Introduction to Scripting

The purpose of this exercise is to introduce bash scripting for creating command line workflows.

OUTLINE

  1. Why scripts are useful - a prerequisite to scripting.
  2. Text editor and bash scripting basics.

Script: A series of commands or instructions to automate a task. The commands are written in a text file that is then executed by a program without being first compiled (converted into the binary machine code).

Scripting language: A computer programming language that supports scripts. The scripts are typically interpreted by the program and do not have to be compiled.

Part 1. Prerequisite to scripting

In this exercise, we’ll discuss variable assignment and manipulation. The goal is to demonstrate the need to automate repeated tasks, which will be done in the next exercise in which we introduce scripts.

1.1. Open a shell or terminal window.

1.2. Assign a DNA sequence to a variable (sequence):

Variable assignment in bash does not allow spaces before or after =. The variable will be stored until you terminate your terminal session.

To reference a stored variable, we need to prepend a $ to it. So to get the stored value of sequence, we would use $sequence. See below.

1.3. Complement the sequence using tr:

tr generally makes single character substitutions, in this case, substituting each character in the left set of brackets with a character in the same position in the right set of brackets. The number of characters must be the same in each set of brackets. The brackets themselves are optional.

1.4. Reverse the sequence using rev:

The rev command reverses a string of characters.

1.5. Reverse complement the sequence using tr and rev:

Notice that we can redirect the output from one command into another using a pipe (|).

1.6. Calculate the length of the sequence using wc:

The word count command, wc, returns the number of lines, words, and characters in that order.

This exercise required multiple steps and if it is something we wanted to repeat, it could be easily automated with a script, as demonstrated below.

Part 2. Bash scripting

In this exercise, we will write a bash script to identify the length, the complement, the reverse, and the reverse complement of a DNA sequence.

TEXT EDITORS

A good text editor is essential for writing scripts. Microsoft Word and Mac TextEdit should not be used for writing scripts.

Common Free Text Editors:

2.1. Open a shell or terminal window.

2.2. Create a new directory named bash_scripts using mkdir.

2.3. Change into the bash_scripts directory using cd.

2.4. Open a text editor, such as BBEDIT or Notepad++, and use it to create a new file called revcomp.sh within the bash_scripts directory. In the next steps, we will write a bash script by adding commands to the file.

2.5. Confirm that you are in the bash_scripts directory using pwd and that the revcomp.sh script is in the directory using ls.

2.6. Within revcomp.sh prompt the user for a sequence and store as a variable (seq) using the commandread -p:

The -p flag allows you to print a message to the terminal at the prompt, in this case "Enter a sequence: ". Whatever the user enters at the prompt will be saved in the variable seq.

2.7. Complement the sequence using tr and store it as a new variable (comp):

Notice that to capture the output of a command, we need to enclose the command in backticks ( ` ).

2.8. Reverse the sequence using rev and store it as a new variable (reverse):

2.9. Reverse complement a sequence using tr and rev and store it as a new variable (revcomp):

revcomp=echo $seq | tr [ACTGactg] [TGACtgac] | rev

2.10. Compute the length of the sequence using wc -m (the -m flag modifies wc so that it only returns the number of characters) and store as a new variable (length):

length=echo -n $seq | wc -m

2.11. Print to the shell the values of the variables generated in each of the above steps along with a brief description:

What does the command echo "" print to the terminal?

2.12. Return to your terminal window and execute the script using bash revcomp.sh.

2.13. Looping is a technique in computer programming that allows you to repeat (loop) over a task. Edit revcomp.sh to repeat indefinitely using a while loop as follows (this is called an infinite loop). The format of a while loop in bash is:

while some condition is true
    do 
        some block of code to run
    done

2.13.a. Add while before the read command.

In this example, we have an infinite loop becasue the condition read -p "Enter a sequence: " seq will always evaluate to True. We'll demonstrate how to break out of the loop below.

2.13.b. Insert a new line with the command do directly after the previous line containing while:

2.13.c. At the end of the script add a new line with the command done:

2.14. Return to your terminal window and run the script again.

Notice that because of the infinite while loop, the only way to quit the program is to abort the command with ctrl+c. Let's fix that issue.

2.15. Edit revcomp.sh to exit if the user just hits return instead of entering a sequence using a conditional statement.

Like loops, conditionals are a common feature of programming languages. In bash, a conditional statement has the following syntax:

if [[ something == something ]] # == means equal to
    then
        some block of code
fi # closes the conditional

2.15.a. Directy after the read and do statements, check if the variable contains an empty string, which is what will occur if the user just hits return, using a conditional statement:

if [[ $seq == "" ]] then echo "" echo "todo listo. adios!" echo "" exit fi

exit will abort the script.

2.16. Return to your terminal window and execute the script using bash revcomp.sh. Does the program terminate if you hit return without entering a sequence?

Your final script should look something like this:

while read -p "Enter a sequence: " seq
    do
        if [[ $seq == "" ]]
            then
                echo ""
                echo "todo listo. adios!"
                echo ""
                exit
        fi
        comp=`echo $seq | tr [ACTGactg] [TGACtgac]`
        reverse=`echo $seq | rev`
        revcomp=`echo $seq | tr [ACTGactg] [TGACtgac] | rev`
        length=`echo -n $seq | wc -m`
        echo ""
        echo "Original sequence: $seq"
        echo "Complement: $comp"
        echo "Reverse: $reverse"
        echo "Reverse complement: $revcomp"
        echo "Length: $length"
        echo ""
    done

Note that bash doesn't care about indentation so it's technically optional.