Introduction to R. Dr. Emile R. Chimusa Department of Integrative Biomedical Sciences University of Cape Town. May 9, 2016

Transcription

1 Introduction to R Dr. Emile R. Chimusa Department of Integrative Biomedical Sciences University of Cape Town May 9,

2 CONTENTS CONTENTS Contents 1 Getting started in R-RStudio Getting R and RStudio Started on your PC R Packages Key Things to Know About R Getting Help R as a Calculator Assignment, Object names and Basic data Types Computing on data vector 8 3 Functions and Expressions Conditions Statements and Loops in R Data Manupilations Load and Read R data Using Your Own Data Basic Data Operations Basic Operations with Matrices Running scripts 24 6 Important R Tips 24 7 Tutorial 25 2

3 Getting started in R-RStudio CONTENTS 1 Getting started in R-RStudio 1.1 Getting R and RStudio Started on your PC R is a system for statistical computation and graphics. RStudio is an alternative graphical interface to R. We use R-RStudio for several reasons: (1) R is open-source and freely available for Mac, PC, and Linux machine. (2) R is user-extensible and user extensions can easily be made available to others. (3) It is the package of choice for many statisticians and those who use statistics frequently. (4) R is becoming very popular with statisticians and scientists, especially in certain subdisciplines,like genetics. (5) It is gaining new features every day. New statistical methods are often available first in R. You can downloaded R freely from (Windows, Linux or MacOS). Follow the instructions and after a little patience you should be able to start R after which a screen is opened with the prompt >. It is also possible to download RStudio server and set up your own server or RStudio desktop for stand-alone processing. Once you have logged in to an RStudio server, you will see something like in Figure below. 3

4 1.2 R Packages CONTENTS Notice that RStudio divides its world into four panels. Several of the panels are further subdivided into multiple tabs. RStudio offers the user some control over which panels are located where and which tabs are in which panels, so you initial configuration might not be exactly like the one illustrated here. The console panel is where we type commands that R will execute. 1.2 R Packages All functionalities of R are well-organized in so-called packages and R provides many more features through a (large) number of packages. To use a package, it must be installed (one time), and loaded (each session). A number of packages are already available in RStudio. The packages tab in RStudio will show you the list of installed packages and indicate which of these are loaded. Alternatively, use the function library() to see which packages are currently installed on your operating system. 4

5 1.3 Key Things to Know About R CONTENTS You can install other packages by clicking on the Install Package button in RStudio and following the directions or to download a specific package you can use the following command. > rep <-" > #install.packages(c("teachingdemos"),repo=rep,dep=true) From the button Packages at cran.r-project.org it can be seen that R has a huge number of packages available for a wide scale of statistical procedures. 1.3 Key Things to Know About R (1) R is case-sensitive. If you mis-capitalize something in R it won t do what you want. (2) Functions in R use the following syntax: > functionname( argument1, argument2,... ) ˆ The arguments are always surrounded by (round) parentheses and separated by commas. Some functions (like data()) have no required arguments, but you still need the parentheses. ˆ If you type a function name without the parentheses, you will see the code for that function (this probably isn t what you want at this point). (3) TAB completion and arrows can improve typing speed and accuracy. If you begin a command and hit the TAB key, RStudio will show you a list of possible ways to complete the command. If you hit TAB after the opening parenthesis of a function, it will show you the list of arguments it expects. The up and down arrows can be used to retrieve past commands. (4) If you see a + prompt, it means R is waiting for more input. Often this means that you have forgotten a closing parenthesis or made some other syntax error. If you have messed up and just want to get back to the normal plot, hit the escape key and start the command fresh. In this course, we will often use packages from Bioconductor, a very useful open source software project for the analysis and comprehension of genomic data. All these packages are asuming to be installed already from the preamble material. To follow the course it is essential to install Bioconductor on your PC or network. Bioconductor is primarily based on R and can be installed, as follows. > source(" > bioclite() Then to download ALL packages from a repository to your system, to load it, and to make the ALL data (Chiaretti, et. al, 2004) available for usag, you can use the following > bioclite("all") > library(all) > data(all) 5

6 1.4 R as a Calculator CONTENTS Getting Help If something doesn t go quite right, or if you can t remember something, it s good to know where to turn for help. In addition to asking your friends and neighbors, you can use the R help system. To get help on a specific function or data set, simply precede its name with a?: >?plot() If you don t know the exact name of a function, you can give part of the name and R will find all functions that match. Quotation marks are mandatory here. > apropos('hist') # must include quotes. If the above fails, you can do a broader search using help.search(), which will find matches not only in the names of functions and data sets, but also in the documentation for them. >??histogram # any of these will work >??"histogram" >??'histogram' > help.search('histogram') In addition, to obtain an overview of the content of a package use ls(package:stats) or library(help= stats ). 1.4 R as a Calculator R can be used as a calculator. Try typing the following commands in the console panel. > 15.3 * 23.4 [1] > sqrt(16) [1] 4 You can save values to named variables for later reuse. > my_product = 15.3 * 23.4 # save result > my_product # show the result [1] > product < * 23.4 # <- is assignment operator, same as = > product [1]

7 1.5 Assignment, Object names and Basic data Types CONTENTS > 15.3 * > newproduct # -> assigns to the right > newproduct [1] Once variables are defined, they can be referenced with other operators and functions. > 5 * product # half of the product [1] > log(product) # (natural) log of the product [1] > log10(product) # base 10 log of the product [1] > log(product, base=2) # base 2 log of the product [1] The semi-colon can be used to place multiple commands on one line. One frequent use of this is to save and print a value all in one go: > 15.3 * > product; product # save result and show it [1] Assignment, Object names and Basic data Types It is often convenient to assign numbers and values to variables (objects) to be used later. The proper way to assign values to a variable is with the <- operator (with a space on either side). The = symbol works too, but it is recommended by the R masters to reserve = for specifying arguments to functions. > x <- 7*41/pi # don't see the calculated value > x # take a look [1] By choosing a variable name you can use letters, numbers, dots., or underscore characters. You cannot use mathematical operators, and a leading dot may not be followed by a number. Examples of valid names are: x, x2, z.value, and z hat. Objects can be of many types, modes, and classes. At this level, it is not necessary to investigate all of the intricacies of the respective type, but there are some with which you need to become familiar: 7

8 Computing on data vector CONTENTS (1) integer: the values 0, ±1, ±2,...; these are represented exactly by R. (2) double: real numbers (rational and irrational); these numbers are not represented exactly (save integers or fractions with a denominator that is a power of 2). (3) character: elements that are wrapped with pairs of or ; (4) logical: includes TRUE, FALSE, and NA (which are reserved words); the NA stands for not available, i.e., a missing value. You can determine an object s type with the typeof function. In addition to the above, there is the complex data type: > sqrt(-1) # isn't defined > sqrt(-1+0i) # is defined > sqrt(as.complex(-1)) # same thing > (0 + 1i)^2 # should be -1 > typeof((0 + 1i)^2) Note that you can just type (1i) 2 to get the same answer. The NaN stands for not a number ; it is represented internally as double. 2 Computing on data vector All of this time we have been manipulating vectors of length 1. Now let us move to vectors with multiple entries. A data vector is simply a collection of numbers obtained as outcomes from measurements. If you would like to enter the data 74, 31, 95, 61, 76, 34, 23, 54, 96 into R, you may create a data vector with the c function (which is short for concatenate). > my_vector <- c(74,31,95,61,76,34,23,54,96) > my_student <-c("emile","eric","peter","anna") This can be illustrated by a simple example on expression values of a gene. Suppose that gene expression values 1, 1.5, and 1.25 from the persons Eric, Peter, and Anna are available. To store these in a vector we use the concatenate command c(), as follows. > gene1 <- c(1.00,1.50,1.25) > gene_person <-c("eric", 'Peter',"Anna") Now we have created the object gene1 containing three gene expression values. We can compute the sum, mean, and standard deviation of the gene expression values we use the corresponding built-in-functions (We see it later on). In order to compute so-called quantiles of distributions (see later on) or plots of functions (see next sections), we may need to generate sequences of numbers. The easiest way to construct a sequence of numbers is by > 1:10 8

9 Computing on data vector CONTENTS [1] This sequence can also be produced by the function seq, which allows for various sizes of steps to be chosen. For example, we may want to generate numbers between zero and one with step size equal to 0.1. > seq(0,1,0.1) [1] > seq(from=1, to = 5) [1] > x <- seq(from = 2,by =-0.1, length.out =4) Indexing data vectors Sometimes we do not want the whole vector, but just a piece of it. We can access the intermediate parts with the [] operator. Observe (with x defined above) > x <- c(74, 31, 95, 61, 76, 34, 23, 54, 96) > x[1] [1] 74 > x[2:4] [1] > x[c(1, 3, 4, 8)] [1] In addition, the vector LETTERS has the 26 letters of the English alphabet in uppercase and letters has all of them in lowercase. > LETTERS [1] "A" "B" "C" "D" "E" "F" "G" "H" "I" "J" "K" "L" "M" "N" "O" "P" "Q" "R" "S" [20] "T" "U" "V" "W" "X" "Y" "Z" > LETTERS[1:5] [1] "A" "B" "C" "D" "E" We can use the minus sign to specify those elements that we do not want. > x[-c(1, 3, 4, 8)] [1]

10 Functions and Expressions CONTENTS > letters[-(6:24)] [1] "a" "b" "c" "d" "e" "y" "z" Another type of sequence, called a factor. factor is designed to indicate an experimental condition of a measurement or group to which a patient (observation) belongs. When, for example, for each of 7 experimental conditions there are measurements from 6 patients, the corresponding factor can be generated as follows. > factor <- gl(7,6) The 7 conditions are often called levels of a factor. Each of these levels has 6 repeats corresponding to the number of observations (patients) within each level (type of disease). We shall further illustrate the idea of a factor soon because it is very useful for purposes of visualization. 3 Functions and Expressions A function takes arguments as input and returns an object as output. There are functions to do all sorts of things. We show some examples below. > x <- c(74, 31, 95, 61, 76, 34, 23, 54, 96) > sum(x) [1] 544 > length(x) [1] 9 > min(x) # max(x) [1] 23 > mean(x) # sample mean [1] > sd(x) # sample standard deviation [1]

11 Functions and Expressions CONTENTS > plot(x,x) x x By typing the name of the function without any parentheses or arguments, if you are lucky then the code for the entire function will be printed, right there looking at you. For instance, suppose that we would like to see how the intersect function works: > intersect function (x, y) { y <- as.vector(y) unique(y[match(as.vector(x), y, 0L)]) } <bytecode: 0x12947e8> <environment: namespace:base> You can extend the R functions and language by writing your own functions. Bellow is the syntax for designing your own functions in R namefunction <- function(args) {... code... } 11

12 Functions and Expressions CONTENTS Example of your Functions in R: > #Example 1 of functions: > y<- c(3.1,10.5,14,30,15,19) > x<- c(4,12,12,20,16,22) > z<- cbind(x,y) > circle.area <- function(radius) { + area <- pi*radius^2 + return(area) + } > circle.area(4) # calling or using your function [1]

13 Functions and Expressions CONTENTS > #Example 2 of functions: > mystudy <- function(x){ + par(mfrow=c(3,1)) + hist(x[,1]) + hist(x[,2]) + plot(x[,1],x[,2]) + par(mfrow=c(1,1)) + apply(x,2,summary) + } > mystudy(z) x y Min st Qu Median Mean rd Qu Max Histogram of x[, 1] Frequency x[, 1] Histogram of x[, 2] Frequency x[, 2] x[, 2] x[, 1] Figure 1: Multi-plots in one figure. 13

14 3.1 Conditions Statements and Loops in R CONTENTS 3.1 Conditions Statements and Loops in R The if condition and statement syntax if (...condition...) {...code 1... } else {...code 2... } The while loop syntax while (...condition...) {...code...} The for loop syntax for(rank of indices) {...code...} Example 1: If conditions > x <- 10 > y <- 2 > if (y >1){ + x <- 2*x + y <- 2*y + } else{ + x < x <-2*x + } > x [1] 20 > y [1] 4 Example 2: The for loop > cunt <- c(0,0,0,0) > n <- c(2,4,6,4) > for(i in 1:length(n)){ + cunt <- c(cunt,rep(i,n[i])) + } > cunt 14

15 Data Manupilations CONTENTS [1] Example 3: The for and while loops > for (i in 1:10){print(i)} > n<-10 > while (n > 0) { + print(n,"is greater than 0 \n") + n <- n-1 + } 4 Data Manupilations 4.1 Load and Read R data Data analysis involves a large amount of manupilation and cleaning to facilitate downstream data analysis. This section covers basic data manupilation using R default functions. Many packages contain data sets. You can see a list of all data sets in all loaded packages using > data() You can use data sets by simply typing their names. But if you have already used that name for something or need to refresh the data after making some changes you no longer want, you can explicitly load the data using the data() function with the name of the data set you want. > data(iris) Data sets are usually stored in a special structure called a data frame. Data frames have a 2-dimensional structure. (1) Rows correspond to observational units (people, animals, plants, or other objects we are collecting data about). (2) Columns correspond to variables (measurements collected on each observational unit). R comes with some data and ready for to be used. For Example, the iris data frame contains 5 variables measured for each of 150 iris plants (the obervational units). The iris data set is included with the default R installation and located in a package called datasets which is always available. There are several ways we can get some idea about what is in the iris data frame. > str(iris) 'data.frame': 150 obs. of 5 variables: $ Sepal.Length: num $ Sepal.Width : num $ Petal.Length: num $ Petal.Width : num $ Species : Factor w/ 3 levels "setosa","versicolor",..:

16 4.1 Load and Read R data CONTENTS > summary(iris) Sepal.Length Sepal.Width Petal.Length Petal.Width Min. :4.300 Min. :2.000 Min. :1.000 Min. : st Qu.: st Qu.: st Qu.: st Qu.:0.300 Median :5.800 Median :3.000 Median :4.350 Median :1.300 Mean :5.843 Mean :3.057 Mean :3.758 Mean : rd Qu.: rd Qu.: rd Qu.: rd Qu.:1.800 Max. :7.900 Max. :4.400 Max. :6.900 Max. :2.500 Species setosa :50 versicolor:50 virginica :50 > head(iris) Sepal.Length Sepal.Width Petal.Length Petal.Width Species setosa setosa setosa setosa setosa setosa > View(iris) # In interactive mode, you can also try >?iris # to get the documentation about for the data set. Access to an individual variable in a data frame uses the $ operator in the following syntax: dataframe$variable or with(dataframe, variable). For example, either of > iris$sepal.length[1:10] or [1] > with(iris, Sepal.Length)[1:10] [1] The attach() function in R can be used to make objects within dataframes accessible in R with fewer keystrokes, but we strongly discourage its use, as it often leads to name conflicts. 16

17 4.2 Using Your Own Data CONTENTS 4.2 Using Your Own Data RStudio will help you import your own data. To do so use the Import Dataset button in the Workspace tab. You can load data from text files, from the web, or from google spreadsheets. Using read.csv() and read.table() If you are not using RStudio, or if you want to automate the loading of data from files, instead of using the RStudio menus, you can read files using read.csv() or read.table() (for white space delimited files). Now we can load data file class data.csv, it contains 50 observation or indivuals and 13 variables include years, birthmonth, gender, siblings, height, handspan, footlength, breath, armcross, tongue, dice, beans and handed > mytable <- read.csv("class_data.csv", header = TRUE) > head(mytable) years birthmonth gender siblings height handspan footlength breath armcross 1 26 September M right 2 28 August F right 3 26 January M right 4 28 May M left 5 25 M left 6 30 April M right tongue dice beans handed 1 yes yes yes no yes yes Each of these functions also accepts a URL in place of a file name, which provides an easy way to distribute data via the Internet: > web.data <-' > births <- read.table(web.data, header=true) > head(births) # number of live births in the US each day of 1978 date births datenum dayofyear 1 1/1/ /2/ /3/ /4/ /5/ /6/ The mosaic package includes a function called read.file() that uses slightly different default settings and infers whether it should use read.csv(), read.table(), or load() based on the file 17

18 4.3 Basic Data Operations CONTENTS name. load() is used for opening files that store R objects in native format. Using RStudio Server menus (1) Get the file onto the server. Upload (in the Files tab) your csv file to the server, where you can create folders and store files in your personal account. (2) Load the data from the server into your R session. Now import from a text file in the Workspace tab. In either case, be sure to do the following: 1. Choose good variables names. 2. Put your variables names in the first row. 3. Use each subsequent row for one observational unit. 4. Give the resulting data frame a good name. write.csv(mydata, file = "MyData.csv",row.names=FALSE) write.table(mydata, file = "MyData.csv",row.names=FALSE, na="",col.names=false, sep=", Where "MyData" should be a R data frame object. 4.3 Basic Data Operations R has a number of default funtions to deal with variables, below are some of them (1) rbind: combines rows of data. (2) merge: match merges two data frames. (3) dimnames: lists or assigns names of data frames. (4) cbind: combines columns of data. (5) sapply: applies a function to elements of a list. (6) tapply: applies a function to each cell of a ragged array. (7) factor: creates a categorical variable with value labels if desired. (8) table: creates frequency table. (9) head: display first n observations. (10) colmeans: column means. (11) colsums: column sums. 18

19 4.3 Basic Data Operations CONTENTS (11) rowsums: row sums. (12) length: calculates the count of an object uch list, vectors etc. (13) names: list all varaible of data frame. Now let use, the dataset class data.csv to illustrate basic data manupilation in R. Let load it again > gl <- read.csv("class_data.csv",header=t) > attach(gl) > names(gl) [1] "years" "birthmonth" "gender" "siblings" "height" [6] "handspan" "footlength" "breath" "armcross" "tongue" [11] "dice" "beans" "handed" To remove rows from gl with missing data, the R function to check for this is complete.cases() or na.omit() > gl <- gl[complete.cases(gl), ] # or gl <- na.omit(gl) To keep only the observations from gl where the siblings score is 5 or higher, we can do the follow, > gl_sub <- gl[siblings >= 5, ] > nrow(gl_sub);nrow(gl) [1] 17 [1] 44 To separate the data frame gl into two groups, female when tongue is yes and male where dice is equal to 5 or greater than 5 > gl_female <- gl[gender== "F" & tongue == "yes", ] > gl_male <- gl[gender=="m" & dice >= 5,] > length(gl_male);length(gl_female) [1] 13 [1] 13 To use the rbind function when we stack data because we combine rows of data from gl male and gl female, as follow > gl_mf <- rbind(gl_female,gl_male) > nrow(gl_mf) [1] 22 19

20 4.3 Basic Data Operations CONTENTS Let keep only the variables years, birthmonth, footlength and breath from the gel data frame. > gl <- read.csv("class_data.csv",header=t) > gl1.kept <- gl[, c(1, 2, 7, 8)] > names(gl1.kept) [1] "years" "birthmonth" "footlength" "breath" Keeping only the variables year, gender, siblings, handspan, armcross, tongue from the gel data frame. > gl2.kept <- gl[, c(1,3,4,6,9,10)] > names(gl2.kept) [1] "years" "gender" "siblings" "handspan" "armcross" "tongue" To dropping some variables from data.frame use -c() Merge two data frames (gl1.kept and gl2.kept) on a variable (or a list of variables). We use variable year which has the same name in both data sets. Specifying T in the all argument indicates that we want to keep all the observations from each data set rather than only keeping the observations that came from both data sets. > merge_gl.kept <- merge(gl1.kept, gl2.kept, by="years", all=t) > names(merge_gl.kept) [1] "years" "birthmonth" "footlength" "breath" "gender" [6] "siblings" "handspan" "armcross" "tongue" Note: 1. To dropping some variables from data.frame use -c() 2. If the variable that we were merging on had different names in each data frame then we could use the by.x and by.y arguments. In the by.x argument we would list the name of the variable(s) that was in the data frame listed first in the merge function and in the by.y argument we would name the variable(s) that was in the data frame listed second. > gl2.kept <- gl2.kept[,2:4] > merge_gl.kept2 <- merge(gl1.kept, gl2.kept, by.x="years", by.y="gender", all=t) > names(merge_gl.kept2) [1] "years" "birthmonth" "footlength" "breath" "siblings" [6] "handspan" > nrow(merge_gl.kept2) 20

21 4.3 Basic Data Operations CONTENTS [1] 100 Let us look at different way to grub a portion of a data frame and print them using an R package xtable that can also a latex file, > detach() # This make us unable to access gl data frame using its variable names. W > library(xtable) > data1= gl[,1:4] > s=summary(data1) # basic summary of the data > tab = xtable(s, caption = "My Tables", align =c(" c", " c", " c", " c", " c ")) # > print(tab, file = "assign.tex", append = T, table.placement = "h", caption.placeme Let re-attach the gl data frame, look at some frequency of armcross variable and plot the observed frequencies of armcross with respect to months of year. However, we will explore more visualization in the next chapter. > attach(gl) # so can use names directly > tabarmcross = table(armcross) # create frequency table of values > par(mfrow=c(1,3)) > pie(tabarmcross,main="arm on top when crossing") > barplot(tabarmcross,main="arm on top when crossing") > # turn birthmonth into 'ordered factor' called month > month=factor(birthmonth,levels=c("january","february","march","april","may","june" > boxplot(as.vector(table(month))) Arm on top when crossing Arm on top when crossing right left left right Let us plot footlength against height by differentiate female and male by color, 21

22 4.4 Basic Operations with Matrices CONTENTS > plotcolours=c(1,2)[gender] # chooses 1 or 2 according to Gender > plot(height,footlength,pch=16,col=plotcolours,cex=1.5) # big coloured blobs footlength height 4.4 Basic Operations with Matrices Impoortant, we can also convert the data frame to matric using the R function as.matrix(), ad follow > data_matrix <- gl[,4:7] To illustrate matrices operation, we will use a variable footlength from gel data frame to create 7x7 matrice (a matrice of 7 rows and columns. > length(footlength) [1] 50 > x <- footlength[1:49] > x <- matrix(data=x,nrow=7,ncol=7) > dimnames(x) <- list(c("r1","r2","r3","r4","r5","r6","r7"),c("a","b","c","d","e","f 22

23 4.4 Basic Operations with Matrices CONTENTS > apply(x,1,sum) # sum across the 1st dimension, namely rows r1 r2 r3 r4 r5 r6 r > apply(x,2,sum) # sum across the 2nd dimension, columns a b c d e f g > apply(x,1,min) r1 r2 r3 r4 r5 r6 r Basic Linear Algebra: > gl <- read.csv("class_data.csv",header=t) > attach(gl) > x <- footlength[1:49] > x <- matrix(data=x,nrow=7,ncol=7) > t(x) # transpose a matrix [,1] [,2] [,3] [,4] [,5] [,6] [,7] [1,] [2,] [3,] [4,] [5,] [6,] [7,] > diag(x) # diagonal matrix [1] > sum(diag(x)) # trace of a matrix [1] > x %*% x [,1] [,2] [,3] [,4] [,5] [,6] [,7] [1,] [2,] [3,] [4,] [5,] [6,] [7,]

24 Running scripts CONTENTS > det(x) # determinant of a matrix [1] > eigen(x) # eigenvalues and eigenvectors $values [1] i i i [4] i i i [7] i $vectors [,1] [,2] [,3] [,4] [1,] i i i i [2,] i i i i [3,] i i i i [4,] i i i i [5,] i i i i [6,] i i i i [7,] i i i i [,5] [,6] [,7] [1,] i i i [2,] i i i [3,] i i i [4,] i i i [5,] i i i [6,] i i i [7,] i i i 5 Running scripts It is very convenient to use a plain text writer like Notepad, gedit, Kate, Emacs, or WinEdt for the formulation of several consecutive R commands as separated lines (called scripts and your script must have extension.r or.r). Such command lines can be executed by simply using copy and paste into the command line editor of R. Another possibility is to execute a script from a file (source(my script.r) To illustrate the latter consider the following. We can load the a script file calles Example1.R 6 Important R Tips (1) It is unnecessary to retype commands repeatedly, since R remembers what you have recently entered on the command line. (2) To cycle through the previous commands just push the (up arrow) key. More generally, the command history() will show a whole list of recently entered commands. 24

25 Tutorial CONTENTS (3) To find out what all variables are in the current work environment, use the commands objects() or ls(). These list all available objects in the workspace. If you wish to remove one or more variables, use remove(var1, var2, var3), or more simply use rm(var1,var2, var3), and to remove all objects use rm(list = ls()). (4) Use of scan is when you have a long list of numbers (separated by spaces or on different lines) already typed somewhere else, say in a text file. To enter all the data in one fell swoop, first highlight and copy the list of numbers to the Clipboard with Edit Copy (or by right-clicking and selecting Copy ). Next type the x <- scan() command in the R console, and paste the numbers at the 1: prompt with Edit Paste. All of the numbers will automatically be entered into the vector x. (5) Ctrl+l to clear the screen, Ctrl+l (6) When exiting R the user is given the option to save the workspace. I recommend that beginners DO NOT save the workspace when quitting. If Yes is selected, then all of the objects and data currently in R s memory is saved in a file located in the working directory called.rdata. This file is then automatically loaded the next time R starts (in which case R will say [previouslysavedworkspacerestored]). This is a valuable feature for experienced users of R, but I find that it causes more trouble than it saves with beginners. 7 Tutorial 0. What is the meaning of the following abbreviations: rm, sum, prod, seq, sd, nrow, grep, apply, gl, library, source, setwd, history, str. 1. Reading data into R (a) Use the file Women.txt from the course website and read this into R using read.table(), calling the new R object women. (b) What is the class and dimension of the object women? 2. Matrix manipulations (a) Use the file Women.txt from the course website and read this into R using A new woman joined the study, she is 66 tall, 165 lbs and is 34 years. Use rbind to append a row, containing her information to women (b) Use the file Women.txt from the course website and read this into R using How many women have a weight under 140? (c) Use the file Women.txt from the course website and read this into R using What is the average height of women who weigh between 135 and 145 pounds (hint: first select the data and then find the mean. See the section in lecture 1 on Boolean terms and subsetting). 25

26 Tutorial CONTENTS (d) Use the file Women.txt from the course website and read this into R usingget help on the command colnames. (e) Use the file Women.txt from the course website and read this into R using Change the rownames of women to the letters of the alphabet, eg A, B, C, D etc. (f) Use the file Women.txt from the course website and read this into R using There is a correction to the women is row D, her age should be 39. Change the age in row D to 39. (g) Use the file Women.txt from the course website and read this into R using Sort the matrix women by weight and store the result in newwomen 3. Matrix manipulations Using apply, loop, and writing an R function (a) Use apply to generate a summary report, with the mean, median, sd of height, weight and age. (Hint: use the apply function to get the mean, median and sd of the columns and use rbind to create a matrix with rownames; mean, median and sd). (b) Write a function to calculate BMI. The function should have 2 inputs; weight(lb) and height(in) and should return one value; BMI. The formula for BMI is: bmi = (weight(lb)/[height(in)] 2 ) 703 So for example, if weight = 150 lbs, height = 65. The BMI is ( which is The input to your new function bmi should (65) 2 be > bmi(weight=150, height=65) [1] (c) Do the women have a BMI within the recommend range for their height? (Normal )? (Hint: create women$bmi < bmi(women$weight, women$height) and then test if women$bmi were within normal range). (d) Create a data.frame of 5 columns, called df1, which contains 100 random numbers drawn from the normal distribution with a mean of 8.2. (e) Write a function, called cumsumfn to print the cumulative sum of the row means of this data.frame. Hint: create a new function. Within it, first use apply to get the row means (rmeans). Then write a for loop, which iterates over rmeans to add them to the cumulative sum. (f) Is your output equal to cumsum(rowmeans(df1))? 4. Construct a factor. Construct factors that correspond to the following setting. (a) An experiment with two conditions each with four measurements. (b) Five conditions each with three measurements. (c) Three conditions each with five measurements. 26