file_E_2.Rmd

---
title: "Data Viz/Table1"
author: "David Sasson"
date: "5/11/2018"
output:
  html_document: default
  pdf_document: default
  word_document: default
---

```{r setup, include=FALSE}
# knitr::opts_knit$set(root.dir = "~/Desktop/MIT/HST936/Project/")
packs <- c("xlsx", "dplyr", "ggplot2", "gmodels", "ggcorrplot", "caret", "caTools", "gridExtra", "ggthemes", "grid", "ggpubr", "tableone", "knitr", "cowplot")
lapply(packs, require, character.only = TRUE)
set.seed(123)
```

```{r reading_file}
TB <- read.xlsx("data/tb_data.xlsx", sheetIndex = 1, stringsAsFactors = FALSE, header = TRUE)
```

```{r def_outcome}
# unique(TB$outcome) -- ignoring notreported and unknown
TB$trtfailure[TB$outcome == "completed" | TB$outcome == "cured" | TB$outcome == "still_on_treatment"] <- 0
TB$trtfailure[TB$outcome == "failure" | TB$outcome == "death" | TB$outcome == "default" | TB$outcome == "died"] <- 1
TB$trtfailure <- as.factor(TB$trtfailure)

# Remove duplicate entries from the same patient
TB <- TB[!duplicated(TB$patient_id), ]
```




```{r data cleaning}
# Country
TB$country <- as.factor(TB$country)

# Age
TB$age_of_onset <- as.numeric(TB$age_of_onset)

# Sex
TB$gender <- as.factor(TB$gender)

# Education
TB$education <- as.factor(TB$education)

# Employment
TB$employment <- as.factor(TB$employment)

# Number of daily contacts
TB$number_of_daily_contacts <- as.numeric(TB$number_of_daily_contacts)

# Type of resistance
TB$type_of_resistance <- as.factor(TB$type_of_resistance)

# BMI
TB$bmi <- as.numeric(TB$bmi)

# Localization of the lesion in the brain
TB$lung_localization <- as.factor(TB$lung_localization)

# Dissemination
TB$dissemination <- as.factor(TB$dissemination)

# Lung cavity size
TB$lungcavity_size <- as.factor(TB$lungcavity_size)

# Shadow pattern
TB$shadow_pattern <- as.factor(TB$shadow_pattern)

# Pneumothorax
TB$pneumothorax <- as.factor(TB$pneumothorax)

# Pleuritis
TB$plevritis <- as.factor(TB$plevritis)

# Node calcinosis
TB$nodicalcinatum <- as.factor(TB$nodicalcinatum)

# Caverna
TB$totalcavernum <- as.factor(TB$totalcavernum)

# Social risk factors
TB$social_risk_factors <- as.factor(TB$social_risk_factors)

# Treatment 
TB$regimen_drug <- as.factor(TB$regimen_drug)

TB$number_of_children <- as.numeric(TB$number_of_children)
TB$x_ray_count <- as.numeric(TB$x_ray_count)
TB$ct_count <- as.numeric(TB$ct_count)
TB$registration_date <- as.Date(TB$registration_date)

# check results
# sapply(TB, class)
```






```{r age}
ggplot(TB, aes(age_of_onset, fill=gender)) + 
  geom_histogram(binwidth=2) +
  facet_wrap(~country, nrow = 2) + 
  xlab("Age of Onset")+
  ylab("Count") + 
  theme_pubr()+ guides(fill=guide_legend(title="")) + 
  scale_fill_manual(values=c("indianred1", "steelblue"))
```



```{r outcome}
ggplot(TB, aes(x = outcome, y = type_of_resistance)) +
  geom_count(col="indianred1") + 
  scale_size_area() +
  coord_flip()  + 
  ylab("Type of Resistance") + 
  xlab("Outcome") + 
  theme_pubr() 
```



```{r missing data}
g <-ggplot(TB, aes(x = outcome, fill = genomic_data_exists)) +
  geom_bar() + 
  coord_flip() + 
  ggtitle("Genomic Data")+
  guides(fill=FALSE) + ylab("") + xlab("") + 
  theme_pubr() + 
  scale_fill_manual(values=c("indianred1", "steelblue"))

x <- ggplot(TB, aes(x = outcome, fill = x_ray_exists)) +
  geom_bar() + 
  coord_flip() + 
  ggtitle("X-Ray Data") +
  guides(fill=FALSE)+ ylab("")+  
  theme_pubr() + 
  scale_fill_manual(values=c("indianred1", "steelblue"))


c <- ggplot(TB, aes(x = outcome, fill=ct_exists)) +
  geom_bar() + 
  coord_flip()+ 
  guides(fill=FALSE)+ 
  ggtitle("CT Scan Data")  + xlab("")+ 
  theme_pubr() + 
  scale_fill_manual(values=c("indianred1", "steelblue"))

plot_grid(g, x, c, ncol=1) 

```




```{r}
ggplot(TB) +
  geom_bar(aes(outcome), fill="indianred1") + 
  xlab("Outcome") +
  ylab("Count") +
  coord_flip() + 
  facet_wrap(~country)  + 
  theme_pubr()
```


```{r tableone}
# smaller table
listVars <- c("gender", "bmi","age_of_onset" , "country", "social_risk_factors","education", "regimen_drug")

table1 <- CreateTableOne(vars = listVars, data = TB,  strata = c("trtfailure"), test=F)

table1  %>% print(
  printToggle      = FALSE,
  showAllLevels    = TRUE,
  cramVars         = "kon"
) %>%
{data.frame(
  variable_name   = gsub(" ", "&nbsp;", rownames(.), fixed = TRUE), .,
  row.names        = NULL,
  check.names      = FALSE,
  stringsAsFactors = FALSE)} %>% kable()
```



```{r appendix}
# all vars for appendix
allVars <- c("age_of_onset", "gender", "country", "affect_pleura", "bmi", "employment", "totalcavernum", "nodicalcinatum", "type_of_resistance", "lung_capacity_decrease","lungcavity_size", "pneumothorax", "dissemination", "regimen_drug", "shadow_pattern", "ct_count", "microscopy", "number_of_daily_contacts", "lung_localization", "x_ray_count", "process_prevalence","plevritis", "social_risk_factors","education", "regimen_drug")

table1.full <- CreateTableOne(vars = allVars, data = TB, strata = c("trtfailure"), test=F)

table1.full  %>% print(
  printToggle      = FALSE,
  showAllLevels    = TRUE,
  cramVars         = "kon"
) %>%
{data.frame(
  variable_name   = gsub(" ", "&nbsp;", rownames(.), fixed = TRUE), .,
  row.names        = NULL,
  check.names      = FALSE,
  stringsAsFactors = FALSE)} %>% kable()

```