CCR NAME: _____________________________

R Reference Sheet
Eric Pitman Annual Summer Workshop in Computational Science

Author: C. Ryan Mraz

Summer 2013
 R Reference Sheet

---------------------------------------------------------------------RStudio Tips-------------------------------------------------------------------

There is no editor window until you open up a file! To do so, click:

Here:
Then Here:

To see your history (commands you have already issued), click the history pane or

simply click the up arrow on your keyboard while on the command line

To change the relative sizes of each window, hover the mouse over the window border

until appears.

Is your project loaded? Check the upper right Corner:

There are two ways to load csv files in Rstudio:

1) In the RStudio Workspace:

Select Import Dataset: From Text File

Select a .csv file to Open

Use Heading=Yes

2) From the command line:

Set the Working Directory

Load command:

> drop=read.csv(drop.csv)

*Keep Your Projects Tidy!!

To clear the Console window, use: ctrl + L
To clear individual items in the Workspace, use: r m(variable_name)
To clear all items in the Workspace or plotspace, use:
 R Reference Sheet

-----------------------------------------------------------------Common functions--------------------------------------------------------------

length() # How many elements

dim() # Retrieve the dimension of an object.

class() # Class of the vector (=class of its elements)

str() # Number of elements, type, and contents

sum() # Sum of all element values

length() # Number of elements

unique() # Generate vector of distinct values

diff() # Generate vector of first differences

sort() # Sort elements, omitting NAs

order() # Sort indices, with NAs last

rev() # Reverse the element order

na.omit() # Removes rows containing any "NA" values

which(x==#) # Finds indices that satisfy a condition

table() # Creates frequency or contingency tables for your data

levels() # Displays the values that a categorical variable may hold

mean() # Computes and Reports Average Value

median() # Computes and Reports Median Value

range() # Reports min and max:

min() # Minimum value

max() # Maximum value

var() and sd() # Variance, standard deviation

summary() # Reports Combination of measures

cor(X,Y) # Reports Pearson correlation coefficient

 R Reference Sheet

-----------------------------------------------------------Conditionals/Function Calls---------------------------------------------------------

if (condition is true) {

# do something

functionName =function(inputs) {

---------------- # do something

return (------) #result

----------------------------------------------------------------Common Plots--------------------------------------------------------------------

example scatterplot:

data(diamonds, package=ggplot2)

plot( formula=price~carat,

data=diamonds,

col=darkblue,

pch=20,

main="Diamond Price with Size"

example barplot:

data(diamonds, package=ggplot2)

ideal=diamonds[diamonds$cut=="Ideal","color"]

barplot( table(ideal),

xlab="color",

ylab="count",

main="Ideal cut diamonds by Color",

col="hotpink" )
 R Reference Sheet

example histogram:

data(Cars, package=MASS)

hist( Cars93$RPM,

breaks = 4,

xlab="RPM",

main="histogram of engine RPM",

col="red"

example density plot:

data(Cars, package=MASS)

plot( density(Cars93$RPM,bw=200),

main="Density Curve of Engine RPMs of 93 Cars",

xlab="RPM",

col="blue"

example boxplot:

boxplot(formula=mpg~gear,

data=mtcars,

main="Mileage by Gear Number",

xlab="Number of Gears",

ylab="Miles Per Gallon",

col=c("red","green","blue")

)
 R Reference Sheet

example ROC Curve:

library(pROC)

plot.roc( roc(exp$human_crystal, exp$class3_crystal),

ylab="Sensitivity (True Positive Rate)",

xlab="Specificity (1 - False Positive Rate)",

print.auc = TRUE,

print.auc.col="red",

main='Generation 8 ROC curve: 13 proteins, 2 time points each',

print.thres=TRUE,

print.thres.col="blue",

grid=TRUE

#------------------------------------------------------------Common Graph Modifiers-------------------------------------------------------

abline(lm(y~x)) # prints linear regression line on graph

pch=# # Chooses the type of point character to plot

cex = # # Magnifies text or labels on a graph/chart [smaller<(default=1)<larger]

par(mfrow=c(rows,collumns)) # Prints multiple graphs/charts on one sheet

par(mar=c(#,#,#,#)) # Changes margins sizes starting at bottom

legend(x="location", # location of legend

title = "---",

c("Label.1","Label.2",etc.), # separation labels

fill = c("Color.1","Color.2",etc.)

*N.B. There are practically endless possibilities for making graphs and plots pretty!! Play around and find out how!!
 R Reference Sheet

#--------------------------------------------------------------Apply Family---------------------------------------------------------

# There are many types of the function apply, but for our purposes, we will only be using sapply.

sapply

The apply() family of functions can be used to call some other function multiple times on a dataset, with several
different arguments. sapply() returns a vector or matrix result. You can use sapply() on a native R function, or
on a function you wrote yourself.

EXAMPLE:

> u=c(33,45,37,50) # Creating Vector u

> v=c(2,5,8,11) # Creating Vector v
> d=data.frame(u=u,v=v) # Creating Data frame d from Vectors u and v
>
> d # This is what our data frame looks like:
u v # 4 rows of 2 columns
1 33 2
2 45 5
3 37 8
4 50 11
>
>
> sapply( d, mean) # Here, we apply the mean function to our data frame
# using sapply

u v # sapply applies the mean function to each column of

41.25 6.50 # the data frame and outputs each answer in a user-
# friendly format