#This tutorial focuses on the use of dplyr for data manipulation

Step 1: Read in files from yeast expression experiment.

myconditions <- read.csv("conditions_annotation.csv")
myexpression <- read.csv("SC_expression.csv")

Step 2: Limit your search to one condition (pick a new condition for your script)

myfilter <- myconditions[grepl("wildtype",myconditions$primary),]

Step 3: Select expression data from only the annotations/column 1 of filtered list

myexpression2 <-
  myexpression%>%
  select(myfilter$ID)

Step 4: Make data tidy (one observation per row) to use ggplot/dplyr

tidyExpression <- myexpression2 %>% pivot_longer(cols = everything())

Step 5: Create summary of expression value stats in an easy to read tibble (formatted dataframe)

by_treatment <- tidyExpression %>%
  group_by(name)

by_treatment  %>%
  summarise_all(list(mean = mean, median  = median, n=length))
## # A tibble: 4 × 4
##   name    mean median     n
##   <chr>  <dbl>  <dbl> <int>
## 1 AFIQBR  165.   2.13  6071
## 2 AFIQCI  165.   3.31  6071
## 3 QCAQFI  165.   5.36  6071
## 4 QCAQFQ  165.   7.37  6071

Step 5: Plot violin plot of expression data

ggplot(tidyExpression, aes(x=name,y=log(value))) + 
  geom_violin()
## Warning: Removed 1098 rows containing non-finite outside the scale range
## (`stat_ydensity()`).

Step 6: Pick a new condition and regenerate the tibble and plots for this condition.