Pairwise Sequence Alignment Online Tool (Needleman-Wunsch Global Alignment)

Needleman-Wunsch is the standard algorithm for global alignment, using dynamic programming to guarantee the optimal alignment under a given scoring scheme — not a heuristic approximation.

The algorithm rests on a single recurrence. Let H(i,j) be the optimal score for the first i residues of sequence a and the first j residues of sequence b:

H(i,j) = max{ H(i-1,j-1) + s(aᵢ,bⱼ),    align
              H(i-1,j)   + gap,          gap in b
              H(i,j-1)   + gap }         gap in a

Once the matrix is filled, trace back from the bottom-right corner to recover the alignment.

Global or local? This is the most common source of errors:

• Global alignment (this tool): forces both sequences to be aligned end-to-end. Best for sequences of similar length with high overall homology — e.g., two alleles of the same gene, or a plasmid fragment before and after cloning.
• Local alignment (Smith-Waterman / BLAST): finds only the most similar region. Use this when sequences differ greatly in length or share homology only over a subset of positions. Comparing a short primer against a long template with global alignment will give a result swamped by gaps.

Scoring parameters change the result. The defaults (match +1, mismatch −1, gap −2) are adequate for a quick look. For formal analysis, use a tool with a substitution matrix (EDNAFULL for nucleotides, BLOSUM62 for proteins) and separate gap-open and gap-extend penalties — this tool uses a linear gap penalty, which can give substantially biased results for sequences containing long insertions or deletions.

Computation runs entirely in the browser; time and space are both O(mn), so sequence length is capped at 2000 bp.