Multi-Sequence Aligner & Colored Alignment Viewer

Optional · Load a reference from the human genome (hg38)

For Sanger confirmation: fetch an exon (plus intronic flank) or any region from GRCh38, then align your reads to it. Differences are reported in genomic coordinates. Exon numbers follow the Ensembl canonical (MANE Select) transcript, the same numbering the Gene Panel Designer uses. Needs internet (Ensembl + UCSC).

1 · Load sequences

Paste multi-FASTA text, open a local FASTA/text file, and/or add sequences one at a time with a label. Nothing leaves your browser — all parsing and alignment happens locally, with no network calls.

Loaded sequences

2 · Alignment results

Differences from the reference

Most variable columns

Colored alignment view

Aligned FASTA (plain text, equal-length, gaps as "-")

Method & disclosures

• Alignment method: a real Needleman-Wunsch global pairwise aligner extended to multiple sequences via the standard "center-star" heuristic — every other sequence is pairwise-aligned to one chosen reference/center sequence, and the resulting gap patterns are merged into one shared coordinate system. This is a legitimate, well-known MSA approximation, but it IS a heuristic, not a true multi-sequence optimization like MUSCLE, Clustal Omega, or MAFFT. It works well for closely related sequences (e.g. barcode markers within a genus or family); for a divergent or large sequence set, or for anything going into a publication/phylogenetic analysis, re-align with a dedicated MSA tool and treat this as a quick first look.
• Scoring: affine gaps (Gotoh): match +2, mismatch −3, gap opening −7 and extension −2 per base (NCBI blastn defaults), so an indel is scored as one event rather than scattered single gaps. IUPAC ambiguity codes (e.g. R, Y from Sanger heterozygous calls) are kept and score as partial matches; N scores 0. By default overhangs at either end are free (semi-global), which suits reads and amplicons aligned to a longer reference. Reads in the opposite orientation are detected by shared 8-mers and reverse-complemented automatically.
• Column coloring: a column is shown as "conserved" only when every loaded sequence has the identical base there AND no sequence has a gap there; otherwise it is "variable". This is a simple identity-based heuristic, not a substitution-matrix-weighted conservation score.
• %identity: the summary panel's average pairwise %identity is the mean of every pairwise sequence's own Needleman-Wunsch global-alignment identity (matches / aligned columns for that pair) — not derived from the multi-sequence alignment's columns, so it is a global-alignment identity, not a phylogenetic distance metric.
• Performance: the colored view is capped to a maximum number of rendered columns for very long alignments (noted on-screen when this happens) so the browser never has to build an enormous DOM — the downloaded aligned FASTA is always complete and uncapped regardless of what the on-screen view shows.
• Scope: built for short-to-medium nucleotide sequences (barcode markers, single genes, amplicons) — not for BAM/VCF/FASTQ files or whole genomes, and this tool does not do read mapping, variant calling, or genome assembly of any kind.
• Everything runs client-side in your browser; no sequence data is uploaded or sent anywhere.