READING FILES IN R

*CSV files-read.csv()
data<-read.csv(“data.csv”)

*Excel files-library(readxl)
data<-read(“data.xlsx”)

*Other data-read.delim()
data<-read.delim(“datasetfile.text”,header=FALSE)
data
file.choose(),read_tsv()-readr package,read_file()
Writing files in R
*CSV-write.csv()
data<-data.frame(Name=c(“Meera”,”Vani”),Age=c(25,25))
write.csv(data,”output.csv”,row.names=FALSE)
*Excel-xlsx package
write.xlsx(data,file=”result.xlsx”,sheetName=”mydata”,app
end=FALSE)
*Text-.txt
Programming with input and output statements
readline()-input in string and convert to desired data
type
Functions
*n=readline();
*as.integer(n);
*as.numeric(n);
*as.complex(n);
*as.character(n);
as.Date(n);
example
> n<-readline()
555
>n
[1] "555"
> n=as.integer(n)
>n
[1] 555
scan()
input method
list=scan()
list
Reading file using scan()
sf=scan(“fileString.txt”,what=” “)
df=scan(“fileDouble.txt”,what=double())
OUTPUT STATEMENTS IN R

print()
cat()
message()
sprintf()
write.table()
{
str="Hello Hi";
num=123;
dec=34.4;
str;
print(str);
sprintf(“%s is a string",str);
}
[1] "Hello Hi is a string"
sprintf("%d is a number",num);
[1] "123 is a number"
sprintf("%f is a dloat value",dec);
[1] "34.400000 is a dloat value"
> x<-42
> y<-"Hello"
> print(x)
[1] 42
> cat("the value of x is",x,"and y is",y,"\n")
the value of x is 42 and y is Hello
CONDITION STATEMENTS

*If statement
*if-else statement
*if -else if -else statement
*switch statement
if statement
executes the statement if condition is true
Syntax- if(condition)
{
-------code------
}

> x<-10
> if(x>5)
+{
+ cat("x is greater than 5
\n")
+}
x is greater than 5
if- else statement
else block gets executed if the condition is false
syntax- if(condition)
{
---code---
}
else
{
---code---
}

> x<-3
> if(x>5)
+{
+ cat("x is greater than 5\n")
+ }else{
+ cat("x is lesser than 5 \n")
+}
x is lesser than 5
if-else if-else statement
test multiple conditions
syntax-if(condition1){
code---
}else if(condition2){
code---
}else{
code
}
> x<-8
> if(x>10){
+ cat("x is greater than 10 \n")
+ }else if(x>5){
+ cat("x is greater than 5 but lesser than 10\n")
+ }else{
+ cat("x is not greater than 5 \n")
+}
x is greater than 5 but lesser than 10
switch statement
compares different values ,multiway bracnch statemen t
syntax-
switch(expression,
value1={
code---
},
value2={
code---
},
default={
code--
}
}
)
> day<-"sunday"
> switch(1,
+ "monday"=cat("2nd day of the week \n"),
+ "saturday"=cat("weeend \n"),
+ default=cat("regular day \n")
+ )
2nd day of the week
 LOOPS
Repeats a block code multiple times ,iteration,condition is met.

*for loop
*while loop
*repeat loop
for loop
entry controlled loop ,condition is tested first and then body is executed

syntax-
for(var in vector)
{
statements;
}
(or)
for(var in seq){
code---
}
for loop 1 to 5 list in concatenation
> for(x in 1:5) > for(x in c(-
+{ 5,6,9,15,20))
+ print(x^2) +{
+} + print(x)
[1] 1 +}
[1] 4 [1] -5
[1] 9 [1] 6
[1] 16 [1] 9
[1] 25 [1] 15
[1] 20
nested for loop for loop for list

> for(x in 1:4) > for (x in

+{ list("INDIA","US","UK"))
+ for(y in 1:x) +{
+{
+ print(x*y)
+ print(x)
+} +}
+} [1] "INDIA"
[1] 1 [1] "US"
[1] 2 [1] "UK"
[1] 4
[1] 3
[1] 6
[1] 9
[1] 4
[1] 8
[1] 12
[1] 16
for loop with break for loop with next
statement statement
> for(x in c(2,4,6,8,10,12))
+{ > for(x in c(2,4,6,8,10,12))
+ if(x==0) +{
+{ + if(x==0)
+ break +{
+} + next
+ print(x) +}
+} + print(x)
[1] 2 +}
[1] 4 [1] 2
[1] 6 [1] 4
[1] 8 [1] 6
[1] 10 [1] 8
[1] 12 [1] 10
[1] 12
for loop to create histogram multiple plots
mat<-matrix(rnorm(50),ncol=2)
> par(mfrow=c(2,2))
> for(i in 1:2){
hist(mat[, i],main=paste("column",i),xlab="values",col="red")
}
while loop
no of iterations of a loop is unknown
syntax
while(condition){
statements
}

> x<-1
> while(x<6)
+{
+ print(x)
+ x=x+1
+}
[1] 1
[1] 2
[1] 3
[1] 4
[1] 5
repeat loop
excutes till break statement

no condition for exit

syntax
repeat{
statements
if(condition
){
break
}
}
> count<-1 > data<-c("banglore")
> repeat{ > x<-1
+ cat("count",count, "\n") > repeat{
+ count<-count+1 + print(data)
+ if(count>5){ + x<-x+1
+ break + if(x>5)
+} +{
+} + break
count 1 +}
count 2 +}
count 3 [1] "banglore"
count 4 [1] "banglore"
count 5 [1] "banglore"
[1] "banglore"
[1] "banglore"
FUNCTIONS
*Perform task multiple times
*blocks of code can be called multiple times
defining a function

function()

syntax
function_name<-function(arg1,arg2...)
{ code---
return(result)
}
calling a function
> add<-function(a,b){
+ result<-a+b
+ return(result)
+}
> result<-add(3,5)
> cat("the result",result,"\n")
the result 8
>
Built in functions
*sum() > print(sum(5:10))
*max() [1] 45
*min() > print(max(5:10))
*seq() [1] 10
*mean() > print(min(5:10))
[1] 5
> print(seq(5,10))
[1] 5 6 7 8 9 10
> print(mean(5:10))
[1] 7.5
user defined functions

function arguments

named and unamed arguments

res<-add(a=5,b=3)
res1<-add(5,3)
function with default values
power<-function(base,exponent=2){
res<-base^exponent
return(res)
}

ex-2
res1<-power(2)
res2<-power(2,3)
function with single parameter
area_square<-
function(side){
+ area=side*side
+ return(area)
+}
> print(area_square(8))
64
multiple parameter
> si=function(p,r,t)
+{
+ res=(p*r*t)/100
+ return(res)
+}
> print(si(1000,10,2))
[1] 200
without parameter and without return

> fun=function()
+{
+ print("hello")
+}
> print(fun())
[1] "hello"
[1] "hello"
variable scoping
lexical scoping-variables are local by default
<<- assign global variables

Inline functions fun=function(x)x^2*5+x/

3
function() print(fun(5))
lambda or anonymous functions print(fun(-4))
syntax
function(arg1,arg2,....){
code
}
sapply()
numbers<-c(1,2,3,4,5)
sqaure<-sapply(numbers,function(x)x^2)
cat(“squared numbers:”,square,”\n”)
Exceptions
*runtime error due to some conditions at run time
*0 division,file not found,array range.
*Mechanism for handling runtime errors or exceptional
situations that may occur during execution of a
program .
*tryCatch()
*try()
*withCallingHnadlers()
try()
*Helps to continue with program execution even when error occurs
*expression expecting exception can be written in try block
x<-5
> x<6
[1] TRUE
> x>6
[1] FALSE
> try(x>7)
[1] FALSE
tryCatch()
Handles the conditions and controls what happens based
on condition
Main conditions -errors,warnings
Primary tool
structure res<-tryCatch({
x<-5/0
res<-tryCatch({ },error=function(err){
code...
},error=function(err){ cat(“error”,msg(err),”\n”)
code--- }, finally={
},finally={
code-- cat(“the operation is
} complete \n”)
}
})
withCallingHandlers()
alternative to tryCatch() for local handlers tryCatch for existing
check<-function(expression){
withCallingHandlers(expression,
warning=function(w){
message(“warning\n”,w)
},
error=function(e){
message(“error\n”,e)
},
finally={
message(“completed”)
}
}
}
check({10/2})
check({10/0})
check({10/’abc’})
Exception classes
conditionClass()

custom exception-more information

custerror<-function(message){
structure{
list(message=message,call=sys.c
all(-1)),
class=”custerror”
}}
tryCatch({
stop(“custerror(“custom
message”))
},error=function(err){
cat(“custom error
msg”,conditionMessage(err),”\n”)
})
TIMINGS
Timings function

sys.time():current time,elapsed time

system.time()-CPU time for R,execution time ,returns object of
user time ,system time , elapsed time
proc.time()-CPU time,expression sequence ,function call
> system.time(print(5>1))
[1] TRUE > start2<-proc.time()
user system elapsed > si=function(p,r,t)
0 0 0 +{
> start<-Sys.time() + res=(p*r*t)/100
> print(10>=5) + return(res)
[1] TRUE +}
> Sys.time()-start > print(si(1000,10,2))
Time difference of 29.94285 secs [1] 200
> start2<-proc.time() > proc.time()-start2
> print(10>=5) user system elapsed
[1] TRUE 0.11 0.21 50.95
> proc.time()-start2 > proc.time()-start2
user system elapsed user system elapsed
0.05 0.08 28.75 0.14 0.25 70.47
>
 OPTIMIZING TIMING * To improve speed and efficiency
*Execution time

*Vectorized operation-Use vectors instead of loops

*Avoid global variables-function arguments instead of global variables
*Efficient Data Structures-data frames,matrix
*Memory management-data.table for large code,ff.Use rm(),gc().
*Parallel processing-parallel package , foreach package.
*Optimize functions- profvis,microenchmark,bench .duplicate caalculations,simplify logic
*Avoid Unnecessary data copies-data.table::set()- modify
*Use effiecient package -data .table
*Caching and Memorization-resuse intermediate result .
*Profiling -Rprof,profvis,system
 PACKAGES
Repositoties of R programming packages

CRAN-Comprehensive R Archive Network

*FTP network
Bioconductor-topic specific repository
OSS for bioinformatics
Github-oss projects,unlimited space
 Installing packages
install.packages(“package name”)

Update,check installed packages

check- installed.packages()
update-update.packages()

loading packages in R

library(dplyr)
require(dplyr)
Profvis package
Profiling R,visualizing code execution

> profvis({
+ for(i in 1:1000000){
+ sqrt(i)
+}
+}

Microbenchmark package
measuring execution time
Benchmarking with bench package
detailed information
Benchmarking with rbenchmark package
compares execution time of multiple functions