snps_roc_analysis

library(tidyverse)

## Loading tidyverse: ggplot2
## Loading tidyverse: tibble
## Loading tidyverse: tidyr
## Loading tidyverse: readr
## Loading tidyverse: purrr
## Loading tidyverse: dplyr

## Conflicts with tidy packages ----------------------------------------------

## filter(): dplyr, stats
## lag():    dplyr, stats

data = read.table("master_output.txt", header=T)

data %>% ggplot(aes(x=sn,y=ppv, col=as.factor(rna_cov))) + geom_line() +
  ggtitle("Precision-Recall plot")

data = data %>% mutate(fdr=1-ppv)
data %>% ggplot(aes(x=fdr,y=sn, col=as.factor(rna_cov))) + 
  geom_line() +
  geom_point() +
  ggtitle("fdr vs. sn")

data %>% ggplot(aes(x=fp,y=tp, col=as.factor(rna_cov))) + 
  geom_line() +
  geom_point() +
  ggtitle("tp ~ fp")

## add false positive rate
data = data %>% mutate(tn = 3.2e9-(tp+fn+fp)) # size of human genome = 3.2e9
data = data %>% mutate(fpr = fp / (fp+tn))

# plot actual ROC

data %>% ggplot(aes(x=fpr, y=sn, col=as.factor(rna_cov))) + 
  geom_line() +
  geom_point() +
  ggtitle("ROC")