Homework answer key - Introduction to R practice

Creating vectors/factors and dataframes

  1. We are performing RNA-Seq on cancer samples being treated with three different types of treatment (A, B, and P). You have 12 samples total, with 4 replicates per treatment. Write the R code you would use to construct your metadata table as described below.
    • Create the vectors/factors for each column (Hint: you can type out each vector/factor, or if you want the process go faster try exploring the rep() function).
       sex <- c("M", "F",...) # saved vectors/factors as variables and used c() or rep() function to create
    • Put them together into a dataframe called meta.
      meta <- data.frame(sex, stage, treatment, myc) # used data.frame() to create the table
    • Use the rownames() function to assign row names to the dataframe (Hint: you can type out the row names as a vector, or if you want the process go faster try exploring the paste() function).
      rownames(meta) <- c("sample1", "sample2",... , "sample12") # or use:
     
  rownames(meta) <- paste("sample12", 1:12, sep="")

    Your finished metadata table should have information for the variables sex, stage, treatment, and myc levels:

      sex stage treatment myc
    sample1 M I A 2343
    sample2 F II A 457
    sample3 M II A 4593
    sample4 F I A 9035
    sample5 M II B 3450
    sample6 F II B 3524
    sample7 M I B 958
    sample8 F II B 1053
    sample9 M II P 8674
    sample10 F I P 3424
    sample11 M II P 463
    sample12 F II P 5105

Subsetting vectors/factors and dataframes

  1. Using the meta data frame from question #1, write out the R code you would use to perform the following operations (questions DO NOT build upon each other):

    • return only the treatment and sex columns using []:
      meta[ , c(1,3)]
    • return the treatment values for samples 5, 7, 9, and 10 using []:
      meta[c(5,7,9,10), 3]
    • use filter() to return all data for those samples receiving treatment P:
      filter(meta, treatment == "P")
    • use filter()/select() to return only the stage and treatment data for those samples with myc > 5000:
      filter(meta, myc > 5000) %>% select(stage, treatment)
    • remove the treatment column from the dataset using []:
      meta[, -3]
    • remove samples 7, 8 and 9 from the dataset using []:
      meta[-7:-9, ]
    • keep only samples 1-6 using []:
      meta [1:6, ]
    • add a column called pre_treatment to the beginning of the dataframe with the values T, F, F, F, T, T, F, T, F, F, T, T (Hint: use cbind()):
      pre_treatment <- c(T, F, F, F, T, T, F, T, F, F, T, T)
     
  cbind(pre_treatment, meta)
    • change the names of the columns to: “A”, “B”, “C”, “D”:
      colnames(meta) <- c("A", "B", "C", "D")

Extracting components from lists

  1. Create a new list, list_hw with three components, the glengths vector, the dataframe df, and number value. Use this list to answer the questions below . list_hw has the following structure (NOTE: the components of this list are not currently named):

       [[1]]
   [1]   4.6  3000.0 50000.0 

   [[2]]
          species  glengths 
     1    ecoli    4.6
     2    human    3000.0
     3    corn     50000.0

   [[3]]
   [1] 8

Write out the R code you would use to perform the following operations (questions DO NOT build upon each other):

 list_hw[[2]]

 list_hw[[1]][3]

 list_hw[[2]][2, 1]

 names(list_hw) <- c("genome_lengths","genomes","record")
 
 list_hw$record

Creating figures with ggplot2

plot_image

  1. Create the same plot as above using ggplot2 using the provided metadata and counts datasets. The metadata table describes an experiment that you have setup for RNA-seq analysis, while the associated count matrix gives the normalized counts for each sample for every gene. Download the count matrix and metadata using the links provided.

    Follow the instructions below to build your plot. Write the code you used and provide the final image.

    • Read in the metadata file using: meta <- read.delim("Mov10_full_meta.txt", sep="\t", row.names=1)

    • Read in the count matrix file using: data <- read.delim("normalized_counts.txt", sep="\t", row.names=1)

    • Create a vector called expression that contains the normalized count values from the row in data that corresponds to the MOV10 gene.

      expression <- data["MOV10", ]
    • Check the class of this expression vector. data.frame

    Then, will need to convert this to a numeric vector using as.numeric(expression)

      class(expression)
     
  expression <- as.numeric(expression)
     
  class(expression)
    • Bind that vector to your metadata data frame (meta) and call the new data frame df.
      df <- cbind(meta, expression) #or
     
  df <- data.frame(meta, expression)

    • Create a ggplot by constructing the plot line by line:

      • Initialize a ggplot with your df as input.

      • Add the geom_jitter() geometric object with the required aesthetics

      • Color the points based on sampletype

      • Add the theme_bw() layer

      • Add the title “Expression of MOV10” to the plot

      • Change the x-axis label to be blank

      • Change the y-axis label to “Normalized counts”

      • Using theme() change the following properties of the plot:

        • Remove the legend (Hint: use ?theme help and scroll down to legend.position)

        • Change the plot title size to 1.5x the default and center align

        • Change the axis title to 1.5x the default size

        • Change the size of the axis text only on the y-axis to 1.25x the default size

        • Rotate the x-axis text to 45 degrees using axis.text.x=element_text(angle=45, hjust=1)

         ggplot(df) +
     geom_jitter(aes(x= sampletype, y= expression, color = sampletype)) +
     theme_bw() +
     ggtitle("Expression of MOV10") +
     xlab(NULL) +
     ylab("Normalized counts") +
     theme(legend.position = "none",
           plot.title=element_text(hjust=0.5, size=rel(1.5)),
           axis.text=element_text(size=rel(1.25)),
           axis.title=element_text(size=rel(1.5)),
           axis.text.x=element_text(angle=45, hjust=1))

    plot_image