BLAST (Basic Local Alignment Search Tool) finds regions of similarity between your sequence and the sequences in NCBI’s databases. Run a search in the official NCBI BLAST tool embedded below, then use the guide to choose the right program and read your results with confidence.
Open NCBI BLAST in a new tab ↗ (recommended on small screens).
Which BLAST program should I use?
Pick the program by the type of query and database:
- blastn — nucleotide query vs nucleotide database. Identifying a DNA sequence, finding homologous genes.
- blastp — protein query vs protein database. Identifying a protein, finding homologs.
- blastx — translated nucleotide query vs protein database. Finding the protein a DNA/EST sequence encodes.
- tblastn — protein query vs translated nucleotide database. Locating a protein’s coding region in genomic/EST data.
- tblastx — translated nucleotide query vs translated nucleotide database. Discovering distant coding similarity.
How to run a BLAST search
- Choose the program above (e.g. blastn for DNA).
- Enter your query as raw sequence, FASTA, or an accession number.
- Select a database — e.g. nr/nt for general searches, RefSeq for curated sequences, or a specific organism.
- (Optional) tune the parameters — organism filter, expect threshold, or a more/less sensitive algorithm (megablast vs blastn).
- Run BLAST and wait for the results page.
Reading your BLAST results
- E-value (Expect): the number of hits this good you’d expect by chance. Lower is better — values near 0 (e.g. 1e−50) indicate a strong, significant match; values above ~1 are likely noise.
- Bit score: a normalised score of alignment quality; higher is better and is comparable across searches.
- Percent identity: the fraction of identical residues in the aligned region.
- Query cover: how much of your query the alignment spans. A high identity over a low query cover may only match a domain, not the whole sequence.
Read these together: a hit with low E-value, high bit score, high identity and high query cover is a confident match.
Video tutorials
Frequently asked questions
What does the E-value mean?
It is the number of matches as good as this one you would expect to find by chance in the database. The lower the E-value, the more significant the hit; values near zero indicate a strong match.
blastn or megablast?
megablast is faster and intended for highly similar sequences (e.g. the same or closely related species); use traditional blastn for more divergent matches.
Which database should I search?
nr/nt is the broad default. Use RefSeq for curated, non-redundant references, or restrict by organism to speed up and focus the search.
High identity but low query cover — is it a match?
It means only part of your query aligns well, often a shared domain. Check whether the covered region is the part you care about before concluding the sequences are homologous.
