-
Notifications
You must be signed in to change notification settings - Fork 0
/
Accuracy_scores.py
132 lines (103 loc) · 4.64 KB
/
Accuracy_scores.py
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
def check_overlap(range1_start, range1_end, range2_start, range2_end):
overlap = max(0, min(range1_end, range2_end) - max(range1_start, range2_start) + 1)
return overlap
def nucleotide_metrics(start_pos_pred, end_pos_pred, start_pos_true, end_pos_true, seq_len):
true_positives = check_overlap(start_pos_pred, end_pos_pred, start_pos_true, end_pos_true)
# true_negatives = check_overlap(0, start_pos_pred - 1, 0, start_pos_true - 1) + \
# check_overlap(0, start_pos_pred - 1, end_pos_true + 1, seq_len) + \
# check_overlap(end_pos_pred + 1, seq_len, 0, start_pos_true - 1) + \
# check_overlap(end_pos_pred + 1, seq_len, end_pos_true + 1, seq_len)
false_negatives = check_overlap(0, start_pos_pred - 1, start_pos_true, end_pos_true) + \
check_overlap(end_pos_pred + 1, seq_len, start_pos_true, end_pos_true)
false_positives = check_overlap(start_pos_pred, end_pos_pred, 0, start_pos_true - 1) + \
check_overlap(start_pos_pred, end_pos_pred, end_pos_pred + 1, seq_len)
try:
performance_coefficient = true_positives / (true_positives + false_positives + false_negatives)
except ZeroDivisionError:
performance_coefficient = 0
try:
sensitivity = true_positives / (true_positives + false_negatives)
except ZeroDivisionError:
sensitivity = 0
try:
specificity = true_positives / (true_positives + false_positives)
except ZeroDivisionError:
specificity = 0
return performance_coefficient, sensitivity, specificity
def nucleotide_accuracy(pred_df, mode='nPC', score_group=None):
if score_group == None:
pred_df = pred_df.copy()
else:
mask = pred_df['Subgroup'].str.contains('_' + str(score_group) + '_')
pred_df = pred_df[mask].copy()
pred_df.reset_index(inplace=True, drop=True)
# print(pred_df.Subgroup)
total_sites_num = pred_df.shape[0]
if mode in ['nSp', 'nSn', 'nPC']:
sums = int(pred_df[mode].sum(axis=0))
else:
sums = int(pred_df['nPC'].sum(axis=0))
accuracy = sums / total_sites_num
return accuracy
def site_accuracy(pred_df, threshold=1, mode='sPC', score_group=None):
if score_group == None:
pred_df = pred_df.copy()
else:
mask = pred_df['Subgroup'].str.contains('_' + str(score_group) + '_')
pred_df = pred_df[mask].copy()
pred_df.reset_index(inplace=True, drop=True)
true_positives = 0
false_positives = 0
target_sequence_overlap = {}
for idx, row in pred_df.iterrows():
start_pred = row['Predicted_Start_pos']
stop_pred = row['Predicted_Stop_pos']
start_true = row['Target_Start_pos']
stop_true = row['Target_Stop_pos']
pred_width = stop_pred-start_pred+1
seq_id = row['Subgroup'].split('_')[2]
overlap = check_overlap(start_pred, stop_pred, start_true, stop_true)
# if pred_width - overlap <= threshold:
if overlap >= threshold:
true_positives += 1
if seq_id in list(target_sequence_overlap.keys()):
target_sequence_overlap[seq_id] += 1
else:
target_sequence_overlap[seq_id] = 1
else:
false_positives += 1
if seq_id not in list(target_sequence_overlap.keys()):
target_sequence_overlap[seq_id] = 0
false_negatives = list(target_sequence_overlap.values()).count(0)
try:
sPC = true_positives / (true_positives + false_positives + false_negatives)
except ZeroDivisionError:
sPC = 0
try:
sSn = true_positives / (true_positives + false_negatives)
except ZeroDivisionError:
sSn = 0
try:
sSp = true_positives / (true_positives + false_positives)
except ZeroDivisionError:
sSp = 0
total_motif_groups_num = len(set(pred_df['Subgroup'].str.split('_')[0]))
if mode == 'sSp':
accuracy = sSp / total_motif_groups_num
elif mode == 'sSn':
accuracy = sSn / total_motif_groups_num
else:
accuracy = sPC / total_motif_groups_num
return accuracy
#### Spyros ####
def nucleotide_acc_weights(pred_df, mode='nPC', score_group=None):
if score_group is not None:
pred_df = pred_df[pred_df['Subgroup'].str.contains('_' + str(score_group) + '_')].copy()
pred_df.reset_index(drop=True, inplace=True)
total_sites_num = pred_df['Weights'].sum()
if mode in ['nSp', 'nSn', 'nPC']:
sums = (pred_df[mode] * pred_df['Weights']).sum()
else:
sums = (pred_df['nPC'] * pred_df['Weights']).sum()
accuracy = sums / total_sites_num
return accuracy