covid-19-3d-plot-ts-animation.Rmd

---
title: "COVID-19  World Stats and Time Series"
author: "Divyanshu Marwah"
date: "4/17/2020"
output: flexdashboard::flex_dashboard
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(warning = FALSE,
                      error = FALSE,
                      echo = FALSE)
```

```{r install_packages,include=FALSE}
#install.packages("tidyverse")
#install.packages("lubridate")
#install.packages("plotly")
#install.packages("highcharter")
#install.packages("DT")
#install.packages("gifski")
#install.packages("ggthemes")
#install.packages("gganimate")
#install.packages("flexdashboard")
#to view the outputs of every code chunk, remove include false from r blocks
```

```{r load_packages,include=FALSE}
library(flexdashboard)
library(tidyverse)
library(lubridate)
library(plotly)
library(DT)
library(ggthemes)
library(gganimate)
library(highcharter)
library(RColorBrewer)
```
```{r time_series_confirmed,include=FALSE}
# time series of confirmed cases in each area
ts_confirmed <- read_csv(
  file = "https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_confirmed_global.csv",
  col_types = cols(
    .default = col_double(),
    `Province/State` = col_character(),
    `Country/Region` = col_character()
  )) %>%
  pivot_longer(cols = -c(`Province/State`, `Country/Region`, Lat, Long), names_to = "Date", values_to = "Confirmed") %>%
  mutate(Date = mdy(Date))

head(ts_confirmed)
```
```{r time_series_recovered,include=FALSE}
# time series of recovered cases in each area
ts_recovered <- read_csv(
  file = "https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_recovered_global.csv",
  col_types = cols(
    .default = col_double(),
    `Province/State` = col_character(),
    `Country/Region` = col_character()
  )) %>%
  pivot_longer(cols = -c(`Province/State`, `Country/Region`, Lat, Long), names_to = "Date", values_to = "Recovered") %>%
  mutate(Date = mdy(Date))
head(ts_recovered)
```
```{r time_series_deaths,include=FALSE}
# time series of deaths cases in each area
ts_deaths <- read_csv(
  file = "https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_deaths_global.csv",
  col_types = cols(
    .default = col_double(),
    `Province/State` = col_character(),
    `Country/Region` = col_character()
  )) %>%
  pivot_longer(cols = -c(`Province/State`, `Country/Region`, Lat, Long), names_to = "Date", values_to = "Deaths") %>%
  mutate(Date = mdy(Date))
head(ts_deaths)
```
```{r time_series,include=FALSE}
# combining all three
ts_all <- ts_confirmed %>%
  left_join(ts_recovered) %>%
  mutate(Recovered = replace_na(Recovered, replace = 0)) %>%
  left_join(ts_deaths) %>%
  mutate(Deaths = replace_na(Deaths, replace = 0))
head(ts_all)
```

```{r time_series_region,include=FALSE}
# group by region and fetch the latest data
region_recent <- ts_all %>%
  filter(Date == max(Date)) %>%
  group_by(`Country/Region`, Date) %>%
  summarise(Confirmed = sum(Confirmed),
            Recovered = sum(Recovered),
            Deaths = sum(Deaths))

head(region_recent,include=FALSE)
```
```{r country_code,include=FALSE}
# get country code
code <- read_csv('./novel-corona-virus-2019-dataset/country_codes.csv',
                 col_types = cols(
                   COUNTRY = col_character(),
                   `GDP (BILLIONS)` = col_double(),
                   CODE = col_character())) %>%
  select(COUNTRY, CODE) %>%
  rename(Region = COUNTRY,
         Code  = CODE) %>%
  rownames_to_column("id")
code$id <- as.integer(code$id)
head(code)
```
```{r replace_region_name,include=FALSE}
# Rename the unmatched region name in code
code$Region <- code$Region %>%
  str_replace(pattern = "United States", replacement = "US") %>%
  str_replace(pattern = "Macedonia", replacement = "North Macedonia") %>%
  str_replace(pattern = "Czech Republic", replacement = "Czechia") %>%
  str_replace(pattern = "Taiwan", replacement = "Taiwan*") %>%
  str_replace(pattern = "West Bank", replacement = "West Bank and Gaza") %>%
  str_replace(pattern = "Congo, Democratic Republic of the", replacement = "Congo (Kinshasa)") %>%
  str_replace(pattern = "Congo, Republic of the", replacement = "Congo (Brazzaville)") %>%
  str_replace(pattern = "Bahamas, The", replacement = "Bahamas") %>%
  str_replace(pattern = "Swaziland", replacement = "Eswatini") %>%
  str_replace(pattern = "Gambia, The", replacement = "Gambia")
```
```{r recent_case_cumulative,include=FALSE}
# recent cumulative case and join daily case summary with code name
region_recent_code <- region_recent %>%
  left_join(code, by = c("Country/Region" = "Region")) %>%
  arrange(desc(Confirmed))
```
```{r set_map,include=FALSE}
#Set country boundaries as light grey,  hex code #D3D3D3
line <- list(color = toRGB("#D3D3D3"), width = 0.2)

#Specify map projection and options
geo <- list(
     showframe = FALSE,
     showcoastlines = FALSE,
     projection = list(type = 'orthographic'),
     resolution = '100',
     showcountries = TRUE,
     countrycolor = '#D3D3D3',
     showocean = TRUE,
     oceancolor = '#006699',
     showlakes = TRUE,
     lakecolor = '#a9d6f5',
     showrivers = TRUE,
     rivercolor = '#99c0db',
     bgcolor = '#e8f7fc')
```

3D Visualization - Latest
============================================

Column 1
--------------------------------------------------

### 3D Map for Confirmed Cases in Each Region
```{r confirm_map_3d,include=FALSE}
confirm_map_3d <- plot_geo() %>%
  layout(geo = geo,
         paper_bgcolor = '#e8f7fc') %>%
  add_trace(data = region_recent_code,
            z = ~Confirmed, 
            color = ~Confirmed, 
            colors = 'Reds',
            text = ~`Country/Region`,
            locations = ~Code, 
            marker = list(line = line))
```


```{r view_confirm_map_3d}
confirm_map_3d
```

Column 2
--------------------------------------------------

### Search by Region

```{r table_case,include=FALSE}
regions_interactive_table <- region_recent_code %>%
  select(`Country/Region`,Code, Date, Confirmed, Recovered, Deaths) %>%
  arrange(desc(Confirmed)) %>%
  datatable(
    rownames = FALSE,
    fillContainer = TRUE,
    options = list(
      autoWidth = TRUE,
      bPaginate = TRUE,
      pageLength = 10)
  )
```
```{r view_interactive_table}
regions_interactive_table
```

### Tree Map showing Density of Confirmed Cases
```{r tree_map,echo=FALSE}
region_recent_code  %>%
  hchart(type = "treemap",
         hcaes(name = `Code`,
               value = Confirmed,
               colorValue = Confirmed)) %>%
  hc_colorAxis(minColor = brewer.pal(10, "Reds")[1],
               maxColor = brewer.pal(10, "Reds")[9])

```

```{r barchart_cum_all,include=FALSE}
ts_all_date <- ts_all %>%
  rename(Region = `Country/Region`) %>%
  group_by(Date) %>%
  summarise(Confirmed = sum(Confirmed),
            Deaths = sum(Deaths),
            Recovered = sum(Recovered))
```

World Trend Charts
============================================

Column 1
--------------------------------------------------

### Stacked Bar Chart showing the trend of Cumulative Csases
```{r barchart_cum_date,include=FALSE}
#transforming the df  to get  the  cumulative number date-wise
ts_date_long <- ts_all_date %>% 
    select(-Confirmed) %>%
    pivot_longer(cols = -Date, names_to = "Status", values_to = "Cases")
head(ts_date_long)
```



```{r barchart_cum,include=FALSE}
barchart <- ggplot(data = ts_all_date, aes(x = Date)) +
  geom_bar(aes(y = Confirmed), position = "stack", stat = "identity", fill = "#1b9e77") + 
  geom_bar(data = ts_date_long, aes(y = Cases, fill = Status), position = "stack", stat = "identity") +
  scale_fill_manual(values = c("#d95f02", "#7570b3")) + #dark2 theme from colorbrewer2.org
  scale_y_continuous(labels = scales::label_number_si(accuracy = 0.1)) +
  theme_gdocs(base_size = 10)+
  theme(axis.title = element_blank())
show_barchart <- ggplotly(barchart)%>%  
  layout(legend = list(orientation = 'h'))
```


```{r barchart_cum_show}
show_barchart
```

Column 2
--------------------------------------------------

### Animated Time Series Increment in the World 

```{r ts_increment,include=FALSE}
ts_increment_long <- ts_all_date %>%
  mutate(Confirmed = Confirmed - lag(Confirmed,1), #for creating a lagged version for Ts
            Deaths = Deaths - lag(Deaths,1),
            Recovered = Recovered - lag(Recovered,1)) %>%
  filter(Date != min(Date)) %>%
  pivot_longer(-Date, names_to = "Case", values_to = "Increment")
```

```{r animate_world_increment,include=FALSE}
animate_world_increment <- 
  ggplot(data = ts_increment_long, 
         mapping = aes(x = Date, y = Increment, group = Case, color = Case)) +
  geom_line() +
  scale_color_brewer(palette = "Dark2") +
  geom_segment(aes(xend = max(Date)+1, yend = Increment), linetype = 2, colour = 'grey') +
  geom_text(aes(x = max(Date)+1, label = Case), hjust = 0) +
  coord_cartesian(xlim = c(min(ts_increment_long$Date), max(ts_increment_long$Date)+7), clip = 'off') +
  theme(legend.position = "none", axis.title.x = element_blank()) +
  guides(size = FALSE) + 
  geom_point(aes(size = Increment)) +
  scale_size(range = c(2, 10)) +
  transition_reveal(Date) +
  labs(title = 'World Case Increment at Date: {frame_along}')
world_animated_ts <- animate(animate_world_increment, nframes = 200, fps = 5, end_pause = 80)
anim_save("animate_world_increment.gif", animation = world_animated_ts)
```


```{r view_animate_world_increment}
world_animated_ts
```

```{r ts_top_10_countries,include=FALSE}
countries  <- c("US","Spain","Italy","France","Germany",
                "China","United Kingdom","Iran","Turkey","Belgium")

ts_countries  <- ts_all%>%
  filter(`Country/Region` %in% countries)

head(ts_countries)
```
```{r  join_by_country_code,include=FALSE}
ts_countries <- ts_countries %>%
  rename(Region = `Country/Region`)%>%
  filter(Confirmed > 0) %>%
  left_join(code)

head(ts_countries)
```
```{r combine_states,include=FALSE}
new_ts_country <- ts_countries%>%
  group_by(Code,Date)%>%
  summarize(Confirmed = sum(Confirmed),
            Recovered = sum(Recovered),
            Deaths = sum(Deaths))

head(new_ts_country)
```

Time Series Plot - Most Affected
============================================

Column 1
--------------------------------------------------

### Time Series Increment in Most Affected Countries - Animated

```{r ts_animate_confirmed,include=FALSE}
animate_confirmed <- 
  ggplot(data = new_ts_country, 
         mapping = aes(x = Date, y = Confirmed, group = Code, color = Code)) +
  geom_line() +
  scale_color_brewer(palette = "Paired") +  # palette from https://colorbrewer2.org/
  scale_y_log10() +
  geom_segment(aes(xend = max(Date)+1, yend = Confirmed), linetype = 2, colour = 'grey') +
  geom_text(aes(x = max(Date)+1, label = Code), hjust = 0) +
  coord_cartesian(clip = 'off') +
  theme(legend.position = "none", axis.title.x = element_blank()) +
  guides(size = FALSE) + 
  geom_point(aes(size = Confirmed), alpha = 0.7) +
  scale_size(range = c(2, 10)) +
  transition_reveal(Date) +
  labs(title = 'Confirmed Cases at Date: {frame_along}')

coun_ts_confirmed_anim <- animate(animate_confirmed, nframes = 200, fps = 5, end_pause = 80)
anim_save("./animate_confirmed_coun.gif", animation = coun_ts_confirmed_anim)
```
```{r view_ts_animate_confirmed}
coun_ts_confirmed_anim
```

Column 2
--------------------------------------------------

### Deaths Cases in Most Affected Countries - Animated

```{r ts_animate_deaths,include=FALSE}
animate_deaths <- 
  ggplot(data = new_ts_country, 
         mapping = aes(x = Date, y = Deaths, group = Code, color = Code)) +
  geom_line() +
  scale_color_brewer(palette = "Paired") +  # palette from https://colorbrewer2.org/
  scale_y_log10() +
  geom_segment(aes(xend = max(Date)+1, yend = Deaths), linetype = 2, colour = 'grey') +
  geom_text(aes(x = max(Date)+1, label = Code), hjust = 0) +
  coord_cartesian(clip = 'off') +
  theme(legend.position = "none", axis.title.x = element_blank()) +
  guides(size = FALSE) + 
  geom_point(aes(size = Deaths), alpha = 0.7) +
  scale_size(range = c(2, 10)) +
  transition_reveal(Date) +
  labs(title = 'Death Cases at Date: {frame_along}')

coun_ts_death_anim <- animate(animate_deaths, nframes = 200, fps = 5, end_pause = 80)
anim_save("./animate_death_coun.gif", animation = coun_ts_death_anim)
```


```{r view_ts_animate_deaths}
coun_ts_death_anim
```

### Recovered Cases Most Affected Countries - Animated

```{r ts_animate_recovery,include=FALSE}
animate_recovered <- 
  ggplot(data = new_ts_country, 
         mapping = aes(x = Date, y = Recovered, group = Code, color = Code)) +
  geom_line() +
  scale_color_brewer(palette = "Paired") +  # palette from https://colorbrewer2.org/
  scale_y_log10() +
  geom_segment(aes(xend = max(Date)+1, yend = Recovered), linetype = 2, colour = 'grey') +
  geom_text(aes(x = max(Date)+1, label = Code), hjust = 0) +
  coord_cartesian(clip = 'off') +
  theme(legend.position = "none", axis.title.x = element_blank()) +
  guides(size = FALSE) + 
  geom_point(aes(size = Recovered), alpha = 0.7) +
  scale_size(range = c(2, 10)) +
  transition_reveal(Date) +
  labs(title = 'Recovered Cases at Date: {frame_along}')

coun_ts_recov_anim <- animate(animate_recovered, nframes = 200, fps = 5, end_pause = 80)
anim_save("./animate_recovery_coun.gif", animation = coun_ts_recov_anim)
```


```{r view_ts_animate_recovery}
coun_ts_recov_anim
```