Pairwise Alignment

Dynamic programming algorithms for aligning two sequences.

Global Alignment (Needleman-Wunsch)

from Bio import pairwise2
from Bio.pairwise2 import format_alignment

seq1 = "HEAGAWGHEE"
seq2 = "PAWHEAE"

# Simple global alignment
alignments = pairwise2.align.globalxx(seq1, seq2)
for align in alignments[:3]:
    print(format_alignment(*align))

Local Alignment (Smith-Waterman)

# Local alignment finds best local match
alignments = pairwise2.align.localxx(seq1, seq2)
best = alignments[0]
print(f"Score: {best[2]}")
print(f"Start: {best[3]}, End: {best[4]}")

Custom Scoring

# Custom match/mismatch scores
alignments = pairwise2.align.globalms(seq1, seq2,
                                       match=2,      # Match score
                                       mismatch=-1,  # Mismatch penalty
                                       open=-0.5,    # Gap open penalty
                                       extend=-0.1)  # Gap extend penalty

Using Substitution Matrices

from Bio.Align import substitution_matrices

blosum62 = substitution_matrices.load("BLOSUM62")

# Alignment with BLOSUM62
alignments = pairwise2.align.globalds(seq1, seq2, blosum62, -10, -0.5)

Alignment Statistics

def calculate_identity(align1, align2):
    """Calculate percent identity between aligned sequences."""
    matches = sum(1 for a, b in zip(align1, align2) if a == b and a != '-')
    length = len(align1.replace('-', ''))
    return (matches / length) * 100