mrbles.report module

Quality Control Report Classes and Functions

This file stores the quality control report classes and functions for the MRBLEs Analysis module.

class mrbles.report.BeadsReport(data, images, masks, assay_channel, codes=None, files=None, sort=True)

Bases: future.types.newobject.newobject

Per-MRBLE images report.

This method generates the selected images per-MRBLE.

This method can take a lot of time, since it will generate images per-MRBLE. It takes about 5 minutes per 1,000 beads, for 12 images each, which makes a total of 11,000 images.

Parameters:

• data (Pandas DataFrame) – Contains all the dimension, posotional, and intensity data per-MBRLE.
• images (mrbles.ImageDataFrame, Xarray DataArray) – Contains images.
• masks (Xarray DataArray) – Contains masks.
• assay_channel (str) – Assay channel name, e.g. ‘Cy5_FF’
• codes (int, list of int) – Integer or list of integers with selected codes. Defaults to None.
• files (int, list of int) – Integer or list of integers with selected files. Defaults to None.
• sort (boolean) – Sort by code. Defaults to True.
• Methods
• ——–
• generate() (method)

Variables:

• time_sec (float) – Time required per-image generated in seconds. For instance, 300 beads times 12 images is 3600 images. Defaults to 0.0275
• parallelize (boolean) – Wether to use parallelization. Can be slowing down on low-power computers. Defaults to True.

generate(filename)

Per-MRBLE images report.

This method generates the selected images per-MRBLE.

This method can take a lot of time, since it will generate images per-MRBLE. It takes about 5 minutes per 1,000 beads, for 12 images each, which makes a total of 12,000 images.

class mrbles.report.ClusterCheck(decode_object, *args, **kwargs)

Bases: mrbles.data.TableDataFrame

Cluster check reporting.

plot_2D(colors, ci_trace=None, min_prob=None)

Plot 2D clusters.

plot_3D(min_prob=None)

Plot ratio clusters in 3D.

Parameters:min_prob (float) – Set minimal probability level.

static scatter(data, target, codes=None, title=None, axes_names=None)

Plot clusters in scatter plot 2D or 3D.

class mrbles.report.GenerateCodes(colors, s0s, slopes, nsigma)

Bases: future.types.newobject.newobject

Generate bead code set.

Parameters:

• colors (list of str) – List of coding colors in a list of strings.
• s0s (liat of float) – List of standard deviations (SD) at intensity 0 for each encoding color.
• slopes (float) – List of slopes of the SDs versus intensity for each encoding color.
• nsigma (float) – The number of SD to separate coding levels.

Examples

>>> code_set_gen = GenerateCodes(['Dy', 'Sm', 'Tm'],
                                 [0.0039, 0.0055, 0.0029],
                                 [0.022, 0.016, 0.049], 6.4)
>>> code_set_gen.result
           Dy        Sm        Tm
0    0.000000  0.000000  0.000000
>>> code_set_gen = GenerateCodes(['Dy', 'Sm'],
                                 [0.0039, 0.0055],
                                 [0.022, 0.016], 8.4)
>>> code_set_gen.generate()
Number of codes:  24
>>> code_set_gen.result
          Dy        Sm
0   0.000000  0.000000
1   0.000000  0.106747
2   0.000000  0.246642
......................
>>> code_set_gen.iterate(28)
....................
Number of codes:  26
Number of codes:  26
Number of codes:  28
Final nsigma:  8.09
Iterations  :  31

axis

Return number of axis (colors).

colors

Return color names.

generate(nsigma=None, depends=None)

Generate codes with default nsigma or given nsigma.

Parameters:

• nsigma (float, optional) – The number of SD to separate coding levels. Defaults to initial nsigma.
• depends (any, experimental, optional) – Used for Tm (3rd in array) dependence on Dy (1st array).

static get_levels(std0, slope, nsigma)

Predict the number of levels of a coding color.

The coding levels are based on s0, the standard deviation (SD) at intensity 0, and the slope between intensity and SD.

Parameters:

• std0 (float) – The SD at intensity 0.
• slope (float) – The slope of the SD versus intensity.
• nsigma (float) – The number of SD to separate levels.

Returns:

levels – Returns list of codes values for a given coding color.

Return type:

list of float

iterate(num, nsimga_start=None, nsigma_step=0.01, max_iter=1000)

Iterate nsigma until number of codes are found.

Does not work with dependence, such as Tm dependence on Dy.

Parameters:

• num (int) – Number of required codes.
• nsigma_start (float, optional) – Start value of nsigma. Defaults to initial nsigma.
• nsigma_step (float, optional) – Iterarion step. Defaults to 0.01
• max_iter (int, optional) – Maximum iteration steps. Defaults to 1000.

levels

Return number of levels.

static recursive_looper(iterators, pos=0)

Recursive looper.

Implements the same functionality as nested for loops, but is more dynamic. Iterators can either be a list of methods which return iterables, a list of iterables, or a combination of both.

result

Return resulting ratios.

to_csv(filename)

Export to CSV.

to_csv_rep(filename, repeats, labels=None, pos=True)

Export to CSV, with repeated ratios for bead synthesis.

class mrbles.report.PeptideScramble(seq)

Bases: future.types.newobject.newobject

Randomizes amino acid sequence

Parameters:seq (string) – Inset amino acid sequence as a string. Returns:seq – Returns string of shuffled amino acid sequence. Return type:string

random(seq=None)

Return randomized amino acid sequence.

class mrbles.report.QCReport(data)

Bases: future.types.newobject.newobject

MRBLE library Quality Control report.

Parameters:data (Pandas DataFrame) – Per beads data.

assay_contamination(pdf_object)

bead_size()

Bead size distribution plot.

beads_per_code()

Beads per-code distribution plot.

code_stability()

generate(filename, savefig=False)

Generate QC report.

Parameters:

• filename (str) – Filename to save QC PDF report to.
• savefig (boolean) – Save figures separately to ‘report’ folder. Defaults to False.

html_report()

Generate HTML report.

npl_covar_plots()

npl_plots(pdf_object)

mrbles.report.circle_overlay(image, dims=None, ring=None)

Circle Overlay Image.

Overlay image with circles of labeled mask.

mrbles.report.image_overlay(image, image_blend, alpha=0.2, cmap_image='Greys_r', cmap_blend='jet')

Overlay 2 Images. Overlay of 2 images using alpha blend.