fake-data.Rmd

---
title: "Data Example"
author: "Mark Dunning"
date: "2 February 2016"
output: html_document
---


```{r}
library(wakefield)
set.seed(666)

random_patients <- function(n) {
  as.data.frame(r_data_frame(
    n,
    id,
    name,
    race,
    age,
    sex,
    smokes,
    height_cm(name = "Height"),
    normal(name="Weight"),
    birth(random = TRUE, x = NULL, start = Sys.Date() - 365 * 45, k = 365*2,by = "1 days"),
    state,
    pet,
    grade_level(x=1:3),
    died,    
    normal(name="Count"),
    date_stamp)
  )
  
  
  
}

patients <- random_patients(100)
patients$Height <- ifelse(patients$Sex == "Male",rnorm(n=100,mean=175.3,sd=6),rnorm(n=100,161.9,sd=3))
patients$Weight <- ifelse(patients$Sex == "Male",rnorm(n=100,mean=84,sd=6),rnorm(n=100,69,sd=3))
patients$Weight <- paste0(patients$Weight, "kg")
```

1. Change some values of `Smokes` to yes/no

```{r}

rand_rows <- sample(1:100,10)
patients$Smokes[rand_rows] <- sample(c("Yes","No"),10,replace=TRUE)
```

2. Put some zeros in the `Age` column

```{r}

rand_rows <- sample(1:100,10)
patients$Age[rand_rows] <- 0
```

```{r}
patients$Height <- as.character(patients$Height)
patients$Height <- paste0(patients$Height,"cm")
```


3. Add whitespace to Male / Female columns

```{r}
rand_rows <- sample(1:100,10)
patients$Sex <- as.character(patients$Sex)
patients$Sex[rand_rows] <- paste0(" ", patients$Sex[rand_rows])
```

4. upper-case convert some of the pets

```{r}
rand_rows <- sample(1:100,10)
patients$Pet <- as.character(patients$Pet)
patients$Pet[rand_rows] <- toupper(patients$Pet[rand_rows])
```

5. convert some of the `None` pets to NA or NULL

```{r}
patients$Race <- as.character(patients$Race)
error_row <- sample(which(patients$Pet != "None"),1)
patients$Race[error_row] <- patients$Pet[error_row]

no_pet <- which(patients$Pet == "None")
patients$Pet[sample(no_pet,2)] <- "NA"
patients$Pet[sample(no_pet,3)] <- "NULL"
  
```


6. Make some of the grades '99'

```{r}
rand_rows <- sample(1:100,7)
patients$Grade_Level <- as.numeric(patients$Grade_Level)
patients$Grade_Level[rand_rows] <- 99
```

7. convert IDs to numbers

```{r}
patients$ID <- as.numeric(patients$ID)
```



```{r}
head(patients)
write.table(patients, file="patient-data.txt",sep="\t",row.names = FALSE)
```

Generate some random gene measurements

```{r}
library(stringr)
counts <- matrix(rnorm(820),ncol=82)
colnames(counts) <- str_pad(sample(1:100,82),pad="0",width=3)

counts <- data.frame(GeneID = LETTERS[1:10],counts)
write.table(counts, file="gene-counts.txt",sep="\t",quote=FALSE)
```