CladeOScope/utils.py at main · dst1/CladeOScope · GitHub

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
146
147
148
149
150
151
152
153
154
155
156
157
158
159
from bisect import bisect

import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
from sklearn.metrics import (average_precision_score, precision_recall_curve,
                             roc_auc_score, roc_curve)
from tqdm import tqdm


def argsort_mat(NPP_cor_mat):
    # prep argsort:
    adj_argsort = np.zeros((NPP_cor_mat.shape[0], NPP_cor_mat.shape[0]))
    for i in tqdm(range(NPP_cor_mat.shape[0])):
        inds = np.argsort(-NPP_cor_mat[i, :])
        adj_argsort[i, inds] = np.arange(NPP_cor_mat.shape[0])
    return adj_argsort


class PairsIndex:
    """
    Pairs index caches the indices for each genes and contains several utilities to match a pair with a sorted all pairs index
    """

    def __init__(self, genes, sort=True):
        """
        Initializes a new instance.
        An instance contains the list of genes and a mapping from gene to place in list

        :param genes: list, all genes in index
        :param sort: whether to use a sorted index
        """
        if sort == True:
            self.genes = sorted(genes)
        else:
            self.genes = genes

        self.ind = {k: i for i, k in enumerate(self.genes)}
        self.rev_ind = {i: k for i, k in enumerate(self.genes)}
        self.n = len(genes)
        self.ind_size = (self.n * (self.n - 1)) // 2
        self.i0_inds = [((2 * self.n - i_0 - 1) * i_0) //
                        2 for i_0 in range(self.n)]

    def get_ind_solo(self, genes):
        """
        Returns the place on the genes in the list

        Assumes genes are found on the list!!

        :param genes: genes to query
        :return: indexes for these genes
        """
        if type(genes) == str:
            return self.ind[genes]
        else:
            return [self.ind[x] for x in genes]

    def _calc_pair_loc(self, genes):
        """
        Internal function, gets an ordered tuple and calculates its location
        :param genes: list, gene[0]<gene[1], also not the same..
        :return: int, location in pairs list
        """

        i_0, i_1 = self.get_ind_solo(genes)
        ind = self._calc_pair_loc_from_inds(i_0, i_1)
        return ind

    def _calc_pair_loc_from_inds(self, x, y):
        """
        Internal function, gets an ordered tuple and calculates its location
        :param genes: list, gene[0]<gene[1], also not the same..
        :return: int, location in pairs list
        """

        i_0, i_1 = sorted([x, y])
        ind = ((2 * self.n - i_0 - 1) * i_0) // 2 + (i_1 - i_0 - 1)
        return ind

    def get_genes_solo(self, inds):
        """
        Returns the gene in place ind

        Assumes genes are found on the list!!

        :param genes: genes to query
        :return: indexes for these genes
        """
        if type(inds) == int:
            return self.rev_ind[inds]
        else:
            return [self.rev_ind[x] for x in inds]

    def _calc_pair_genes(self, ind, return_inds=False):
        """
        Internal function, gets a location and calulate the indexes of the genes
        returns the genes as a tuple
        :param ind: location in pairs list
        :return: tuple, genes
        """
        i_0 = bisect(self.i0_inds, ind) - 1
        i_1 = ind - self.i0_inds[i_0] + i_0 + 1

        if return_inds:
            return (i_0, i_1)
        return (self.get_genes_solo(i_0), self.get_genes_solo(i_1))


def plot_roc_single(y_test, probas, target, title="ROC", label=None,
                    ax=None, lw=2, figsize=(10, 10), max_fpr=1.0):

    if type(y_test) == pd.core.frame.DataFrame:
        y_test = y_test.values

    if ax is None:
        fig, ax = plt.subplots(1, 1, constrained_layout=True, figsize=figsize)

    fpr, tpr, _ = roc_curve(y_test, probas)
    roc_auc = roc_auc_score(y_test, probas, max_fpr=max_fpr)
    if label is None:
        ax.plot(fpr, tpr, lw=lw,
                label='{0} (AUC = {1:0.3f})'.format(target, roc_auc))
    else:
        ax.plot(fpr, tpr, lw=lw, label=label)
    ax.plot([0, 1], [0, 1], 'k--', lw=2)
    ax.set_xlim([0.0, max_fpr])
    ax.set_ylim([0.0, 1.05])
    ax.set_xlabel('False Positive Rate')
    ax.set_ylabel('True Positive Rate')
    ax.set_title(title)
    ax.legend()
    return ax


def plot_PR_single(y_test, probas, target, label=None,
                   title="Precision - Recall",
                   ax=None, lw=2, figsize=(10, 10)):

    if type(y_test) == pd.core.frame.DataFrame:
        y_test = y_test.values

    if ax is None:
        fig, ax = plt.subplots(1, 1, constrained_layout=True, figsize=figsize)

    prec, rec, _ = precision_recall_curve(y_test, probas)
    AP = average_precision_score(y_test, probas)
    if label is None:
        ax.plot(rec, prec, lw=lw,
                label='{0} (AP = {1:0.2f})'.format(target, AP))
    else:
        ax.plot(rec, prec, lw=lw, label=label)
    ax.set_xlabel('Recall')
    ax.set_ylabel('Precision')
    ax.set_ylim([0.0, 1.05])
    ax.set_xlim([0.0, 1.0])
    ax.set_title(title)
    ax.legend()
    return ax