In avallecam/incidenceflow: What the Package Does (One Line, Title Case)

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

library(incidenceflow)

library(tidyverse)
library(readxl)
library(lubridate)
library(aweek)
library(janitor)
# library(avallecam)
library(compareGroups)
library(incidence)
library(patchwork)
# library(avallecam)
library(EpiEstim)
library(colorspace)
theme_set(theme_bw())

# example outbreak --------------------------------------------------------

library(outbreaks) #sample data
library(incidence) #core functions
# library(avallecam) #improvements
outbreak_data <- ebola_sim$linelist$date_of_onset

workflow 01

# packages ----------------------------------------------------------------

if(!require("devtools")) install.packages("devtools")
# if(!require("avallecam")) devtools::install_github("avallecam/avallecam") #improvements

library(tidyverse) #magrittr and purrr packages
library(lubridate) #ymd
library(outbreaks) #sample data
library(incidence) #core functions

# example outbreak --------------------------------------------------------

dat <- ebola_sim$linelist$date_of_onset
i.7 <- incidence(dat, interval=7)
# plot(i.7)
f1 <- fit(i.7[1:20])
f2 <- fit_optim_split(i.7)

# broom like functions ----------------------------------------------------

# tidy
f1 %>% tidy_incidence()
f2 %>% pluck("fit") %>% tidy_incidence()

# glance
f1 %>% glance_incidence()
f2 %>% pluck("fit") %>% glance_incidence()


# using purrr -------------------------------------------------------------

# using purrr::map family function allows easy stratification
# for gender and could be extrapolated to administrative levels
# in country level analysis

incidence_purrr <- ebola_sim$linelist %>% 
  as_tibble() %>% 
  #filter observations explicitly before incidence()
  filter(date_of_onset<lubridate::ymd(20141007)) %>% 
  #stratify by any group of covariates
  group_by(gender) %>% 
  nest() %>% 
  mutate(incidence_strata=map(.x = data,
                              .f = ~incidence(.x %>% pull(date_of_onset),
                                              interval=7))) %>% 
  mutate(strata_fit=map(.x = incidence_strata,
                            .f = fit)) %>% 
  mutate(strata_fit_tidy=map(.x = strata_fit,
                                  .f = tidy_incidence)) %>% 
  mutate(strata_fit_glance=map(.x = strata_fit,
                                       .f = glance_incidence))

# keep only the tibbles
incidence_purrr_tibble <- incidence_purrr %>% 
  select(-data,-incidence_strata,-strata_fit)

# tidy_incidence output
incidence_purrr_tibble %>% 
  unnest(cols = c(strata_fit_tidy))

# glance_incidence output
incidence_purrr_tibble %>% 
  unnest(cols = c(strata_fit_glance))

workflow 02

linelist_raw <- ebola_sim$linelist %>% 
  as_tibble() %>% 
  #filter observations explicitly before incidence()
  # filter(date_of_onset<lubridate::ymd(20141007)) %>% 
  mutate(all="all")

dictionary <- linelist_raw %>% 
  count(all,gender) %>% 
  rownames_to_column(var = "code")

linelist_raw %>% 
  group_by(gender) %>% 
  skimr::skim()

time_delay_set = 7
#### execute -------------------------------

nest_dynamics <- create_nest_dynamics(linelist = linelist_raw,
                                      dictionary = dictionary,
                                      strata_major = all,
                                      strata_minor = gender,
                                      strata_minor_code = code, # unico para diccionario
                                      date_incidence_case = date_of_onset,
                                      date_of_analysis_today=FALSE,
                                      issue_number_set = 0)

nest_dynamics %>% glimpse()

#### nested figures -------------------------------
# nest_summary <- create_nest_summary_map(nest_dynamics = nest_dynamics,
#                                                    geometry = ubigeo_geometria_per2,
#                                                    strata_major=nm_pais,
#                                                    strata_minor=nm_depa)

nest_summary <- create_nest_summary(nest_dynamics = nest_dynamics,
                                    time_limit_fig02 = Inf)

nest_summary %>% glimpse()

region_name <- "all"

nest_summary %>% 
  filter(strata_major==region_name) %>% 
  pull(fig01) %>% pluck(1)
nest_summary %>% 
  filter(strata_major==region_name) %>% 
  pull(fig02) %>% pluck(1)
nest_summary %>% 
  filter(strata_major==region_name) %>% 
  pull(fig03) %>% pluck(1)
# nest_summary %>% 
#   filter(strata_major==region_name) %>% 
#   pull(fig04) %>% pluck(1)

nest_summary %>% 
  filter(strata_major==region_name) %>% 
  pull(tab01) %>% pluck(1)
nest_summary %>% 
  filter(strata_major==region_name) %>% 
  pull(tab02) %>% pluck(1)
nest_summary %>% 
  filter(strata_major==region_name) %>% 
  pull(tab03) %>% pluck(1)
nest_summary %>% 
  filter(strata_major==region_name) %>% 
  pull(tab04) %>% pluck(1)

nest_dynamics %>%
  slice(1) %>%
  pull(incidence_fit_figure) %>%
  pluck(1)
nest_dynamics %>%
  slice(1) %>%
  pull(rt_figure) %>%
  pluck(1)
nest_dynamics %>%
  slice(2) %>%
  pull(rt_figure) %>%
  pluck(1)