exercise10_regex.Rmd

## 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.
<br>Remember you can comment the code using #.


<details>
<summary>
*Answer*
</summary>

```{r, eval=F}
getwd()
setwd("~/Rcourse/ExtraModule")
```

</details>


**1- Play with grep**

* Create the following data frame

```{r}
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.

<details>
<summary>
*Answer*
</summary>

```{r}
# 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)), ]
```

</details>

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

<details>
<summary>
*Answer*
</summary>

```{r, eval=F}
df2[grep("a[a-z]*i", df2$citizenship),]
```

</details>

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

<details>
<summary>
*Answer*
</summary>

```{r, eval=F}
df2[grep("[gi]l", df2$citizenship),]
```

</details>

**2- Play with gsub**

Build this vector of file names:

```{r}
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".

<details>
<summary>
*Answer*
</summary>

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

</details>