<\body> What did we learn last time? <\itemize-dot> matrices: create with . Address entries in an array: 8]> Apply a function on all rows, or all column of an array:\ <\itemize-dot> - calculate mean of all rows. - calculate mean of all columns. All this is very useful for bootstraping! Functions: Read data from file: Till now we saw 3 types of vectors: numbers, booleans, and strings <\input| >> a=1:4 <\input| >> a <\input| >> a=c("hello","there") <\input| >> a <\input| >> a=(1:4)\2 <\input| >> a <\input| >> \; > Now we'll learn about factors. Factors are somewhere between a string and a boolean. We use them when we have values, such as strings, that have several different types <\input| >> a=sample(c("summer","winter"),20,rep=T) <\input| >> a <\input| >> \; > This is a vector of strings. Now we want to use the fact that we have just 2 types: <\input| >> a=factor(a) <\input| >> a <\input| >> a[1] <\input| >> as.numeric(a) <\input| >> levels(a) <\input| >> levels(a)=c("s","w") <\input| >> a <\input| >> \; > This has several advantages. The main one is that it is much faster to check if factors are equal to one another - you just have to compare numbers. Anoter comes up later, when we learn about ANOVA.\ We want to read the following table: (switch to excel) \; We saved it as comma separated values. Let us first see how we can list files in a directory: <\input| >> list.files() > Very easy. How can we see in which directory we are? <\input| >> getwd() > And how to change it? <\input| >> setwd("~/R-course-2006/lecture3") > These are a bit hard to remember: , and for set working directory, and get working directory. <\input| >> a=read.table("example3.csv",head=T,row.names=1,as.is=3, sep=",") <\input| >> a <\input| >> \; > Normaly, R will convert a string column that is read in into categorical data. To prevent R from doing that, you should give the argument, with the columns that should be read as strings. If a column has values that are essencially all different, store as string. If some values are the same and you will compare between values, use categorical data (i.e. no ) The are many ways to access the data in the table: <\input| >> a$height <\input| >> a$hei <\input| >> a$hai <\input| >> \; > The method using the dollar sign takes a column. You can see that you can abbreviate the name. <\input| >> a[,1] <\input| >> a[1:2,] <\input| >> a[,"email"] <\input| >> a[,"ema"] <\input| >> \; > You can not abreviate when you use . We want to sort the table by height. How do we do that? Here the function helps. It gives us the order of elements. <\input| >> a <\input| >> order(a$height) <\input| >> \; > So will be sorted by height if we write first the 5th row, then the 3rd, then the 6th, then the 1st, and so on. we can do that like this: <\input| >> i=order(a$height) <\input| >> i <\input| >> a[i,] <\input| >> a[ order( a$height ) , ] <\input| >> \; > In R, almost everything can have names: <\input| >> a=c(2,3,4) <\input| >> a <\input| >> a=c(x=2,y=3,z=4) <\input| >> a <\input| >> names(a) <\input| >> a=1:4 <\input| >> names(a) <\input| >> names(a)=c("a","b","c","d") <\input| >> a <\input| >> names(a) <\input| >> \; > Now we see that each entry has a name. \; This is useful when functions return values, because it tells us what each value is: <\input| >> a=rnorm(100) <\input| >> summary(a) <\input| >> \; > is a function, and the values that it returns have names: <\input| >> sum.a=summary(a) <\input| >> names(sum.a) <\input| >> sum.a["Mean"] <\input| >> sum.a["Max."] <\input| >> \; > matrices and data.frames have names for the rows and columns: <\input| >> a=matrix(1:6,2,3) <\input| >> a <\input| >> rownames(a)=c("first row","second row") <\input| >> colnames(a)=c("a","b","c") <\input| >> a <\input| >> a[,"a"] <\input| >> a[,"b"] <\input| >> \; > Row names are especially important when we want to use data from different sources: for example when we have results of one experiment for some genes, and more results of from a second experiment. The row names allow us then to quickly connect the results for the same genes. Last time we talked a bit about functions. Now some more. Let us define a simple function: <\input| >> f=function(x) x+1 <\input| >> f(4) <\input| >> f(9.3) <\input| >> x=17 <\input| >> f(2) <\input| >> x <\input| >> \; > So, a function is very easy to define: We just say followed by an expression. <\input| >> square= function(z) z^2 <\input| >> square(3) <\input| >> square(1:5) <\input| >> square > We can see that functions can simply be stored in variables, like numbers or strings. Functions can also take several arguments: <\input| >> mult = function(x,y) x*y <\input| >> mult(2,3) <\input| >> mult(1:4,2:5) <\input| >> \; > Applying a function to a vector <\input| >> a=1:10 <\input| >> f=function(x) x+1 <\input| >> sapply(a,f) <\input| >> \; > applies the function to each element in the vector, and returns the result. If the result is a vector, then we get a matrix: <\input| >> g=function(x) c(x,x+1) <\input| >> g(2) <\input| >> g(5) <\input| >> sapply(1:4,g) <\input| >> \; > Sometimes we need more than one expression for the calculation. In that case we can enclose the calculations with . The result will be the last expression. <\input| >> die.roll.6=function(){ x=sample(1:6,1); x==6 } <\input| >> die.roll.6() <\input| >> \; > Assignments in functions do not affect the outside: <\input| >> x=100 <\input| >> die.roll.6() <\input| >> x <\input| >> \; > To see what is happenening in a function, we can add print statements: <\input| >> die.roll.6=function(){ x=sample(1:6,1); print(x); x==6 } <\input| >> die.roll.6() <\input| >> \; <\input| >> \; > R also has conditional statements, and loops: <\input| >> f = function(x) { if( x \ 3) 4 else 5} <\input| >> f(2) <\input| >> f(6) <\input| >> \; > <\input| >> for(i in 1:10) print(i) <\input| >> a=1 <\input| >> for(i in 1:10) a = a * i <\input| >> a \; > has the following structure: you give a variable that will iterate over a vector or list of things.\ The expression will be called with each i. \; <\input| >> a > \; \; \; <\initial> <\collection> <\references> <\collection> > > > > > > > > |?>> |?>> |?>> |?>> |?>> |?>> |?>> |?>> <\auxiliary> <\collection> <\associate|toc> |math-font-series||Lecture 3> |.>>>>|> |factors |.>>>>|> > |read.table |.>>>>|> > |names |.>>>>|> > |What is the difference between matrices and data.frames? |.>>>>|> > |Functions |.>>>>|> > |Conditionals |.>>>>|> > |Loops |.>>>>|> >