--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/5_per_base_sequence_content.Rmd	Tue Aug 08 10:14:46 2017 -0400
@@ -0,0 +1,45 @@
+---
+title: "Per Base Sequence Content"
+output: html_document
+---
+
+```{r setup, include=FALSE, warning=FALSE, message=FALSE}
+knitr::opts_chunk$set(echo = ECHO)
+```
+
+## Per Base Sequence Content
+
+```{r}
+PBSC_df = data.frame()
+PBSC_file_paths = read.csv('PBSC_file_paths.txt',
+                           header = TRUE, stringsAsFactors = FALSE)
+for(i in 1:nrow(PBSC_file_paths)) {
+  # file_path = paste0('REPORT_OUTPUT_DIR/', PBSC_file_paths[i,2])
+  file_path = PBSC_file_paths[i,2]
+  pbsc_df = read.csv(file_path,
+                     sep='\t', header=TRUE, stringsAsFactors = FALSE) %>%
+    mutate(Base1=as.numeric(str_split_fixed(X.Base, '-', 2)[,1]),
+           Base2=as.numeric(str_split_fixed(X.Base, '-', 2)[,2])) %>%
+  (function (df) {
+    df1 = select(df, -Base2)
+    df2 = select(df, -Base1) %>% filter(Base2 != '')
+    colnames(df1) = c(colnames(df1)[1:5], 'Base')
+    colnames(df2) = c(colnames(df2)[1:5], 'Base')
+    res = rbind(df1, df2) %>% arrange(Base)
+    return(res)
+  })
+  pbsc_df$sample_id = rep(PBSC_file_paths[i,1], nrow(pbsc_df))
+  PBSC_df = rbind(PBSC_df, pbsc_df)
+}
+```
+
+
+```{r out.width="100%"}
+PBSC_df_2 = select(PBSC_df, -X.Base) %>%
+  melt(id = c('Base', 'sample_id'), value.name = 'base_percentage')
+p = ggplot(data = PBSC_df_2, aes(x = Base, y = base_percentage, group = variable, color = variable)) +
+  geom_line() +
+  facet_wrap(~ sample_id)
+ggplotly(p)
+```
+