Matrices

 

These exercises are about the matrices sections of Introduction to R.

Exercise 1

• Create a matrix containing information on gene names Gene_1, Gene_2, Gene_3,Gene_4 expression values 1000, 3000, 10000, 12000 and gene lengths 100, 3000, 200, 1000

geneNames <- c("Gene_1", "Gene_2", "Gene_3","Gene_4")
expression <- c(1000, 3000, 10000, 12000)
geneLengths <- c(100, 3000, 200, 1000)
geneMatrix <- cbind(expression,geneLengths)
rownames(geneMatrix) <- geneNames
geneMatrix

##        expression geneLengths
## Gene_1       1000         100
## Gene_2       3000        3000
## Gene_3      10000         200
## Gene_4      12000        1000

• Update the matrix to contain the expression over the gene length for all genes (Length normalised expression) as final column.

HINT - We calculated LNE before in vectors exercises’ bonus question

lne <- geneMatrix[,"expression"]/geneMatrix[,"geneLengths"]
geneMatrix <- cbind(geneMatrix,lne)
geneMatrix

##        expression geneLengths lne
## Gene_1       1000         100  10
## Gene_2       3000        3000   1
## Gene_3      10000         200  50
## Gene_4      12000        1000  12

• Create a smaller matrix containing genes longer than 200

smallGeneMatrix <- geneMatrix[geneMatrix[,"geneLengths"] > 200,]
smallGeneMatrix

##        expression geneLengths lne
## Gene_2       3000        3000   1
## Gene_4      12000        1000  12

• Create a smaller matrix with just expression and lne columns containing genes longer than 200 and expression greater than 300.

smallGeneMatrix <- geneMatrix[geneMatrix[,"geneLengths"] > 200 & geneMatrix[,"expression"] > 300,c("expression","lne")]
smallGeneMatrix

##        expression lne
## Gene_2       3000   1
## Gene_4      12000  12

• Add an additional column to the orginal matrix which contains some GO terms associated with these genes: “Positively Regulates Transcription”, “Negatively Regulates Transcription”, “Positively Regulates Transcription”, “Regulates Autophagy”

GOterms <- c("Positively Regulates Transcription", "Negatively Regulates Transcription", "Positively Regulates Transcription", "Regulates Autophagy")
geneMatrix_GO <- cbind(geneMatrix,GOterms)
geneMatrix_GO

##        expression geneLengths lne  GOterms                             
## Gene_1 "1000"     "100"       "10" "Positively Regulates Transcription"
## Gene_2 "3000"     "3000"      "1"  "Negatively Regulates Transcription"
## Gene_3 "10000"    "200"       "50" "Positively Regulates Transcription"
## Gene_4 "12000"    "1000"      "12" "Regulates Autophagy"

• What is the data type? Use the class(),is.numeric() and is.character() functions to investigate. Compare this to the original matrix.

class(geneMatrix_GO)

## [1] "matrix" "array"

is.numeric(geneMatrix_GO)

## [1] FALSE

is.character(geneMatrix_GO)

## [1] TRUE

class(geneMatrix)

## [1] "matrix" "array"

is.numeric(geneMatrix)

## [1] TRUE

is.character(geneMatrix)

## [1] FALSE

• Subset the original matrix to only genes that are involved in Transcription. Make sure the result is a numeric matrix.

geneMatrix_transcription <- geneMatrix[grepl("transcription",GOterms, ignore.case = T),]

geneMatrix_transcription

##        expression geneLengths lne
## Gene_1       1000         100  10
## Gene_2       3000        3000   1
## Gene_3      10000         200  50

is.numeric(geneMatrix_transcription)

## [1] TRUE

Bonus Question

- Calculate the sum of expression and length columns for only genes with length > 100.

expressionSum <- sum(geneMatrix[geneMatrix[,"geneLengths"] > 100,"expression"])
geneLengthSum <- sum(geneMatrix[geneMatrix[,"geneLengths"] > 100,"geneLengths"])
#OR
expressionAndGeneLengthSum <- colSums(geneMatrix[geneMatrix[,"geneLengths"] > 100,c("expression","geneLengths")])
expressionAndGeneLengthSum

##  expression geneLengths 
##       25000        4200