MegabLAST

MegaBLAST is a variant of the BLAST family designed to rapidly align highly similar nucleotide sequences. It is part of the NCBI BLAST+ toolkit and is optimized for speed when comparing large datasets or large reference databases. MegaBLAST achieves fast performance by using a large seed word size to reduce the search space and then extending candidate matches to produce high-scoring alignments.

The algorithm relies on seed-and-extend principles, using long exact word matches as seeds and applying a sequence extension step to build high-scoring segment pairs and, when appropriate, gapped alignments. This approach makes MegaBLAST particularly effective for finding closely related sequences rather than distant homologs.

Applications of MegaBLAST include genome-to-genome comparisons, strain-level identification, and other tasks where query and database sequences are expected to be highly similar (often 95% identity or higher). It is especially useful for rapid screening of large nucleotide databases and for genome-scale analyses. Its performance advantage comes at the cost of reduced sensitivity to more divergent relationships compared with standard BLASTN.

Output and availability: MegaBLAST is used for nucleotide-to-nucleotide searches against nucleotide databases such as nt or RefSeq within the BLAST+ package. Results are returned in standard BLAST formats, including query coverage, percent identity, alignment length, bit scores, and E-values. It can be run from the command line or via NCBI’s online BLAST interface with the appropriate task selection.