forked from TRACE-LAC/covid19-waves-bogota
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathresults_genomics.py
203 lines (177 loc) · 10.3 KB
/
results_genomics.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
# -*- coding: utf-8 -*-
"""
Created on Fri Oct 21 2022
@author: davidsantiagoquevedo
@author: ntorresd
"""
import yaml
import sys
import pandas as pd
import datetime
from datetime import datetime
import matplotlib.pyplot as plt
import mpl_axes_aligner as mpla
import seaborn as sns
from met_brewer import met_brew
config = yaml.load(open("config.yml", "r"))["default"]
UPDATE = config['UPDATE_DATES']['CONFIRMED_CASES']
DATA_PATH = config['PATHS']['DATA_PATH']
OUT_PATH = config['PATHS']['OUT_PATH'].format(dir = 'genomics')
DATE_GENOMICS = config['UPDATE_DATES']['GENOMICS']
UTILS_PATH = config['PATHS']['UTILS_PATH'].format(dir = 'severe_outcomes')
plt.style.use(config['PATHS']['PLOT_STYLE'])
prop_cycle = plt.rcParams['axes.prop_cycle']
colors = prop_cycle.by_key()['color']
# Import useful functions
sys.path.append(UTILS_PATH)
import utilities_severity as ut
# Load data
## Variants
df_variants = pd.read_csv(DATA_PATH + 'variants-ic-bog_'+ DATE_GENOMICS + '.csv')
df_variants['date'] = pd.to_datetime(df_variants['date'])
## Results from multinomial analysis
df_results = pd.read_csv(OUT_PATH + 'theta.csv')
df_pivot = pd.read_csv(DATA_PATH + 'variants_pivot.csv').rename(columns = {'t' : 'week', 'week': 'week_name'})
df_results = df_results.merge(df_pivot[['week', 'week_name']], on = 'week')
df_results['week_date'] = df_results.week_name.apply(lambda date: datetime.strptime(date + '-1', "%Y-%W-%w"))
df_results = df_results.sort_values(by = 'week_date')
## Cases
df_confirmed_bogota = pd.read_csv(DATA_PATH + 'confirmed_cases_' + UPDATE + '.csv')
df_confirmed_bogota = df_confirmed_bogota.astype({'age':int})
df_confirmed_bogota['onset'] = pd.to_datetime(df_confirmed_bogota['onset'], errors='coerce')
df_confirmed_bogota['death'] = pd.to_datetime(df_confirmed_bogota['death'], errors='coerce')
n_ticks = 10
def plot_multinomial(ax, limits):
# Weekly counts - Incidence
date1 = datetime.strptime(limits[0] + '-1', '%Y-%W-%u')
date2 = datetime.strptime(limits[1] + '-1', '%Y-%W-%u')
df_incidence = df_confirmed_bogota[df_confirmed_bogota['onset'].between(date1,date2)]
df_incidence['week'] = df_confirmed_bogota['onset'].apply(lambda date: date.strftime('%Y-%W'))
df_incidence_count = ut.counts(df_incidence, var='week',columns=['week', 'cases'])
df_incidence_count['week_date'] = df_incidence_count['week'].apply(lambda date: datetime.strptime(date + '-1', '%Y-%W-%u'))
df_incidence_count = df_incidence_count.sort_values(by = 'week_date')
l1 = ax.bar(df_incidence_count['week_date'], df_incidence_count['cases'],
alpha = 0.3, color = "grey", width=5.3, label = 'Casos nuevos semanales')
ax.set_ylabel('Incidencia')
ax.tick_params(axis='x', rotation=90)
ax.xaxis.set_major_locator(plt.MaxNLocator(n_ticks))
labs = ['Casos nuevos semanales']
# Prevalence - Observed
ax1 = ax.twinx()
stat = "mean"
n = 0
variant = 'Alpha'
mask = (df_results.stat == stat) & (df_results.variant == variant)
mask1 = (df_results.stat == 'q025') & (df_results.variant == variant)
mask2 = (df_results.stat == 'q975') & (df_results.variant == variant)
ax1.plot(df_results[mask].week_date, df_results[mask].theta, color = colors[n], label = variant, marker = '')
ax1.fill_between(df_results[mask].week_date, df_results[mask1].theta, df_results[mask2].theta,
color = colors[n], alpha = 0.2)
labs.append(variant)
n = 1
variant = 'Delta'
mask = (df_results.stat == stat) & (df_results.variant == variant)
mask1 = (df_results.stat == 'q025') & (df_results.variant == variant)
mask2 = (df_results.stat == 'q975') & (df_results.variant == variant)
ax1.plot(df_results[mask].week_date, df_results[mask].theta, color = colors[n], label = variant, marker = '')
ax1.fill_between(df_results[mask].week_date, df_results[mask1].theta, df_results[mask2].theta,
color = colors[n], alpha = 0.2)
labs.append(variant)
n = 2
variant = 'Gamma'
mask = (df_results.stat == stat) & (df_results.variant == variant)
mask1 = (df_results.stat == 'q025') & (df_results.variant == variant)
mask2 = (df_results.stat == 'q975') & (df_results.variant == variant)
ax1.plot(df_results[mask].week_date, df_results[mask].theta, color = colors[n], label = variant, marker = '')
ax1.fill_between(df_results[mask].week_date, df_results[mask1].theta, df_results[mask2].theta,
color = colors[n], alpha = 0.2)
labs.append(variant)
n = 3
variant = 'Mu'
mask = (df_results.stat == stat) & (df_results.variant == variant)
mask1 = (df_results.stat == 'q025') & (df_results.variant == variant)
mask2 = (df_results.stat == 'q975') & (df_results.variant == variant)
ax1.plot(df_results[mask].week_date, df_results[mask].theta, color = colors[n], label = variant, marker = '')
ax1.fill_between(df_results[mask].week_date, df_results[mask1].theta, df_results[mask2].theta,
color = colors[n], alpha = 0.2)
labs.append(variant)
n = 4
variant = 'Omicron'
mask = (df_results.stat == stat) & (df_results.variant == variant)
mask1 = (df_results.stat == 'q025') & (df_results.variant == variant)
mask2 = (df_results.stat == 'q975') & (df_results.variant == variant)
ax1.plot(df_results[mask].week_date, df_results[mask].theta, color = colors[n], label = variant, marker = '')
ax1.fill_between(df_results[mask].week_date, df_results[mask1].theta, df_results[mask2].theta,
color = colors[n], alpha = 0.2)
labs.append(variant)
# Multinomial model
n = 0
variant = 'Alpha'
mask = (df_results.stat == stat) & (df_results.variant == variant)
mask1 = (df_results.stat == 'q025') & (df_results.variant == variant)
mask2 = (df_results.stat == 'q975') & (df_results.variant == variant)
ax1.plot(df_results[mask].week_date, df_results[mask][variant]/df_results[mask].weekly_count_variants,
color = colors[n], marker = '^', linestyle = '')
n = 1
variant = 'Delta'
mask = (df_results.stat == stat) & (df_results.variant == variant)
mask1 = (df_results.stat == 'q025') & (df_results.variant == variant)
mask2 = (df_results.stat == 'q975') & (df_results.variant == variant)
ax1.plot(df_results[mask].week_date, df_results[mask][variant]/df_results[mask].weekly_count_variants,
color = colors[n], marker = '^', linestyle = '')
n = 2
variant = 'Gamma'
mask = (df_results.stat == stat) & (df_results.variant == variant)
mask1 = (df_results.stat == 'q025') & (df_results.variant == variant)
mask2 = (df_results.stat == 'q975') & (df_results.variant == variant)
l4 = ax1.plot(df_results[mask].week_date, df_results[mask].theta, color = colors[n], label = variant)
ax1.plot(df_results[mask].week_date, df_results[mask][variant]/df_results[mask].weekly_count_variants,
color = colors[n], marker = '^', linestyle = '')
n = 3
variant = 'Mu'
mask = (df_results.stat == stat) & (df_results.variant == variant)
mask1 = (df_results.stat == 'q025') & (df_results.variant == variant)
mask2 = (df_results.stat == 'q975') & (df_results.variant == variant)
l5 = ax1.plot(df_results[mask].week_date, df_results[mask].theta, color = colors[n], label = variant)
ax1.plot(df_results[mask].week_date, df_results[mask][variant]/df_results[mask].weekly_count_variants,
color = colors[n], marker = '^', linestyle = '')
n = 4
variant = 'Omicron'
mask = (df_results.stat == stat) & (df_results.variant == variant)
mask1 = (df_results.stat == 'q025') & (df_results.variant == variant)
mask2 = (df_results.stat == 'q975') & (df_results.variant == variant)
l6 = ax1.plot(df_results[mask].week_date, df_results[mask].theta, color = colors[n], label = variant)
ax1.plot(df_results[mask].week_date, df_results[mask][variant]/df_results[mask].weekly_count_variants,
color = colors[n], marker = '^', linestyle = '')
ax1.tick_params(axis='x', rotation=90)
ax1.xaxis.set_major_locator(plt.MaxNLocator(n_ticks))
ax1.spines.right.set_visible(True)
ax1.set_ylabel('Prevalencia', rotation = 270, labelpad = 15)
handles = ax.containers + ax1.get_lines()
labels1, handles1 = ax1.get_legend_handles_labels()
ax1.legend(handles, labs, loc='upper right', frameon=False, fontsize=12, ncol = 2)
mpla.align.yaxes(ax, 0, ax1, 0, 0.03)
# Prevalence histogram
def plot_prevalence(ax):
agg_df = df_variants.pivot_table(index = 'date',
columns = 'lineage',
values = 'PointEst',
aggfunc = 'max').reset_index()
initial_date = min(df_variants['date'])
final_date = max(df_variants['date'])
bins = len(df_variants['date'].unique())
variants_hist = sns.histplot(data = df_variants, ax=ax,
multiple = 'stack',
weights = 'PointEst', bins = bins, x = 'date', hue = 'lineage',
legend = True)
ax.set_xticks(pd.date_range(start = min(df_variants['date']), end = final_date, freq = "M"))
ax.set_xlim(left = min(df_variants['date']), right = final_date)
sns.move_legend(variants_hist, 'lower right', bbox_to_anchor = (0.92, 0.95), ncol = 3, title=None)
# Advantage heatmap
def plot_heatmap(ax, n):
df_mean = pd.read_csv(OUT_PATH + 'advantage_mean.csv')
df_mean = df_mean.set_index('pivot_variant')
color_list = met_brew('Cassatt1', n = n, brew_type='continuous')
sns.heatmap(data = df_mean, annot = df_mean, fmt = '.2f', cmap = color_list, ax = ax)
ax.set_ylabel('')
ax.tick_params(axis='x', rotation=90)