BLAST sequence alignment

Suppose you find a mysterious sequence of DNA and want to know what it is. The fastest way is not to think hard about it but to ask a vast library: 'Has anyone seen something like this before?' If your unknown sequence closely matches a gene that someone already studied in a fish or a fly, you instantly inherit a strong guess about what it does. BLAST is the search engine that answers that question by comparing your sequence against millions of others.

Sequence alignment means lining up two sequences to reveal where they match, where they differ, and where letters have been inserted or deleted, so you can measure how similar they are. Doing this carefully against every entry in a giant database would be impossibly slow, so BLAST (Basic Local Alignment Search Tool) uses a clever shortcut: it first finds short exact 'seed' words shared between your query and a database entry, then only bothers to extend a full alignment outward from those promising seeds. It returns hits ranked by an alignment score and, importantly, an E-value — the number of matches this good you would expect to find by pure chance in a database this size.

BLAST is one of the most-used tools in all of biology because similarity is such a powerful clue: closely related sequences usually share ancestry and often share function. But the output must be read with care. A low E-value means a match is unlikely to be coincidence, not that the two sequences do the same job — similar sequence suggests but does not prove similar function. Short queries and repetitive or low-complexity regions can throw up misleading hits, and a strong match to a wrongly-annotated database entry will cheerfully pass that wrong label on to you. Alignment finds resemblance; interpreting what the resemblance means is still your job.