17.6 Exercise 10: Regular expressions

Create the script “exercise8.R” and save it to the “Rcourse/ExtraModule” directory: you will save all the commands of exercise 8 in that script.
Remember you can comment the code using #.

Answer

getwd()
setwd("~/Rcourse/ExtraModule")

1- Play with grep

• Create the following data frame

df2 <- data.frame(age=c(32, 45, 12, 67, 40, 27), 
    citizenship=c("England", "India", "Spain", "Brasil", "Tunisia", "Poland"), 
    row.names=paste(rep(c("Patient", "Doctor"), c(4, 2)), 1:6, sep=""),
    stringsAsFactors=FALSE)

Using grep: create the smaller data frame df3 that contains only the Patient’s but NOT the Doctor’s information.

Answer

# Select row names
rownames(df2)
# Select only rownames that correspond to patients
grep("Patient", rownames(df2))
# Create data frame that contains only those rows
df3 <- df2[grep("Patient", rownames(df2)), ]

• Use grep and one regular expression to retrieve from df2 patients/doctors coming from either Brasil or Spain. Brainstorm a bit!

Answer

df2[grep("a[a-z]*i", df2$citizenship),]

• Use grep and one regular expression to retrieve from df2 patients/doctors coming from either Brasil or England.

Answer

df2[grep("[gi]l", df2$citizenship),]

2- Play with gsub

Build this vector of file names:

vector1 <- c("L2_sample1_GTAGCG.fastq.gz", "L1_sample2_ATTGCC.fastq.gz", 
    "L1_sample3_TGTTAC.fastq.gz", "L4_sample4_ATGGTA.fastq.gz")

Use gsub and an appropriate regular expression on vector1 to retrieve only “sample1,” “sample2,” “sample3” and “sample4.”

Answer

# | is used as OR
gsub(pattern="L[124]{1}_|_[ATGC]{6}.fastq.gz", 
    replacement="", 
    x=vector1)