Tuberculosis_R

# TUBERCULOSIS EVOLUTION 1980 - 2008

#Reading a file and getting a overview of the data
tuber <- read.table(url("http://stat405.had.co.nz/data/tb.csv"), header = T, sep = ",")
head(tuber)
colnames(tuber)
str(tuber)
dim(tuber)

#Rename column of group and age
names(tuber) <- gsub("new_sp_f", "female_", names(tuber))
names(tuber) <- gsub("new_sp_m", "male_", names(tuber))
names(tuber)[3] <- "u_u"

# Reshaping the table
tuber2 <- melt(tuber, id.vars = c("iso2", "year"))
summary(tuber2$variable)
names(tuber2)[3] <- "group_age"

#splitting column "group_age" into two columns
#option 1
library(stringr)
tuber2$group <- str_split_fixed(tuber2$group_age, "_", 2)

#option2
library(tidyr)
tuber2 <- tuber2 %>% separate(group_age, c("group", "age"))

#let's have a look at the quality of the data -> 68% of lines contain NA values
sum(is.na(tuber2))
dim(tuber2)
tuber3 <- na.omit(tuber2)

# CALCULATION
# Number of cases per country and year and visualize it
#option1
tmp <- tapply(tuber2$value, tuber2$year, sum, na.rm = T)
#option2
tmp <- ddply(tuber3, .(year), summarize, total = sum(value))

library(ggplot2)
ggplot(tmp, aes(x = year, y = total)) + geom_point(color = "red") +
  theme_light() +
  ggtitle("Evolution of the Tuberculosis iin the country from 1980 till 2008")+