Sequence alignment

Sequence alignment is the process of arranging two or more DNA, RNA, or protein sequences to identify regions of similarity, along with insertions, deletions, or mismatches, revealing how closely related the sequences are or where a given sequence sits relative to a known reference.

A pairwise alignment compares two sequences directly, commonly used to map short sequencing reads back to a reference genome, while a multiple sequence alignment compares many sequences at once to find conserved regions across them. Algorithms score matches, mismatches, and gaps and search for the highest-scoring arrangement, using heuristics because finding the exact optimal alignment becomes computationally impractical at genome scale.

Sequence alignment is the foundational step behind variant calling, which compares an aligned sample against a reference to find differences, and behind phylogenetic analysis, which uses alignments across species or individuals to infer evolutionary relationships. Pitfalls include repetitive genomic regions that can align ambiguously to multiple locations, and the fact that a poor alignment upstream produces confidently wrong variant calls or evolutionary trees downstream, even though the later statistical analysis may look entirely sound.