Skip to content

Must-known Linux/Unix Tools

File and Directory Management

Linux provides powerful tools for managing files and file systems. Here we will introduce a few essential commands.

pwd: print the current working directory

Usage

$ pwd
/cluster/home/yzhang85
$ cd /cluster/tufts/rt/yzhang85/
$ pwd
/cluster/tufts/rt/yzhang85

cd: change directory

Usage

cd [directory]
If a directory is not supplied as an argument, it will default to your home directory. 
$ pwd
/cluster/tufts/rt/yzhang85
$ cd ..
$ pwd
/cluster/tufts/rt
$ cd  
$ pwd
/cluster/home/yzhang85

Shortcuts

  • .. cd to the parent directory. cd ..
  • ~ cd to the home directory. cd ~
  • - cd to the previous directory. cd -

touch: create new files and update timestamps

touch is used to create new files or to update the timestamps (access and modification times) of existing files.

Create new file

touch newfile.txt

Update timestamps of existing files

touch existingfile.txt

mkdir: create new directory

Usage

mkdir [options] dir_name

Common option

  • -p: Creates parent directories if they don't exist.

$ mkdir -p rnaseq/output
This will create output folder as well as its parent folder rnaseq if it doesn't exist.

mv: move a file/directory to a new location or rename it

Usage

mv [options] source destination

Common option

  • -i: Prompts for confirmation before overwriting an existing file. Useful to avoid accidental data loss.
  • -f: Forces the operation without prompting, even if an existing file would be overwritten. Use with caution!

cp: copy a file/directory

Usage

cp [options] source destination

Common option

  • -r: To copy directory

rm: remove files/directories

Usage

rm [options] file/directory

Common option

  • -r: Deletes recursively any file and subdirectories contained within the given directory.

Text processing

Linux command-line tools are invaluable for bioinformatics text processing due to their efficiency and flexibility. They allow for rapid manipulation and analysis of large biological datasets, such as DNA sequences, protein structures, and gene expression data. Commands like grep, sed, awk, and cut are essential for filtering, extracting, and reformatting text-based biological information.

cat: catenate files (joins their contents)**

Usage

cat [options] file1 file2 …

Common option

  • -n: tells cat to number each line of the output. This is helpful for debugging scripts.

head/tail: display the beginning/end of a file

Usage

head/tail [options] file

Common option

  • -n [number]: Specifies the number of lines to display (default: 10).

less/more: view the content of a file page by page

Usage

less largefile.txt
more largefile.txt
  • Arrow Keys: Use the up and down arrow keys to scroll line by line.
  • Spacebar: Move forward one page.
  • b: Move backward one page.
  • q: Quit and exit less.
  • /search_term: Search for search_term within the file. Press n to go to the next occurrence.
  • g: Go to the beginning of the file.
  • G: Go to the end of the file.
  • Spacebar: Move forward one page.
  • Enter: Move forward one line.
  • q: Quit and exit more.
  • /search_term: Search for search_term within the file (forward only).

cut: Extract sections from each line of files

cut is a text-processing utility used to extract specific sections of lines from a file or standard input. It is commonly used to split lines of text based on delimiters, extract columns of data, and work with fixed-width fields. The cut command is helpful for manipulating text files like TSVs and CSVs.

Usage

cut [OPTIONS] [FILE...]

Common options

  • -d: Specifies the delimiter for field extraction. Default is TAB. Example: -d, (use a comma as the delimiter).
  • -f: Selects specific fields, used with a delimiter (default is TAB). Example: -f 1,3 (extract fields 1 and 3).

Example

cut -f1,3 -d, file.csv ##(Extract columns 1 and 3 from a comma-separated file)

sort: Sort lines of text files

sort is designed to sort plain-text data with columns. Running sort without any arguments simply sorts a file alphabetically. By default, sort treats blank characters (like tab or spaces) as field delimiters. If your file uses another delimiter (such as a comma for CSV files), you can specify the field separator with -t (e.g., -t",").

Usage

Sort lines alphabetically
sort file.txt
Sort lines numerically
sort -n file.txt
Sort by a specific column
sort -k 2 file.txt  # Sort by the second column
Sort by multiple columns
sort -k 1,2 file.txt  # Sort by the first column, then the second

uniq: Report or filter out repeated lines in a file.

uniq is used to filter out or identify duplicate lines in a sorted file. It removes adjacent duplicate lines, so it’s often combined with the sort command to find unique lines in a file effectively.

Usage

sort file.txt | uniq

grep:Extracting lines matching (not matching) a pattern

Usage

grep [options] PATTERN file

Common option

  • -i: ignore cases
  • -v: select non-matching lines.
  • -A NUM: Print NUM lines of trailing context after matching lines.
  • -B NUM: Print NUM lines of leading context before matching lines.

sed: Stream editor for modifying file content

sed (short for stream editor) is a powerful text-processing tool in Bash that allows you to parse and transform text in files or streams. It is commonly used to perform basic text manipulations like search and replace, insert and delete lines, and apply regular expressions on text data.

Substitution (Search and Replace)

Replace the first occurrence of old with new in each line: 

sed 's/old/new/' filename.txt
Replace all occurrences of old with new in each line: 
sed 's/old/new/g' filename.txt

In-place substitution

sed -i 's/old/new/g' filename.txt
Warning: Use this command with caution as it directly modifies the original file. To create a backup, use -i.bak: 
sed -i.bak 's/old/new/g' filename.txt

Delete lines

sed '/pattern/d' filename.txt

sed and awk that we will introduce later are very powerful bash commands. If you have interest in learning more about their usage, below is a very good book.

Data Compression and Archiving

When working with files on Linux, compressing them to save space and bundling multiple files into a single archive is a common practice. The commands gzip, gunzip, and tar are essential tools for file compression and archiving in Bash.

gzip: Compress files

gzip is used to compress files in Linux. It reduces file sizes using the DEFLATE algorithm, resulting in files with the .gz extension.

Usage

Compress file/files
gzip file.txt ## Compress file.txt and replace it with file.txt.gz.
gzip file1.txt file2.txt file3.txt ## Each file will be compressed and replaced with a .gz version.
Keep the Original File
gzip -k file.txt ## Create file.txt.gz while preserving the original file.txt.
Compress the folder
gzip -r directory_name

gunzip: decompress .gz files

Decompress files
gunzip file.txt.gz
gunzip file1.txt.gz file2.txt.gz
Keep the Compressed File
gunzip -k file.txt.gz

tar: Archive multiple files into one or extract them.

tar is used to create, extract, and manipulate archive files. However, tar itself does not compress files; it only archives them by combining multiple files and directories into a single file. This file usually has a .tar extension. However, tar can be used in combination with other compression utilities (like gzip or bzip2) to compress the archive.

Create a .tar archive without compression

tar -cvf archive.tar my_folder

Extract a tar file

tar -xvf archive.tar

Creating a compressed archive(.tar.gz)

tar -cvzf archive.tar.gz my_folder

Extracting a compressed archive(.tar.gz)

tar -xvzf archive.tar.gz

Other useful tools

Redirection: >, >>, <

  • >: Overwrites the contents of a file with the command's output

    cat file1 file2 > files

  • >>: Appends the output to the end of an existing file

​ cat file3 >> files

  • <: Uses the contents of a file as input to a command

​ sort < names.txt

Pipe: |

Pipes in Linux are a powerful feature that allows you to connect the output of one command directly as the input to another command. This is a key concept in Unix/Linux philosophy, which promotes the use of small, modular tools that can be combined to perform complex tasks.

A pipe is represented by the | symbol. When you place a pipe between two commands, the standard output (stdout) of the command on the left of the pipe becomes the standard input (stdin) for the command on the right.

Usage

command1 | command2

sort file.txt | uniq

  • sort file.txt: Sorts the lines in file.txt.

  • uniq: Removes duplicate lines from the sorted output.

Wildcards: selecting multiple files/directories based on patterns

  • *: Represents zero or more characters.

*.fastq.gz matches all fastq.gz files

  • ?: Represents a single character.

file?.txt matches "file1.txt", "fileA.txt", but not "file12.txt".

  • []: Represents a single character within a specified range or set.

[abc]at matches "bat", "cat", or "aat”.

  • [0-9] matches any single digit.

Alias

An alias in Linux is a custom shortcut or abbreviation for a command or a series of commands. Once defined, you can use the alias in place of the original command.

Creating an alias

To create an alias, use the alias command followed by the name you want to give the alias and the command it should execute. 

alias alias_name='command'

Example

alias ll='ls -l'
alias la='ls -a'
alias mav='module avail'
alias ml='module load'

Using alias

[yzhang85@login-prod-03 MPI]$ ll
total 31
-rwxrwx--- 1 yzhang85 yzhang85 16256 May 16 07:57 hello_mpi*
-rw-rw---- 1 yzhang85 yzhang85   731 May 16 07:56 hello_mpi.c
-rwxrwxr-x 1 yzhang85 yzhang85  8392 Dec 23  2023 hello_world*
-rwxrwxr-x 1 yzhang85 yzhang85   731 Oct 31  2023 hello_world.c*
-rwxrwxr-x 1 yzhang85 yzhang85  4053 Oct 31  2023 mpi_hello_world.c*

[yzhang85@login-prod-03 MPI]$ mav blast
-------------------------------- /opt/shared/Modules/modulefiles-rhel6 --------------------------
   blast/2.2.24    blast/2.2.31    blast/2.3.0    blast/2.8.1

------------------------------- /cluster/tufts/hpc/tools/module ---------------------------------
   blast-plus/2.11.0    ncbi-magicblast/1.5.0

------------------------------- /cluster/tufts/biocontainers/modules -----------------------------
   blast/2.15.0 (D)

  Where:
   D:  Default Module

A soft link (also known as a symbolic link or symlink) is a type of file in Linux that points to another file or directory. It’s essentially a shortcut that references the location of another file, allowing you to access it from a different location in the filesystem.

Usage

To create a soft link, you use the ln command with the -s option: 

ln -s target_file link_name

  • target_file: The file or directory you want to link to.
  • link_name: The name of the symlink that will point to the target.

Previous: Files and File system

Next: Loops