Regular Expressions in Python

Examples:

• Find all email addresses in a document
• Validate phone numbers
• Extract gene IDs from biological data
• Find DNA/RNA sequence patterns
• Clean messy text data

Getting Started

Import the Module

import re

Why Raw Strings Matter

# Normal string - \n becomes a newline
print("Hello\nWorld")
# Output:
# Hello
# World

# Raw string - \n stays as literal characters
print(r"Hello\nWorld")
# Output: Hello\nWorld

In regex, backslashes are special! Raw strings prevent confusion:

# ❌ Confusing without raw string
pattern = "\\d+"

# ✅ Clean with raw string
pattern = r"\d+"

Level 1: Literal Matching

The simplest regex matches exact text.

import re

dna = "ATGCGATCG"

# Search for exact text "ATG"
if re.search(r"ATG", dna):
    print("Found ATG!")

Your First Function: re.search()

match = re.search(r"ATG", "ATGCCC")
if match:
    print("Found:", match.group())    # Found: ATG
    print("Position:", match.start())  # Position: 0

Practice

Level 2: The Dot . - Match Any Character

The dot . matches any single character (except newline).

# Find "A" + any character + "G"
dna = "ATGCGATCG"
matches = re.findall(r"A.G", dna)
print(matches)  # ['ATG', 'ACG']

New Function: re.findall()

text = "cat bat rat"
print(re.findall(r".at", text))  # ['cat', 'bat', 'rat']

Practice

Level 3: Character Classes [ ]

Square brackets let you specify which characters to match.

# Match any nucleotide (A, T, G, or C)
dna = "ATGCXYZ"
nucleotides = re.findall(r"[ATGC]", dna)
print(nucleotides)  # ['A', 'T', 'G', 'C']

Character Ranges

Use - for ranges:

re.findall(r"[0-9]", "Room 123")      # ['1', '2', '3']
re.findall(r"[a-z]", "Hello")         # ['e', 'l', 'l', 'o']
re.findall(r"[A-Z]", "Hello")         # ['H']
re.findall(r"[A-Za-z]", "Hello123")   # ['H', 'e', 'l', 'l', 'o']

Negation with ^

^ inside brackets means "NOT these characters":

# Match anything that's NOT a nucleotide
dna = "ATGC-X123"
non_nucleotides = re.findall(r"[^ATGC]", dna)
print(non_nucleotides)  # ['-', 'X', '1', '2', '3']

Practice

Level 4: Quantifiers - Repeating Patterns

Quantifiers specify how many times a pattern repeats.

Examples

# Find sequences of 2+ C's
dna = "ATGCCCAAAGGG"
print(re.findall(r"C+", dna))       # ['CCC']
print(re.findall(r"C{2,}", dna))    # ['CCC']

# Find all digit groups
text = "Call 123 or 4567"
print(re.findall(r"\d+", text))     # ['123', '4567']

# Optional minus sign
print(re.findall(r"-?\d+", "123 -456 789"))  # ['123', '-456', '789']

Combining with Character Classes

# Find all 3-letter codons
dna = "ATGCCCAAATTT"
codons = re.findall(r"[ATGC]{3}", dna)
print(codons)  # ['ATG', 'CCC', 'AAA', 'TTT']

Practice

Level 5: Escaping Special Characters

Special characters like . * + ? [ ] ( ) have special meanings. To match them literally, escape with \.

# ❌ Wrong - dot matches ANY character
text = "file.txt and fileXtxt"
print(re.findall(r"file.txt", text))  # ['file.txt', 'fileXtxt']

# ✅ Correct - escaped dot matches only literal dot
print(re.findall(r"file\.txt", text))  # ['file.txt']

Common Examples

re.search(r"\$100", "$100")           # Literal dollar sign
re.search(r"What\?", "What?")         # Literal question mark
re.search(r"C\+\+", "C++")            # Literal plus signs
re.search(r"\(test\)", "(test)")      # Literal parentheses

Practice

Level 6: Predefined Shortcuts

Python provides shortcuts for common character types.

Examples

# Find all digits
text = "Room 123, Floor 4"
print(re.findall(r"\d+", text))  # ['123', '4']

# Find all words
sentence = "DNA_seq-123 test"
print(re.findall(r"\w+", sentence))  # ['DNA_seq', '123', 'test']

# Split on whitespace
data = "ATG  CCC\tAAA"
print(re.split(r"\s+", data))  # ['ATG', 'CCC', 'AAA']

Practice

Level 7: Anchors - Position Matching

Anchors match positions, not characters.

Examples

dna = "ATGCCCATG"

# Match only at start
print(re.search(r"^ATG", dna))   # Matches!
print(re.search(r"^CCC", dna))   # None

# Match only at end
print(re.search(r"ATG$", dna))   # Matches!
print(re.search(r"CCC$", dna))   # None

# Word boundaries - whole words only
text = "The cat concatenated strings"
print(re.findall(r"\bcat\b", text))  # ['cat'] - only the word
print(re.findall(r"cat", text))      # ['cat', 'cat'] - both

Practice

Level 8: Alternation - OR Operator |

The pipe | means "match this OR that".

# Match either ATG or AUG
dna = "ATG is DNA, AUG is RNA"
print(re.findall(r"ATG|AUG", dna))  # ['ATG', 'AUG']

# Match stop codons
rna = "AUGCCCUAAUAGUGA"
print(re.findall(r"UAA|UAG|UGA", rna))  # ['UAA', 'UAG', 'UGA']

Practice

Level 9: Groups and Capturing ( )

Parentheses create groups you can extract separately.

# Extract parts of an email
email = "user@example.com"
match = re.search(r"(\w+)@(\w+)\.(\w+)", email)
if match:
    print("Username:", match.group(1))   # user
    print("Domain:", match.group(2))     # example
    print("TLD:", match.group(3))        # com
    print("Full:", match.group(0))       # user@example.com

Named Groups

Use (?P<name>...) for readable names:

gene_id = "NM_001234"
match = re.search(r"(?P<prefix>[A-Z]+)_(?P<number>\d+)", gene_id)
if match:
    print(match.group('prefix'))  # NM
    print(match.group('number'))  # 001234

Practice

Level 10: More Useful Functions

re.sub() - Find and Replace

# Mask stop codons
dna = "ATGTAACCC"
masked = re.sub(r"TAA|TAG|TGA", "XXX", dna)
print(masked)  # ATGXXXCCC

# Clean multiple spaces
text = "too    many     spaces"
clean = re.sub(r"\s+", " ", text)
print(clean)  # "too many spaces"

re.compile() - Reusable Patterns

# Compile once, use many times (more efficient!)
pattern = re.compile(r"ATG")

for seq in ["ATGCCC", "TTTAAA", "GCGCGC"]:
    if pattern.search(seq):
        print(f"{seq} contains ATG")

Practice

Biological Examples

Validate DNA Sequences

def is_valid_dna(sequence):
    """Check if sequence contains only A, T, G, C"""
    return bool(re.match(r"^[ATGC]+$", sequence))

print(is_valid_dna("ATGCCC"))  # True
print(is_valid_dna("ATGXCC"))  # False

Find Restriction Sites

def find_ecori(dna):
    """Find EcoRI recognition sites (GAATTC)"""
    matches = re.finditer(r"GAATTC", dna)
    return [(m.start(), m.group()) for m in matches]

dna = "ATGGAATTCCCCGAATTC"
print(find_ecori(dna))  # [(3, 'GAATTC'), (12, 'GAATTC')]

Count Codons

def count_codons(dna):
    """Split DNA into codons (groups of 3)"""
    return re.findall(r"[ATGC]{3}", dna)

dna = "ATGCCCAAATTT"
print(count_codons(dna))  # ['ATG', 'CCC', 'AAA', 'TTT']

Extract Gene IDs

def extract_gene_ids(text):
    """Extract gene IDs like NM_123456"""
    return re.findall(r"[A-Z]{2}_\d+", text)

text = "Genes NM_001234 and XM_567890 are important"
print(extract_gene_ids(text))  # ['NM_001234', 'XM_567890']

Quick Reference

Key Functions Summary

Resources

• Regex 101 Interactive Tutorial
• Regex Tester
• Python re Documentation