1. Open a R session by clicking the R icon.

2. Install Bioconductor in R. (The laptop in the classroom already installed Bioconductor).

   #install bioconductor from R
   source("http://bioconductor.org/biocLite.R")

3. Install affy package in R via Bioconductor. (The laptop in the classroom already installed Bioconductor).

   #install affy package
   biocLite("affy")

4. Install samr package in R via Bioconductor. (The laptop in the classroom already installed Bioconductor).

   #install samr package
   biocLite("samr")

5. Load the affy package.

   #loading affy library
   library(affy)

6. Significance Analysis of Microarrays (SAM) in R is implemented as samr package. Load the samr package.

   #loading samr library
   library(samr)

7. Create a temporary file exprsFile to hold gene expression.

   #create a temporary file
   exprsFile <- tempfile()

8. Read in class 1 samples.

   #reading class 1 samples from file "CLASS1"
   file1 <- read.table("CLASS1", sep = "\t")

9. Check file1.

   #check what is being read in as file1
   head(file1)

10. Read in class 2 samples.

    #reading class 2 samples from file "CLASS2"
    file2 <- read.table("CLASS2", sep = "\t")

11. Check file2.

    #check what is being read in as file2
    head(file2)

12. Assign class 1 samples as "1" and class 2 samples as "2" and store these values in Response.

    #Assign class labels to the column in sequential order accoriding to the number of rows in file1 and file2.
    #samples in file1 will get the label "1"
    #samples in file2 will get the label "2"
    Response <- c(rep(1,nrow(file1)), rep(2,nrow(file2)))

13. Check Response.

    #check what is being assigned as Response
    Response

14. Read in gene expression file AC.txt and assign it as a matrix in file.

    #Read in gene expression profiles from file known as "AC.txt" and assign it to file
    file <- as.matrix(read.table("AC.txt", header = TRUE, row.names = 1))

15. Check the file.

    #check what is being read in as file
    head(file)

16. Create exprsFile

    #write "file" to "exprsFile" to create a R object
    write.table(file, exprsFile, quote=F, sep="\t", row.names=TRUE, col.names=TRUE)

17. Create an affy gene expression object to be used in samr.

    #create an affy expression object to be used in samr
    collapsed <- readExpressionSet(exprsFile, sep="\t", annotation = "hgu133plus2")

18. Check what is in the R object collapsed.

    #check the "collapsed" object
    collapsed

19. Create the samr data file sam.data as a list.

    #create the samr data file
    #x = gene expression profiles obtained from exprs(collapsed)
    #y = class label obtained from "Response"
    #geneid = first column from exprs(collapsed)
    #genenames = row names from exprs(collapsed)
    #logged2 = is the gene expression in log2 format? TRUE or FALSE
    sam.data <- list(x=exprs(collapsed), y=Response, geneid = row.names(exprs(collapsed)), genenames = row.names(exprs(collapsed)), logged2=TRUE)

20. Check the list of data stored in sam.data.

    #check what is being read in as sam.data
    sam.data

    #you will see five lists
    #x, y, geneid, genenames, logged2

21. Run samr using data in the sam.data.

    #Run samr using data in sam.data
    #resp.type = "Two class unpaired"
    #nperms = number of permutations
    #random.seed = set a random number as seed
    samr.obj <- samr(sam.data, resp.type = "Two class unpaired", nperms = 100, random.seed = 12345)
    

22. Compute the delta values and store in delta.table.

    #Compute the delta values and store in delta.table
    delta.table <- samr.compute.delta.table(samr.obj)

23. Check the delta.table.

    #Print out delta.table
    delta.table

24. Select a FDR cut-off and assign the corresponding delta values to delta.

    #select a FDR cut-off and assign the corresponding delta values to delta (Example here, delta = 0.05)
    delta <- 0.05

25. Plot the SAM graph using the cut-off with the delta value using samr.plot.

    #Plot the graph using the cut-off with the delta value
    samr.plot(samr.obj,delta)

    You can plot SAM graph using different delta values, and see the effects on the "dashed lines".

26. List out the significant genes at the cut-off delta value using samr.compute.siggenes.table and write in siggenes.table.

    #List out the significant genes at the cut-off delta in siggenes.table
    siggenes.table <- samr.compute.siggenes.table(samr.obj, delta, sam.data, delta.table)

27. Examine the significant genes in siggenes.table.

    #examine the significant genes in siggenes.table
    siggenes.table

28. Assign the list of up-regulated genes as myupgenes.

    #assign up genes to myupgenes
    myupgenes <- siggenes.table$genes.up

29. Check what is in myupgenes.

    #check what is in myupgenes
    myupgenes

30. Print the list of myupgenes into a tab delimited file myupgenes.txt.

    #print myupgenes into a file
    write.table(myupgenes, file="myupgenes.txt", sep="\t")

31. Assign the list of up-regulated genes as mydowngenes.

    #assign down genes to mydowngenes
    mydowngenes <- siggenes.table$genes.lo

32. Check what is in mydowngenes.

    #check what is in mydowngenes
    mydowngenes

33. Print the list of mydowngenes into a tab delimited file mydowngenes.txt.

    #print mydowngenes into a file
    write.table(mydowngenes, file="mydowngenes.txt", sep="\t")

Related R script

Assignment #3

Your tasks are:

1. Modify the R script to read in these files.
2. List out the significant genes at FDR = 0.05, find out the delta value.
3. Plot the SAM graph at this delta value.
4. How many up-regulated genes? Print out the up-regulated genes.
5. How many down-regulated genes? Print out the down-regulated genes.
6. Compare your gene list with the published gene list. How many genes are overlap?