-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathplot_CAMI2_benchmark.py
259 lines (222 loc) · 12 KB
/
plot_CAMI2_benchmark.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
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
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
import pandas as pd
import os
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
plt.rcParams['pdf.fonttype'] = 42
plt.rcParams['ps.fonttype'] = 42
skin_index = [1, 13, 14, 15, 16, 17, 18, 19, 20, 28]
oral_index = [6,7,8,13,14,15,16,17,18,19]
airways_index = [4,7,8,9,10,11,12,23,26,27]
gastrointestinal_index = [0, 1, 2, 3, 4, 5, 9, 10, 11, 12]
urogenital_index= [0, 2, 3, 5, 6, 21, 22, 24, 25]
def plot_short_reads():
genus_VAMB_list = []
species_VAMB_list = []
strain_VAMB_list = []
for env in ['Airways','Gastrointestinal','Oral','Skin','Urogenital']:
print(env)
data_index = None
if env == 'Skin':
data_index = skin_index
if env == 'Oral':
data_index = oral_index
if env == 'Airways':
data_index = airways_index
if env == 'Gastrointestinal':
data_index = gastrointestinal_index
if env == 'Urogenital':
data_index = urogenital_index
genome_path = os.path.join('data/short_reads/{0}/amber_{1}'.format(env, data_index[0]), 'genome')
method_list = {}
species_list = {}
genus_list = {}
for root, dirs, files in os.walk(genome_path, topdown=False):
for name in dirs:
method_list[name] = []
species_list[name] = []
genus_list[name] = []
for temp in data_index:
taxi = pd.read_csv('data/short_reads/{0}/taxonomic_profile.txt'.format(env), sep='\t', skiprows=3, dtype={'@@TAXID': str, 'TAXPATH': str})
taxi_genus = taxi[taxi['RANK'] == 'genus']['@@TAXID'].values.tolist()
taxi_species = taxi[taxi['RANK'] == 'species'][
'@@TAXID'].values.tolist()
method_path_list = []
genome_path = os.path.join('data/short_reads/{0}/amber_{1}'.format(env, temp),'genome')
for root, dirs, files in os.walk(genome_path, topdown=False):
for name in dirs:
method_path_list.append(os.path.join(root, name))
for method_path in method_path_list:
metric = pd.read_csv(os.path.join(method_path, 'metrics_per_bin.tsv'), sep='\t')
com_90_pur_95 = metric[
(metric['Completeness (bp)'].astype(float) > float(0.9)) & (
metric['Purity (bp)'].astype(float) >= float(0.95))]
strain_list = com_90_pur_95['Most abundant genome'].values.tolist()
method_list[method_path.split('/')[-1]].extend(strain_list)
for temp_strain in strain_list:
if temp_strain in taxi['_CAMI_GENOMEID'].values.tolist():
taxi_split = taxi[taxi['_CAMI_GENOMEID'] == temp_strain]['TAXPATH'].values[0].split('|')
if taxi_split[-2] in taxi_species:
species_list[method_path.split('/')[-1]].append(taxi_split[-2])
if taxi_split[-3] in taxi_genus:
genus_list[method_path.split('/')[-1]].append(taxi_split[-3])
result = {'VAMB': {'strain': list(set(method_list['VAMB'])),
'species': list(set(species_list['VAMB'])),
'genus': list(set(genus_list['VAMB']))},
'SemiBin': {'strain': list(set(method_list['SemiBin'])),
'species': list(set(species_list['SemiBin'])),
'genus': list(set(genus_list['SemiBin']))},
'SemiBin2': {'strain': list(set(method_list['SemiBin2'])),
'species': list(set(species_list['SemiBin2'])),
'genus': list(set(genus_list['SemiBin2']))}}
strain_VAMB = len(result['VAMB']['strain'])
species_VAMB = len(result['VAMB']['species'])
genus_VAMB = len(result['VAMB']['genus'])
strain_SemiBin = len(result['SemiBin']['strain'])
species_SemiBin = len(result['SemiBin']['species'])
genus_SemiBin = len(result['SemiBin']['genus'])
strain_SemiBin2 = len(result['SemiBin2']['strain'])
species_SemiBin2 = len(result['SemiBin2']['species'])
genus_SemiBin2 = len(result['SemiBin2']['genus'])
print(strain_VAMB, species_VAMB, genus_VAMB)
print(strain_SemiBin, species_SemiBin, genus_SemiBin)
print(strain_SemiBin2, species_SemiBin2, genus_SemiBin2)
print('strain_improvement(SemiBin, VAMB): ', (strain_SemiBin2 - strain_SemiBin) / strain_SemiBin, (strain_SemiBin2 - strain_VAMB) / strain_VAMB)
print('species_improvement(SemiBin, VAMB): ', (species_SemiBin2 - species_SemiBin) / species_SemiBin, (species_SemiBin2 - species_VAMB) / species_VAMB)
print('genus_improvement(SemiBin, VAMB): ', (genus_SemiBin2 - genus_SemiBin) / genus_SemiBin, (genus_SemiBin2 - genus_VAMB) / genus_VAMB)
genus_VAMB_list.append((genus_SemiBin2 - genus_VAMB) / genus_VAMB)
species_VAMB_list.append((species_SemiBin2 - species_VAMB) / species_VAMB)
strain_VAMB_list.append((strain_SemiBin2 - strain_VAMB) / strain_VAMB)
line_width = 1
plt.figure(figsize=(4, 4))
plt.plot(['genus', 'species', 'strain'],
[genus_VAMB, species_VAMB, strain_VAMB], label='VAMB', color='#ec7014',
linewidth=line_width, marker='o', )
plt.plot(['genus', 'species', 'strain'],
[genus_SemiBin, species_SemiBin, strain_SemiBin], label='SemiBin', color='#7570b3',
linewidth=line_width, marker='o', )
plt.plot(['genus', 'species', 'strain'],
[genus_SemiBin2, species_SemiBin2,
strain_SemiBin2], label='SemiBin2',
color='#1b9e77', linewidth=line_width, marker='o', )
plt.legend()
plt.title(f"{env}", fontsize=20, alpha=1.0, color='black')
plt.savefig(f'cami2_short_reads_{env}.pdf', dpi=300, bbox_inches='tight')
plt.close()
print('VAMB genus average:', np.mean(genus_VAMB_list))
print('VAMB species average:', np.mean(species_VAMB_list))
print('VAMB strain average:', np.mean(strain_VAMB_list))
def get_number_of_genomes_per_completeness(amber_path, return_pandas=False):
genome_path = os.path.join(amber_path, 'genome')
table = {}
for root, dirs, files in os.walk(genome_path, topdown=False):
for name in dirs:
method_path = os.path.join(root, name)
metric = pd.read_csv(os.path.join(method_path, 'metrics_per_bin.tsv'), sep='\t')
metric = metric.query('`Purity (bp)` >= 0.95')
com_90_pur_95 = metric.eval('`Completeness (bp)` > 0.9').sum()
com_80_pur_95 = metric.eval('`Completeness (bp)` > 0.8').sum() - (
com_90_pur_95)
com_70_pur_95 = metric.eval('`Completeness (bp)` > 0.7').sum() - (
com_90_pur_95 + com_80_pur_95)
com_60_pur_95 = metric.eval('`Completeness (bp)` > 0.6').sum() - (
com_90_pur_95 + com_80_pur_95 + com_70_pur_95)
com_50_pur_95 = metric.eval('`Completeness (bp)` > 0.5').sum() - (
com_90_pur_95 + com_80_pur_95 + com_70_pur_95 + com_60_pur_95)
table[method_path.split('/')[-1]]= [com_90_pur_95, com_80_pur_95, com_70_pur_95, com_60_pur_95, com_50_pur_95]
if return_pandas:
return pd.DataFrame(table, index=[90,80,70,60,50]).T
return table
color_map = ['#01665e', '#5da8a1', '#80cdc1', '#c7eae5']
def plot_long_reads():
for env in ['Skin','Oral','Airways','Gastrointestinal','Urogenital']:
print(env)
data = get_number_of_genomes_per_completeness(f'data/long_reads/amber_{env}', return_pandas=True)
subset = data.loc[[
'VAMB',
'Metadecoder',
'SemiBin',
'Metabat2',
'SemiBin2']]
subset = subset[[90, 80, 70, 60]]
print(subset)
high_quality_list = subset[90].sort_values().values
print('Improvement of best binner over second best: {:.2%}'.format(
(high_quality_list[-1] - high_quality_list[-2]) / high_quality_list[-2]))
ax = subset.plot(kind="barh", stacked=True, legend=False,
color=color_map)
ax.legend(['>90', '>80', '>70', '>60'],
loc='lower right', fontsize=10, title='completeness')
# ax.set_xticks(ticks=y_label)
# ax.set_xticklabels(labels= fontsize=15, color='black')
plt.xticks(fontsize=15, color='black')
ax.set_yticklabels(labels=subset.index, fontsize=15, color='black')
ax.set_xlabel('Bins(< 5% contamination)', fontsize=15, color='black')
ax.set_title(f'{env}', fontsize=15, alpha=1.0)
plt.tight_layout()
plt.savefig(f'cami2_long_reads_{env}.pdf', dpi=300)
plt.close()
plt.show()
def plot_long_reads_update():
for env in ['Skin','Oral','Airways','Gastrointestinal','Urogenital']:
print(env)
data = get_number_of_genomes_per_completeness(f'data/long_reads_update/amber_{env}', return_pandas=True)
subset = data.loc[[
'CONCOCT',
'VAMB',
'Metadecoder',
'MetaBinner',
'SemiBin',
'Metabat2',
'SemiBin2']]
subset = subset[[90, 80, 70, 60]]
print(subset)
high_quality_list = subset[90].sort_values().values
print('Improvement of best binner over second best: {:.2%}'.format(
(high_quality_list[-1] - high_quality_list[-2]) / high_quality_list[-2]))
ax = subset.plot(kind="barh", stacked=True, legend=False,
color=color_map)
ax.legend(['>90', '>80', '>70', '>60'],
loc='lower right', fontsize=10, title='completeness')
# ax.set_xticks(ticks=y_label)
# ax.set_xticklabels(labels= fontsize=15, color='black')
plt.xticks(fontsize=15, color='black')
ax.set_yticklabels(labels=subset.index, fontsize=15, color='black')
ax.set_xlabel('Bins(< 5% contamination)', fontsize=15, color='black')
ax.set_title(f'{env}', fontsize=15, alpha=1.0)
plt.tight_layout()
plt.savefig(f'cami2_long_reads_{env}_update.pdf', dpi=300)
plt.close()
plt.show()
def plot_long_reads_training_compare():
result = {}
for env in ['Skin','Oral','Airways','Gastrointestinal','Urogenital']:
print(env)
data = get_number_of_genomes_per_completeness(f'data/long_reads/amber_{env}', return_pandas=True)
subset = data.loc[[
'SemiBin2(MMseqs2)',
'SemiBin2']]
subset = subset[[90, 80, 70, 60]].values
print(subset)
result[env] = {}
result[env]['SemiBin2(MMseqs2)'] = subset[0][0]
result[env]['SemiBin2'] = subset[1][0]
subset = pd.DataFrame(np.array([[result['Airways']['SemiBin2(MMseqs2)'], result['Airways']['SemiBin2']],
[result['Gastrointestinal']['SemiBin2(MMseqs2)'], result['Gastrointestinal']['SemiBin2']],
[result['Oral']['SemiBin2(MMseqs2)'], result['Oral']['SemiBin2']],
[result['Skin']['SemiBin2(MMseqs2)'], result['Skin']['SemiBin2']],
[result['Urogenital']['SemiBin2(MMseqs2)'], result['Urogenital']['SemiBin2']]]),
columns=['SemiBin2(MMseqs2)', 'SemiBin2'],
index=['Airways', 'Gastrointestinal','Oral','Skin','Urogenital'])
print(subset)
ax = subset.plot(kind='bar', width=0.6,color=['#7570b3','#1b9e77'])
ax.set_xticklabels(labels=['Airways', 'Gastrointestinal','Oral','Skin','Urogenital'], fontsize=15, color='black',rotation = 45)
ax.set_ylabel('High quality strains', fontsize=15, color='black')
ax.set_title('strain', fontsize=20, alpha=1.0, color='black')
plt.savefig('cami2_long_reads_training_compare.pdf', dpi=300,bbox_inches='tight')
plt.close()
if __name__ == '__main__':
# plot_long_reads_training_compare()
# plot_long_reads()
plot_short_reads()
# plot_long_reads_update()