Updated 2021/04/23

The Book of R: chapter 3.4 (arrays), 5.1 (lists), 5.2 (data frames)

Module 2 review

  • Variable types: vector, matrix
  • Vector
    • x = c(1.2,2,54,6)
    • textvec[1:2]
  • Matrix
    • firstmat[1,2]
    • firstmat[1:2,2:3]

Exercise: String commands (10 minutes)

We used this code to reverse a string

paste(rev(strsplit(DNA, NULL)[[1]]), collapse="")

Figure out how this works by trying each of the nested functions, starting from strsplit(DNA, NULL)

Note: [[1]] extracts elements within a list (which is a type of vector that can contain other lists)

Exercise solution (5 minutes)

DNA <- "GATTACA"
str(strsplit(DNA,"A"))
## List of 1
##  $ : chr [1:3] "G" "TT" "C"
str(strsplit(DNA,"A")[1])
## List of 1
##  $ : chr [1:3] "G" "TT" "C"
str(strsplit(DNA,"A")[[1]])
##  chr [1:3] "G" "TT" "C"
str(strsplit(DNA,"A")[[1]][1])
##  chr "G"
strsplit(DNA,"")
## [[1]]
## [1] "G" "A" "T" "T" "A" "C" "A"
identical(strsplit(DNA,NULL),strsplit(DNA,""))
## [1] TRUE
length(NULL)
## [1] 0
rev(strsplit(DNA,"")) #nope!
## [[1]]
## [1] "G" "A" "T" "T" "A" "C" "A"
rev(strsplit(DNA,"")[[1]])
## [1] "A" "C" "A" "T" "T" "A" "G"
paste(rev(strsplit(DNA,"")[[1]]),sep="&")
## [1] "A" "C" "A" "T" "T" "A" "G"
paste(rev(strsplit(DNA,"")[[1]]),collapse="&")
## [1] "A&C&A&T&T&A&G"
paste(rev(strsplit(DNA,"")[[1]]),collapse="")
## [1] "ACATTAG"

Lists

Lists: creating

  • Lists are vectors that contain a mix of variable types
  • Create a list using the list() function
numericVector = c(1, 3, 5)
numericVector
## [1] 1 3 5
#Note: numericVector is just a variable name - we call it numericVector because we define it here to contain numbers, but it could be called any variable name that is allowed.

stringVector = c("abc", "def", "ghi")
stringVector
## [1] "abc" "def" "ghi"
logicVector = c(T, F, T, F)
logicVector
## [1]  TRUE FALSE  TRUE FALSE
listVariable = list(numericVector, stringVector, logicVector, 77)
listVariable
## [[1]]
## [1] 1 3 5
## 
## [[2]]
## [1] "abc" "def" "ghi"
## 
## [[3]]
## [1]  TRUE FALSE  TRUE FALSE
## 
## [[4]]
## [1] 77

Lists: accessing

listVariable
## [[1]]
## [1] 1 3 5
## 
## [[2]]
## [1] "abc" "def" "ghi"
## 
## [[3]]
## [1]  TRUE FALSE  TRUE FALSE
## 
## [[4]]
## [1] 77
listVariable[1]  #access first list element
## [[1]]
## [1] 1 3 5
listVariable[[1]] #access content of first element
## [1] 1 3 5
listVariable[[1]][2] 
## [1] 3
listVariable[[2]][1] = "ACGT"
listVariable[[2]]
## [1] "ACGT" "def"  "ghi"
listVariable[c(2, 4)]
## [[1]]
## [1] "ACGT" "def"  "ghi" 
## 
## [[2]]
## [1] 77

Lists: named elements

listVariable = list(numbers=numericVector, strings=stringVector, logic=logicVector, myNum=77)
listVariable
## $numbers
## [1] 1 3 5
## 
## $strings
## [1] "abc" "def" "ghi"
## 
## $logic
## [1]  TRUE FALSE  TRUE FALSE
## 
## $myNum
## [1] 77
listVariable["numbers"]
## $numbers
## [1] 1 3 5
listVariable["myNum"]
## $myNum
## [1] 77
#These are equivalent for named elements
listVariable[[1]]
## [1] 1 3 5
listVariable[["numbers"]]
## [1] 1 3 5
listVariable$numbers  #easiest access method
## [1] 1 3 5

Exercise: Lists (10 minutes)

  • Create a list containing a vector of odd numbers from 1 to 100
    • Hint: you can use the seq() command
  • Calculate the sum of this list
    • Hint: use the sum() command
  • Repeat this, but name the vector and access its values from the list using its name

Exercise solution (5 minutes)

mylist <- list(seq(1,10))
myvec <- 1:10
identical(mylist[[1]],myvec)
## [1] TRUE
mylist <- list(odds=seq(1,100,by=2),
               evens=seq(2,100,by=2),
               someothersilliness=c("why","are","we","here","?"))

sum(mylist[[1]])
## [1] 2500
identical(sum(mylist[[1]]),sum(mylist$odds))
## [1] TRUE
sum(mylist[["odds"]])
## [1] 2500
paste(mylist[["someothersilliness"]],collapse=" ")
## [1] "why are we here ?"

Arrays

Arrays: creating

  • Like matrix, but more than two dimensions
  • Only one data type (like vector, matrix)
  • Create an array using the array() function
vector1 = c(1,3,5)
vector1
## [1] 1 3 5
vector2 = c(20,21,22,23,24,25)
vector2
## [1] 20 21 22 23 24 25
#Create two 3x3 matrices
arrayVariable = array(c(vector1,vector2),dim = c(3,3,2))
arrayVariable
## , , 1
## 
##      [,1] [,2] [,3]
## [1,]    1   20   23
## [2,]    3   21   24
## [3,]    5   22   25
## 
## , , 2
## 
##      [,1] [,2] [,3]
## [1,]    1   20   23
## [2,]    3   21   24
## [3,]    5   22   25

Arrays: accessing

arrayVariable
## , , 1
## 
##      [,1] [,2] [,3]
## [1,]    1   20   23
## [2,]    3   21   24
## [3,]    5   22   25
## 
## , , 2
## 
##      [,1] [,2] [,3]
## [1,]    1   20   23
## [2,]    3   21   24
## [3,]    5   22   25
#get matrix 1
arrayVariable[,,1]
##      [,1] [,2] [,3]
## [1,]    1   20   23
## [2,]    3   21   24
## [3,]    5   22   25
#get 3rd row of matrix 2
arrayVariable[3,,2]
## [1]  5 22 25
#get element in row 1, column 3 in matrix 1
arrayVariable[1,3,1]
## [1] 23
#get matrix 2
arrayVariable[,,2]
##      [,1] [,2] [,3]
## [1,]    1   20   23
## [2,]    3   21   24
## [3,]    5   22   25

Arrays: named elements

#Create the names
rowNames = c("row1","row2","row3")
rowNames
## [1] "row1" "row2" "row3"
columnNames = c("col1","col2","col3")
columnNames
## [1] "col1" "col2" "col3"
matrixNames = c("matrix1","matrix2")
matrixNames
## [1] "matrix1" "matrix2"
arrayVariable = array(c(vector1,vector2), dim = c(3,3,2), dimnames = list(rowNames, columnNames, matrixNames))
arrayVariable
## , , matrix1
## 
##      col1 col2 col3
## row1    1   20   23
## row2    3   21   24
## row3    5   22   25
## 
## , , matrix2
## 
##      col1 col2 col3
## row1    1   20   23
## row2    3   21   24
## row3    5   22   25
#Access by names:
arrayVariable[,,"matrix1"]
##      col1 col2 col3
## row1    1   20   23
## row2    3   21   24
## row3    5   22   25
arrayVariable["row2","col1","matrix1"]
## [1] 3

Does each matrix in an array need to be the same shape?

array()
## [1] NA

Data frames

Data frames: creating

  • Used to store data tables
  • List of equal sized vectors
  • Create a data.frame using the data.frame() function
name = c("patient1", "patient2", "patient3")
name
## [1] "patient1" "patient2" "patient3"
age = c(46, 49, 50)
age
## [1] 46 49 50
smoker = c(T, F, T)
smoker
## [1]  TRUE FALSE  TRUE
patientRecords = data.frame(name, age, smoker)
patientRecords

Data frames: accessing

patientRecords
#Just like matrices
patientRecords[1,] #row 1
patientRecords[,1] #column 1
## [1] "patient1" "patient2" "patient3"
patientRecords[1,2] #element at row 1, column 2
## [1] 46
patientRecords$age
## [1] 46 49 50

Exercise: Data frames (10 minutes)

Read the “gene_condition_source_id.txt” file used in Assignment 1 into a data frame variable (see assignment 1 for the command)

Look at the top few rows using the head() command

How many rows and columns does the file have? Try the str() command. Also try nrow() and ncol()

Use names() to find the column names

Exercise solution (5 minutes)

wc -l gene_condition_source_id.txt
## 11062 gene_condition_source_id.txt
list.files(path=".") # "." == the current working directory
##  [1] "_site.yml"                    "aaoutput.txt"                
##  [3] "ASCII-Table-wide.svg"         "assignment2_sequence.txt"    
##  [5] "column.txt"                   "docs"                        
##  [7] "fileOutput.png"               "gene_condition_source_id.txt"
##  [9] "GSE13924_RAW.tar"             "GSM350771.CEL.gz"            
## [11] "GSM350772.CEL.gz"             "index.html"                  
## [13] "index.md"                     "mod_me.html"                 
## [15] "mod_me.Rmd"                   "mod_me_files"                
## [17] "mod1.html"                    "mod1.Rmd"                    
## [19] "mod2.html"                    "mod2.Rmd"                    
## [21] "mod2.utf8.md"                 "mod2_files"                  
## [23] "mod3.Rmd"                     "mod4.Rmd"                    
## [25] "mod5.Rmd"                     "mod5exercise.fa"             
## [27] "mod6.Rmd"                     "mod7.knit.md"                
## [29] "mod7.Rmd"                     "mod8.Rmd"                    
## [31] "mod9.Rmd"                     "mod9.utf8.md"                
## [33] "mydata.txt"                   "p4b.Rproj"                   
## [35] "README.md"                    "recipebookTOC.png"           
## [37] "row.txt"                      "RStudioOptions.png"          
## [39] "site_libs"                    "sum.R"                       
## [41] "test.txt"                     "thisisapdf.pdf"              
## [43] "thisisnotapdf.png"            "yeast_properties.txt"
db <- read.table("gene_condition_source_id.txt",
                 header=T,sep="\t",quote="",comment.char="",
                 colClasses="character")
nrow(db)
## [1] 11061
identical(colnames(db),names(db))
## [1] TRUE
str(db)
## 'data.frame':    11061 obs. of  9 variables:
##  $ X.GeneID       : chr  "2" "2" "144568" "53947" ...
##  $ AssociatedGenes: chr  "A2M" "A2M" "A2ML1" "A4GALT" ...
##  $ RelatedGenes   : chr  "" "" "" "" ...
##  $ ConceptID      : chr  "C0002395" "C0002395" "C1833692" "C0599990" ...
##  $ DiseaseName    : chr  "Alzheimer disease" "Alzheimer disease" "Otitis media, susceptibility to" "p phenotype" ...
##  $ SourceName     : chr  "MONDO" "Human Phenotype Ontology" "MONDO" "NCBI curation" ...
##  $ SourceID       : chr  "MONDO:0004975" "HP:0002511" "MONDO:0008162" "" ...
##  $ DiseaseMIM     : chr  "104300" "104300" "166760" "111400" ...
##  $ LastUpdated    : chr  "Feb 19 2020" "Feb 19 2020" "Feb 16 2016" "Feb 16 2016" ...
rownames(db)[1:10]
##  [1] "1"  "2"  "3"  "4"  "5"  "6"  "7"  "8"  "9"  "10"

Assignment 2 hints:

I expect a tarball containing a single folder with any input data required to run your scripts (ie. the FASTA file), any output files generated by your scripts, and your scripts themselves! For this week, there is no report needed. Your scripts should answer any questions themselves (ie. ‘stringnumber_practice.R’ should be filled in with all the code needed to generate solutions for each question).