File size: 4,599 Bytes
5ad11ec
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
import csv
import re
from typing import List
import skimage.io as skio
import numpy as np
import json
import pandas as pds

import matplotlib.colors as mcolors


base_colors = list(mcolors.BASE_COLORS)


def create_image_stack(
    image_format: str,
    fov: str,
    z: str,
    irs: List[str],
    wvs: List[str],
    out_file: str,
    coord_file: str,
):
    """Creates the images in an memory format for convenient data access in later reports

    Args:
        image_format (str): The file image pattern that will be used to construct the image stacks
        fov (str): The Fov of images to use for the stack
        z (str): The z of the images to use for the stack
        irs (List[str]): The list of imaging rounds for the stack
        wvs (List[str]): The list of channels for the stack
        out_file (str): The filename of the image stack produced
        coord_file (str): The filename of the coordinate file produced
    """
    # if any parameter is a single string, then make it into an iterable
    # list object

    if not isinstance(irs, list):
        irs = [irs]
    if not isinstance(wvs, list):
        wvs = [wvs]

    # Extract a test image to get size parameter out of
    test_img = skio.imread(image_format.format(wv=wvs[0], fov=fov, ir=irs[0], z=z))

    xyshape = test_img.shape
    x = xyshape[1]
    y = xyshape[0]

    # Load in all the data

    # Pre-allocate the image stack
    img_stack = np.zeros((y, x, len(wvs), len(irs),))
    for iir, ir in enumerate(irs):
        for iwv, wv in enumerate(wvs):

            img_stack[:, :, iwv, iir] = skio.imread(
                image_format.format(wv=wv, fov=fov, ir=ir, z=z)
            )
    # Drop it all into an npy file
    np.save(out_file, img_stack)

    # Save all the coordinates for the dimensions into a seperate file
    data = {"y": y, "x": x, "wvs": wvs, "irs": irs, "fovs": [fov], "zs": [z]}
    a_file = open(coord_file, "w")
    a_file = json.dump(data, a_file)


class Codebook:
    """The Codebook helper class

    Args:
        codebook_path (str): The path to the codebook file
    """
    def __init__(self, codebook_path: str):
        # TODO this could probably work well as a pandas df
        self.names = []
        self.ids = []
        self.barcode_strings = []
        with open(codebook_path, encoding="utf8") as csv_file:
            codebook_reader = csv.reader(csv_file)
            for i, row in enumerate(codebook_reader):
                if i == 0:
                    self.version = row[1]
                elif i == 1:
                    self.codebook_name = row[1]
                elif i == 2:
                    self.bit_names = row[1:]
                elif i >= 4:
                    self.names.append(row[0])
                    self.ids.append(row[1])
                    self.barcode_strings.append(row[2])
        self.barcode_arrays = [
            np.array([int(char) for char in re.sub(r"\s", "", barcode)], dtype="uint8")
            for barcode in self.barcode_strings
        ]

    def __len__(self):
        return len(self.names)

    def normalize_barcode(self, barcode_array):
        if np.sum(barcode_array) == 0:
            return barcode_array
        return barcode_array / np.sqrt(np.sum(barcode_array ** 2))

    def get_weighted_barcodes(self):
        magnitudes = [np.sqrt(sum(barcodes ** 2)) for barcodes in self.barcode_arrays]
        return [
            (self.barcode_arrays[i] / magnitudes[i]).astype("float16")
            for i in range(len(self.barcode_arrays))
        ]

    def get_single_bit_error_matrix(self, barcode_id):
        barcode_array = self.barcode_arrays[barcode_id]
        bit_error_matrix = [self.normalize_barcode(barcode_array)]
        for i in range(len(barcode_array)):
            corrected_barcode = barcode_array.copy()
            corrected_barcode[i] = np.logical_not(corrected_barcode[i])
            bit_error_matrix.append(self.normalize_barcode(corrected_barcode))
        return np.array(bit_error_matrix)


def read_table(file: str) -> pds.DataFrame:
    """Reads a file differently depending on its extension

    Args:
        file (str): The filename 

    Raises:
        ValueError: If the file extension is unrecognized

    Returns:
        pandas.DataFrame: _description_
    """
    ext = file.split(".")[-1]
    if ext == "csv":
        df = pds.read_csv(file)
    elif ext == "tsv":
        df = pds.read_csv(file, "\t")
    elif ext in {"xls", "xlsx", "xlsm", "xlsb"}:
        df = pds.read_excel(file)
    else:
        raise ValueError("Unexpected file extension")
    return df