merge_disease_states.Rmd

---
title: "merge_disease_states"
output: html_document
---

# check current variables in hgm_subject_df for columns related to disease state/host health
```{r}
# View(as.vector(colnames(hgm_subject_df)))
# df <- hgm_subject_df
# disease_var <- c("host disease", # add!
#                  "study disease", # add
#                  "host disease status", # add
#                  "host disease stage", # na, don't add
#                  "host disease outcome", # na, don't add
#                  "disease stage", # na, don't add
#                  "disease", # na, don't add
#                  "host health state", # na, don't add
#                  "health state", # na, don't add
#                  "host_health", # na, don't add
#                  "gastrointestinal tract disorder", # add
#                  "host phenotype", # na, don't add
#                  "subject is affected", # look into this more... "yes" and "no"
#                  "gastroinstest_disord", # add
#                  "histological_diagnosis", # na, don't add
#                  "ontological term", # na, don't add
#                  "pathotype", # na, don't add
#                  "is the sequenced pathogen host associated?", # na, don't add
#                  "treatment", # na, don't add
#                  "pathovar", # na, don't add
#                  "host_desease") # na, don't add


# View(table(df[,disease_var[1]]))

```

# add values in selected disease variables to merged_health_status
```{r}

df$merged_health_status <- NA

selected_disease_var <- c("host disease",
                          "study disease",
                          "host disease status", 
                          "gastrointestinal tract disorder",
                          "gastroinstest_disord"
                          )

for (variable in selected_disease_var) {
  df$merged_health_status <- ifelse(is.na(df$merged_health_status), 
                                    df[[variable]],
                                    df$merged_health_status)
}

View(table(df$merged_health_status))
```

# convert NA values
```{r}
na_values <- c("missing",
               "Missing", 
               "", 
               " ",
               "not collected",
               "not applicable", 
               "Not Applicable", 
               "Not Collected", 
               "N/A", 
               "n/a",
               "not collected",
               "not determined",
               "not provided",
               "Unknown",
               "unknown",
               "not available",
               "Not collected",
               "Not applicable",
               "Not available",
               "na",
               "NOT COLLECTED")

# nothing new to add to the NA keywords list
# next, let's change the NA keywords to NA in the df!

for (variable in disease_var) {
  for (replace_value in na_values) {
    df[[variable]] <- replace(df[[variable]], df[[variable]] == replace_value, NA)
  }
}
```