-
Notifications
You must be signed in to change notification settings - Fork 1
/
ncov_WA.R
258 lines (204 loc) · 11.8 KB
/
ncov_WA.R
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
library(tidyverse)
library(stringr)
library(foreach)
library(doParallel)
source('utils_inference.R')
source('utils_cluster_alloc.R')
name_variant <- 'Alpha'
i_day <- 10 # Length of time window of analysis
## Input files
file_input_WA_analysis <- '../data/ncov_WA/input_WA_analysis_review.csv'
file_input_cluster_dist <- paste0('../data/ncov_WA/list_df_cluster_dist_by_period_', name_variant, '.rds')
file_proba_trans_before_mut <- '../results/proba_trans_before_mut/df_p_trans_before_mut_with_uncertainty.rds'
df_scenarios <- read_csv(file_input_WA_analysis)
if(! name_variant %in% df_scenarios$variant){
print('Need to pick a variant whose name is in the df_scenarios dataframe!')
}
## Time window of analysis
date_beginning_window <- as.Date(unlist(df_scenarios[df_scenarios$variant == name_variant, 'date_reached_10_seq']))
date_first_end_window <- as.Date(unlist(df_scenarios[df_scenarios$variant == name_variant, 'date_reached_10_seq']))
date_end_window <- date_first_end_window + 60
vec_date_end_window <- seq.Date(date_first_end_window, date_end_window, by = 'day')
## Probability that transmission occurs before mutation
p_trans_before_mut_pre_omicron <- readRDS(file_proba_trans_before_mut) %>%
ungroup() %>%
filter(pathogen == 'SARS-CoV-2 (pre-Omicron)') %>%
select(p_trans_before_mut) %>%
as.numeric()
p_trans_before_mut_post_omicron <- readRDS(file_proba_trans_before_mut) %>%
ungroup() %>%
filter(pathogen == 'SARS-CoV-2 (Omicron)') %>%
select(p_trans_before_mut) %>%
as.numeric()
## Datasets for the analysis
list_df_cluster_dist_by_period <- readRDS(file_input_cluster_dist)
vec_Nextstrain_clades_Omicron <- c("21K (Omicron)", "21L (Omicron)",
"21M (Omicron)",
"22A (Omicron)", "22B (Omicron)",
"22C (Omicron)", "22D (Omicron)",
"22E (Omicron)", "22F (Omicron)")
## Define function used to run analysis
run_analysis_given_cluster_ditrib <- function(list_df_cluster_distrib,
p_trans_before_mut_pre_omicron, p_trans_before_mut_post_omicron,
n_cores = 4){
## Scenarios for the proportion of cases detected
vec_p_detect_cases <- c(0.1, 0.2, 0.5, 0.8)
curr_prop_cases_sequenced <- list_df_cluster_distrib$n_seq_during_timewindow / list_df_cluster_distrib$n_cases_during_timewindow
vec_p_detect <- vec_p_detect_cases * curr_prop_cases_sequenced
## Dataset for the analysis
df_clusters_var_pre_omicron <- list_df_cluster_distrib$df_cluster_distrib %>%
filter(is_variant) %>%
filter(! Nextstrain_clade %in% vec_Nextstrain_clades_Omicron) %>%
group_by(n_in_same_clique) %>%
summarise(count = sum(n_clusters))
df_clusters_var_post_omicron <- list_df_cluster_distrib$df_cluster_distrib %>%
filter(is_variant) %>%
filter(Nextstrain_clade %in% vec_Nextstrain_clades_Omicron) %>%
group_by(n_in_same_clique) %>%
summarise(count = sum(n_clusters))
df_clusters_non_var_pre_omicron <- list_df_cluster_distrib$df_cluster_distrib %>%
filter(! is_variant) %>%
filter(! Nextstrain_clade %in% vec_Nextstrain_clades_Omicron) %>%
group_by(n_in_same_clique) %>%
summarise(count = sum(n_clusters))
df_clusters_non_var_post_omicron <- list_df_cluster_distrib$df_cluster_distrib %>%
filter(! is_variant) %>%
filter(Nextstrain_clade %in% vec_Nextstrain_clades_Omicron) %>%
group_by(n_in_same_clique) %>%
summarise(count = sum(n_clusters))
## Setting up the bounds for the optimization
R_min <- 0.01
R_max <- 5.0
k_min <- 0.001
k_max <- 10.0
max_cluster_size_inference <- 10000
cl <- makeForkCluster(n_cores)
registerDoParallel(cl)
df_inference <- Reduce('bind_rows', foreach(i_p_detect = 1:length(vec_p_detect_cases), .packages = c('dplyr')) %dopar% {
curr_p_detect <- vec_p_detect[i_p_detect]
## Define functions to be minimized
minus_log_lik_combined <- function(param){
R <- param[1]
k <- param[2]
if(nrow(df_clusters_non_var_pre_omicron) == 0){
curr_log_lik_non_var_pre_omicron <- 0.0
} else{
curr_log_lik_non_var_pre_omicron <- get_loglik_obs(vec_cluster_size = df_clusters_non_var_pre_omicron$n_in_same_clique,
vec_n_clusters = df_clusters_non_var_pre_omicron$count,
R = R, k = k,
p = p_trans_before_mut_pre_omicron, p_detect = curr_p_detect,
max_cluster_size = max_cluster_size_inference)
}
if(nrow(df_clusters_non_var_post_omicron) == 0){
curr_log_lik_non_var_post_omicron <- 0.0
} else{
curr_log_lik_non_var_post_omicron <- get_loglik_obs(vec_cluster_size = df_clusters_non_var_post_omicron$n_in_same_clique,
vec_n_clusters = df_clusters_non_var_post_omicron$count,
R = R, k = k,
p = p_trans_before_mut_post_omicron, p_detect = curr_p_detect,
max_cluster_size = max_cluster_size_inference)
}
if(nrow(df_clusters_var_pre_omicron) == 0){
curr_log_lik_var_pre_omicron <- 0.0
} else{
curr_log_lik_var_pre_omicron <- get_loglik_obs(vec_cluster_size = df_clusters_var_pre_omicron$n_in_same_clique,
vec_n_clusters = df_clusters_var_pre_omicron$count,
R = R, k = k,
p = p_trans_before_mut_pre_omicron, p_detect = curr_p_detect,
max_cluster_size = max_cluster_size_inference)
}
if(nrow(df_clusters_var_post_omicron) == 0){
curr_log_lik_var_post_omicron <- 0.0
} else{
curr_log_lik_var_post_omicron <- get_loglik_obs(vec_cluster_size = df_clusters_var_post_omicron$n_in_same_clique,
vec_n_clusters = df_clusters_var_post_omicron$count,
R = R, k = k,
p = p_trans_before_mut_post_omicron, p_detect = curr_p_detect,
max_cluster_size = max_cluster_size_inference)
}
return(- (curr_log_lik_non_var_pre_omicron + curr_log_lik_non_var_post_omicron +
curr_log_lik_var_pre_omicron + curr_log_lik_var_post_omicron))
}
minus_log_lik_split <- function(param){
R_var <- param[1]
R_non_var <- param[2]
k <- param[3]
if(nrow(df_clusters_non_var_pre_omicron) == 0){
curr_log_lik_non_var_pre_omicron <- 0.0
} else{
curr_log_lik_non_var_pre_omicron <- get_loglik_obs(vec_cluster_size = df_clusters_non_var_pre_omicron$n_in_same_clique,
vec_n_clusters = df_clusters_non_var_pre_omicron$count,
R = R_non_var, k = k,
p = p_trans_before_mut_pre_omicron, p_detect = curr_p_detect,
max_cluster_size = max_cluster_size_inference)
}
if(nrow(df_clusters_non_var_post_omicron) == 0){
curr_log_lik_non_var_post_omicron <- 0.0
} else{
curr_log_lik_non_var_post_omicron <- get_loglik_obs(vec_cluster_size = df_clusters_non_var_post_omicron$n_in_same_clique,
vec_n_clusters = df_clusters_non_var_post_omicron$count,
R = R_non_var, k = k,
p = p_trans_before_mut_post_omicron, p_detect = curr_p_detect,
max_cluster_size = max_cluster_size_inference)
}
if(nrow(df_clusters_var_pre_omicron) == 0){
curr_log_lik_var_pre_omicron <- 0.0
} else{
curr_log_lik_var_pre_omicron <- get_loglik_obs(vec_cluster_size = df_clusters_var_pre_omicron$n_in_same_clique,
vec_n_clusters = df_clusters_var_pre_omicron$count,
R = R_var, k = k,
p = p_trans_before_mut_pre_omicron, p_detect = curr_p_detect,
max_cluster_size = max_cluster_size_inference)
}
if(nrow(df_clusters_var_post_omicron) == 0){
curr_log_lik_var_post_omicron <- 0.0
} else{
curr_log_lik_var_post_omicron <- get_loglik_obs(vec_cluster_size = df_clusters_var_post_omicron$n_in_same_clique,
vec_n_clusters = df_clusters_var_post_omicron$count,
R = R_var, k = k,
p = p_trans_before_mut_post_omicron, p_detect = curr_p_detect,
max_cluster_size = max_cluster_size_inference)
}
return(- (curr_log_lik_non_var_pre_omicron + curr_log_lik_non_var_post_omicron +
curr_log_lik_var_pre_omicron + curr_log_lik_var_post_omicron))
}
mle_combined <- optim(par = c(1.0, 0.1),
fn = minus_log_lik_combined,
method = 'L-BFGS-B',
lower = c(R_min, k_min), upper = c(R_max, k_max))
mle_split <- optim(par = c(1.0, 1.0, 0.1),
fn = minus_log_lik_split,
method = 'L-BFGS-B',
lower = c(R_min, R_min, k_min), upper = c(R_max, R_max, k_max))
## Compute likelihood ratio test p-value for split vs combined
LRT_split_vs_combined <- as.numeric(2 * (mle_combined$value - mle_split$value))
p_val_split_vs_combined <- 1.0 - vec_p_quantile[which(vec_quantile_function_chisq > LRT_split_vs_combined)[1]]
c('max_log_lik_combined' = - mle_combined$value,
'max_log_lik_split' = - mle_split$value,
'R_MLE_combined' = mle_combined$par[1],
'k_MLE_combined' = mle_combined$par[2],
'R_var_MLE_split' = mle_split$par[1],
'R_non_var_MLE_split' = mle_split$par[2],
'k_MLE_split' = mle_split$par[3],
'LRT_split_vs_combined' = LRT_split_vs_combined,
'p_val_split_vs_combined' = p_val_split_vs_combined,
'CV_combined' = mle_combined$convergence,
'CV_split' = mle_split$convergence,
'p_detect_cases' = vec_p_detect_cases[i_p_detect]
)
})
stopCluster(cl)
return(df_inference)
}
## Setting up the inference
max_cluster_size_inference <- 10000
n_cores <- 4
vec_p_quantile <- c(0.0, 1e-15, 1e-14, 1e-13, 1e-12, 1e-11,
1e-10, 1e-9, 1e-8, 1e-7, 1e-6, 1e-5, seq(2e-5, 1.0, 1e-5))
vec_quantile_function_chisq <- qchisq(vec_p_quantile, df = 1)
curr_day <- vec_date_end_window[i_day]
list_df_cluster_distrib <- list_df_cluster_dist_by_period[[i_day]]
df_inference <- run_analysis_given_cluster_ditrib(list_df_cluster_distrib,
p_trans_before_mut_pre_omicron, p_trans_before_mut_post_omicron,
n_cores)
print(df_inference)