro.legacy.plot_diff_comparison(post_im, pre_im, scale_ls, scale_robust, title='', sig_mask=None, sample='', dye='')[source]

Two-panel comparison of LS and robust difference images.

Parameters:

  • post_im (ndarray)

  • pre_im (ndarray)

  • scale_ls (float)

  • scale_robust (float)

  • title (str)

  • sig_mask (ndarray or None (retained for API compatibility))

  • sample (str)

  • dye (str)

Return type:

fig, diff_ls, diff_rob

exo2micro.legacy.plot_im_sub(post_im, pre_im, scale, comp=None, scale_robust=None, sample='', dye='')[source]

Subtract a scaled pre-stain image and display panels.

Parameters:

  • post_im (ndarray) – Post-stain and aligned pre-stain images.

  • pre_im (ndarray) – Post-stain and aligned pre-stain images.

  • scale (float) – Least-squares scale factor.

  • comp (ndarray or None) – Comparison image.

  • scale_robust (float or None) – If set, add robust subtraction panels.

  • sample (str) – For title.

  • dye (str) – For title.

Returns:

  • fig (matplotlib.Figure)

  • diff_im (ndarray) – LS difference image.

exo2micro.legacy.plot_ratio_histogram(plot_data, save_path=None)[source]

Diagnostic histogram of post/pre ratio distribution.

Parameters:

  • plot_data (dict) – Data from optimize_subtraction (plot_data output).

  • save_path (str or None) – If set, save to this path.

Returns:

fig

Return type:

matplotlib.Figure

exo2micro.legacy.plot_residual_histogram(plot_data, save_path=None)[source]

Post-subtraction residual histogram comparing LS and robust methods.

Parameters:

  • plot_data (dict) – Data from optimize_subtraction.

  • save_path (str or None) – If set, save to this path.

Returns:

fig

Return type:

matplotlib.Figure or None

exo2micro.legacy.plot_signal_scatter(post_im, pre_im, scale_ls, scale_sig, scale_robust, tissue_mask, signal_percentile=50, n_points=100000, sample='', dye='', save_path=None)[source]

Diagnostic scatter plot of post vs pre brightness.

Shows scale lines for LS, signal-only LS, and robust methods on a log-log density plot.

Parameters:

  • post_im (ndarray) – Post-stain and aligned pre-stain images.

  • pre_im (ndarray) – Post-stain and aligned pre-stain images.

  • scale_ls (float) – LS scale over all tissue pixels.

  • scale_sig (float or None) – LS scale over signal-only pixels.

  • scale_robust (float) – Robust scale.

  • tissue_mask (ndarray of bool) – Tissue mask.

  • signal_percentile (float) – Percentile threshold shown as dashed lines (default 50).

  • n_points (int) – Number of points to display (default 100000).

  • sample (str) – For title.

  • dye (str) – For title.

  • save_path (str or None) – If set, save to this path.

Returns:

fig

Return type:

matplotlib.Figure or None

exo2micro.legacy.plot_stretch_comparison(post_im, pre_im, scale_ls, scale_robust, title='', sample='', dye='')[source]

4-row x 2-column grid comparing stretch methods on LS and robust diffs.

Rows: asinh, signed log, global z-score, local z-score. Columns: LS (diverging cmap), Robust (inferno cmap).

Parameters:

  • post_im (ndarray)

  • pre_im (ndarray)

  • scale_ls (float)

  • scale_robust (float)

  • title (str)

  • sample (str)

  • dye (str)

Returns:

fig

Return type:

matplotlib.Figure

exo2micro.legacy.plot_zoom_region(post_im, pre_im, scale_ls, scale_robust, sig_mask=None, box_size=300, title='', sample='', dye='')[source]

Zoom into the densest signal region and compare LS vs robust subtraction.

Parameters:

  • post_im (ndarray)

  • pre_im (ndarray)

  • scale_ls (float)

  • scale_robust (float)

  • sig_mask (ndarray or None)

  • box_size (int)

  • title (str)

  • sample (str)

  • dye (str)

Return type:

fig, r0, c0

exo2micro.legacy.register_loworder(post_im, pre_im, stopit=500, stopdelta=1e-06, down_scale=0.5, save_prefix=None)[source]

Deprecated since version Use: register_highorder() from exo2micro.alignment instead, which handles both low-order and high-order registration.

Register using Euclidean (rotation + translation) motion model via ECC.

Parameters:

  • post_im (ndarray)

  • pre_im (ndarray)

  • stopit (int)

  • stopdelta (float)

  • down_scale (float)

  • save_prefix (str or None)

Returns:

post_full, pre_aligned

Return type:

ndarray

exo2micro.legacy.residuals_collapsed(scale, post_im, pre_im)[source]

Deprecated since version Use: optimize_subtraction() from exo2micro.scaling instead.

Subtract a scaled pre-stain image and return residuals as 1D array.

Parameters:

  • scale (float)

  • post_im (ndarray)

  • pre_im (ndarray)

Return type:

ndarray