.Rhistory (Chriss-MacBook-Pro.local's conflicted copy 2020-03-13)

select(-key) %>%
select(iso3, country, age_range, sex, nutrient, units_long, units_short, stat, value, everything())
# Format data
data <- data_orig %>%
# Rearrange columns
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
arrange(iso3, age_range) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Add nutrient and stat
left_join(header_info, by=c("key"="col_name")) %>%
# Format units
rename(units_long=units) %>%
mutate(units_short=gsub("/person/day", "", units_long)) %>%
# Arrange columns
select(-key) %>%
select(iso3, country, age_range, sex, nutrient, units_long, units_short, stat, value, everything()) %>%
# Spread stats
spread(key="stat", value="value")
View(data)
# Format data
data <- data_orig %>%
# Rearrange columns
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
arrange(iso3, age_range) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Add nutrient and stat
left_join(header_info, by=c("key"="col_name")) %>%
# Format units
rename(units_long=units) %>%
mutate(units_short=gsub("/person/day", "", units_long)) %>%
# Arrange columns
select(-key) %>%
select(iso3, country, age_range, sex, nutrient, units_long, units_short, stat, value, everything()) %>%
# Spread stats
spread(key="stat", value="value") %>%
rename(value_med=md, value_lo=lo, value_hi=hi)
# Format data
data <- data_orig %>%
# Rearrange columns
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
arrange(iso3, age_range) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Add nutrient and stat
left_join(header_info, by=c("key"="col_name")) %>%
# Format units
rename(units_long=units) %>%
mutate(units_short=gsub("/person/day", "", units_long)) %>%
# Arrange columns
select(-key) %>%
select(iso3, country, age_range, sex, nutrient, units_long, units_short, stat, value, everything()) %>%
# Spread stats
spread(key="stat", value="value") %>%
rename(value_med=md, value_lo=lo, value_hi=hi) %>%
select(iso3:units_short, value_med, value_lo, value_hi, everything())
# Inspect
str(data)
# Example plot
g <- ggplot(data %>% filter(country=="Ghana"), aes(x=age_range, y=value_med, fill=sex)) +
facet_wrap(~nutrient, ncol=4) +
geom_bar() +
theme_bw()
g
# Example plot
g <- ggplot(data %>% filter(country=="Ghana"), aes(x=age_range, y=value_med, fill=sex)) +
facet_wrap(~nutrient, ncol=4) +
geom_bar(stat="identity") +
theme_bw()
g
# Example plot
g <- ggplot(data %>% filter(country=="Ghana"), aes(x=age_range, y=value_med, fill=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity") +
theme_bw()
g
View(data)
# Example plot
g <- ggplot(data %>% filter(country=="Ghana"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity") +
theme_bw()
g
?geom_bar
# Example plot
g <- ggplot(data %>% filter(country=="Ghana" & nutrient=="calcium"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity") +
theme_bw()
g
# Example plot
g <- ggplot(data %>% filter(country=="Ghana" & nutrient=="calcium"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw()
g
# Example plot
g <- ggplot(data %>% filter(country=="Germany" & nutrient=="calcium"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw()
g
# Export
saveRDS(data, file=file.path(outputdir, "genus_nutrient_supplies_by_age_sex_2011.Rds"))
# Example plot
g <- ggplot(data %>% filter(country=="Japan" & nutrient=="calcium"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw()
g
# Example plot
g <- ggplot(data %>% filter(country=="Japan"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw()
g
# Example plot
g <- ggplot(data %>% filter(country=="Japan"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw() +
theme(legend.position = "bottom")
g
# Example plot
g <- ggplot(data %>% filter(country=="Japan"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw() +
labs(x="Age range", y="Daily per capita supply") +
theme(legend.position = "bottom")
g
# Example plot
g <- ggplot(data %>% filter(country=="Japan"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw() +
labs(x="Age range", y="Daily per capita supply") +
theme(legend.position = "bottom",
axis.text.x = element_text(angle = 90, vjust = 0.5))
g
# Example plot
g <- ggplot(data %>% filter(country=="Japan"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw() +
labs(x="Age range", y="Daily per capita supply") +
theme(legend.position = "bottom",
axis.text.x = element_text(angle = 90, vjust = 1))
g
# Example plot
g <- ggplot(data %>% filter(country=="Japan"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw() +
labs(x="Age range", y="Daily per capita supply") +
theme(legend.position = "bottom",
axis.text.x = element_text(angle = 90, vjust = 0.5, hjuts=0))
# Example plot
g <- ggplot(data %>% filter(country=="Japan"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw() +
labs(x="Age range", y="Daily per capita supply") +
theme(legend.position = "bottom",
axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=0))
g
# Example plot
g <- ggplot(data %>% filter(country=="Japan"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw() +
labs(x="Age range", y="Daily per capita supply") +
theme(legend.position = "bottom",
axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))
g
# Example plot
g <- ggplot(data %>% filter(country=="France"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw() +
labs(x="Age range", y="Daily per capita supply") +
theme(legend.position = "bottom",
axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))
g
# Example plot
g <- ggplot(data %>% filter(country=="Canada"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw() +
labs(x="Age range", y="Daily per capita supply") +
theme(legend.position = "bottom",
axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))
g
# Files
files_merge <- list.files(file.path(inputdir, "nutrient_totals_by_age_sex"), pattern=".csv")
# Example data
exdata <- read.csv(file.path(inputdir, "nutrient_totals_by_age_sex", files_merge[1]), as.is=T, na.strings="*")
# Header info
header_info <-  read.csv(file.path(inputdir, "nutrient_totals_by_age_sex", x), as.is=T, nrows=3, header=F, na.strings="*") %>%
select(4:ncol(.)) %>% t() %>%
as.data.frame(stringsAsFactors=F) %>%
rename(nutrient=V1, stat=V2, units=V3) %>%
mutate(stat=recode(stat, "High 95% UI"="hi", "Low 95% UI"="lo", "Median"="md"),
col_name=paste(nutrient, stat, sep="_"))
# Column names
col_names <- c("iso3", "country", "blank", header_info$col_name)
# Merge
x <- files_merge[1]
data_orig <- purrr::map_df(files_merge, function(x) {
# Get sex
sex <- strsplit(x, split="_")[[1]][3] %>%
gsub(".csv", "", .)
# Get age range
age_range <- strsplit(x, split="_")[[1]][2] %>%
gsub("age", "", .)
# Read data
fdata <- read.csv(file.path(inputdir, "nutrient_totals_by_age_sex", x), as.is=T, skip=3, na.strings="*", col.names = col_names) %>%
mutate(age_range=age_range,
sex=sex) %>%
select(age_range, sex, everything())
})
# Format data
data <- data_orig %>%
# Rearrange columns
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
arrange(iso3, age_range) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Add nutrient and stat
left_join(header_info, by=c("key"="col_name")) %>%
# Format units
rename(units_long=units) %>%
mutate(units_short=gsub("/person/day", "", units_long)) %>%
# Arrange columns
select(-key) %>%
select(iso3, country, age_range, sex, nutrient, units_long, units_short, stat, value, everything()) %>%
# Spread stats
spread(key="stat", value="value") %>%
rename(value_med=md, value_lo=lo, value_hi=hi) %>%
select(iso3:units_short, value_med, value_lo, value_hi, everything())
# Inspect
str(data)
# Example plot
g <- ggplot(data %>% filter(country=="Canada"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw() +
labs(x="Age range", y="Daily per capita supply") +
theme(legend.position = "bottom",
axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))
g
# Export
saveRDS(data, file=file.path(outputdir, "genus_nutrient_supplies_by_age_sex_2011.Rds"))
# Example plot
g <- ggplot(data %>% filter(country=="Ghana"), aes(x=age_range, y=value_med, fill=sex, group=sex)) +
facet_wrap(~nutrient, ncol=4, scale="free_y") +
geom_bar(stat="identity", position="dodge") +
theme_bw() +
labs(x="Age range", y="Daily per capita supply") +
theme(legend.position = "bottom",
axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))
g
# Clear workspace
rm(list = ls())
# Setup
################################################################################
# Packages
library(ggplot2)
library(tidyverse)
# Directories
inputdir <- "data/genus/raw"
outputdir <- "data/genus/processed"
# The GENuS Dataverse:
# https://dataverse.harvard.edu/dataverse/GENuS
# GENuS datasets
# 1. Food composition tables for GENuS
# 2. Nutrient Totals (incl. Fortification) by Age and Sex (2011)
# 3. Edible Food by Age and Sex (2011)
# 4. Nutrient Totals by Age and Sex (2011)
# 5. Total Nutrient Supplies by Country and Year
# 6. Nutrient Supplies by Food and Country (2011)
# 7. Total Nutrient Supply including Fortification (2011)
# 8. Edible Food by Country and Year
# Files
files_merge <- list.files(file.path(inputdir, "edible_food_by_age_sex"), pattern=".csv")
# Example data
exdata <- read.csv(file.path(inputdir, "edible_food_by_age_sex", files_merge[1]), as.is=T, na.strings="*")
# Header info
header_info <-  read.csv(file.path(inputdir, "edible_food_by_age_sex", files_merge[1]), as.is=T, nrows=3, header=F, na.strings="*") %>%
select(4:ncol(.)) %>% t() %>%
as.data.frame(stringsAsFactors=F) %>%
rename(food=V1, stat=V2, units=V3) %>%
mutate(stat=recode(stat, "High 95% UI"="hi", "Low 95% UI"="lo", "Median"="md"),
col_names=paste(food, stat, sep="_"))
col_names <- c("iso3", "country", "blank", header_info$col_names)
# Merge data
x <- files_merge[1]
data_orig <- purrr::map_df(files_merge, function(x) {
# Get sex
sex <- strsplit(x, split="_")[[1]][4] %>%
gsub(".csv", "", .)
# Get age range
age_range <- strsplit(x, split="_")[[1]][3] %>%
gsub(".csv", "", .)
# Read data
fdata <- read.csv(file.path(inputdir, "edible_food_by_age_sex", x),
as.is=T, skip=3, na.strings="*", col.names=col_names) %>%
mutate(age_range=age_range,
sex=sex) %>%
select(age_range, sex, everything())
})
# Format data
data <- data_orig %>%
# Remove blank column
select(-blank) %>%
select(iso3, country, age_range, sex, everything())
# Format data
data <- data_orig %>%
# Remove blank column
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.))
data <- data_orig %>%
# Remove blank column
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Arrange
arrange(iso3, age_range) %>%
# Format food/stat
mutate(food_name_messy=gsub("_md|_lo|_hi", "", key),
stat=ifelse(grepl("_md", key), "median",
ifelse(grepl("_lo", key), "low", "high")))
# Format data
data <- data_orig %>%
# Remove blank column
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Arrange
arrange(iso3, age_range) %>%
# Format food/stat
mutate(food_name_messy=gsub("_md|_lo|_hi", "", key),
stat=ifelse(grepl("_md", key), "median",
ifelse(grepl("_lo", key), "low", "high"))) %>%
# Reshape
spread(key="stat", value="value") %>%
rename(value_md=median, value_lo=low, value_hi=high)
View(data)
data <- data_orig %>%
# Remove blank column
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Arrange
arrange(iso3, age_range) %>%
# Format food/stat
mutate(food_name_messy=gsub("_md|_lo|_hi", "", key),
stat=ifelse(grepl("_md", key), "median",
ifelse(grepl("_lo", key), "low", "high")))
View(data)
?spread
vignette("pivot")
data <- data_orig %>%
# Remove blank column
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Arrange
arrange(iso3, age_range) %>%
# Format food/stat
mutate(food_name_messy=gsub("_md|_lo|_hi", "", key),
stat=ifelse(grepl("_md", key), "median",
ifelse(grepl("_lo", key), "low", "high")))
data <- data_orig %>%
# Remove blank column
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Arrange
arrange(iso3, age_range) %>%
# Format food/stat
mutate(food_name_messy=gsub("_md|_lo|_hi", "", key),
stat=ifelse(grepl("_md", key), "median",
ifelse(grepl("_lo", key), "low", "high"))) %>%
select(iso3, country, age_range, sex, food_name_messy, stat, value)
# Format data
data <- data_orig %>%
# Remove blank column
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Arrange
arrange(iso3, age_range) %>%
# Format food/stat
mutate(food_name_messy=gsub("_md|_lo|_hi", "", key),
stat=ifelse(grepl("_md", key), "median",
ifelse(grepl("_lo", key), "low", "high"))) %>%
select(iso3, country, age_range, sex, food_name_messy, stat, value) %>%
# Reshape
spread(key="stat", value="value") %>%
rename(value_md=median, value_lo=low, value_hi=high)
View(data)
# Format data
data <- data_orig %>%
# Remove blank column
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Arrange
arrange(iso3, age_range) %>%
# Format food/stat
mutate(food_name_messy=gsub("_md|_lo|_hi", "", key),
stat=ifelse(grepl("_md", key), "median",
ifelse(grepl("_lo", key), "low", "high"))) %>%
select(iso3, country, age_range, sex, food_name_messy, stat, value) %>%
# Reshape
spread(key="stat", value="value") %>%
rename(value_md=median, value_lo=low, value_hi=high) %>%
# Format age/sex
mutate(age_range=gsub("age", "", age_range),
sex=recode(sex, "bothsexes"="both"))
View(data)
View(header_info)
# Format data
data <- data_orig %>%
# Remove blank column
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Arrange
arrange(iso3, age_range) %>%
# Format food/stat
mutate(food_name_messy=gsub("_md|_lo|_hi", "", key),
stat=ifelse(grepl("_md", key), "median",
ifelse(grepl("_lo", key), "low", "high")))
View(data)
View(header_info)
data <- data_orig %>%
# Remove blank column
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Arrange
arrange(iso3, age_range) %>%
# Format food/stat
left_join(header_info, c("key"="col_names"))
# mutate(food_name_messy=gsub("_md|_lo|_hi", "", key),
#        stat=ifelse(grepl("_md", key), "median",
#                    ifelse(grepl("_lo", key), "low", "high"))) %>%
select(iso3, country, age_range, sex, food, units, stat, value) %>%
# Reshape
spread(key="stat", value="value")
data <- data_orig %>%
# Remove blank column
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Arrange
arrange(iso3, age_range) %>%
# Add food, units, stat
left_join(header_info, c("key"="col_names")) %>%
# mutate(food_name_messy=gsub("_md|_lo|_hi", "", key),
#        stat=ifelse(grepl("_md", key), "median",
#                    ifelse(grepl("_lo", key), "low", "high"))) %>%
select(iso3, country, age_range, sex, food, units, stat, value)
View(data)
data <- data_orig %>%
# Remove blank column
select(-blank) %>%
select(iso3, country, age_range, sex, everything()) %>%
# Convert wide-to-long
gather(key="key", value="value", 5:ncol(.)) %>%
# Arrange
arrange(iso3, age_range) %>%
# Add food, units, stat
left_join(header_info, c("key"="col_names")) %>%
# mutate(food_name_messy=gsub("_md|_lo|_hi", "", key),
#        stat=ifelse(grepl("_md", key), "median",
#                    ifelse(grepl("_lo", key), "low", "high"))) %>%
select(iso3, country, age_range, sex, key, food, units, stat, value)
shiny::runApp('shiny/v2')
# Clear workspace
rm(list = ls())
# Setup
################################################################################
# Packages
library(ggplot2)
library(tidyverse)
# Directories
genusdir <- "data/genus/processed"
# Nutrient supply by country and year
nut_cntry_yr_orig <- read.rds(file.path(genusdir, "genus_nutrient_supplies_by_cntry_year.Rds"))
# Nutrient supply by country and year
nut_cntry_yr_orig <- readRDS(file.path(genusdir, "genus_nutrient_supplies_by_cntry_year.Rds"))
sort(unique(nut_cntry_yr_orig$nutrient))