BLAST is an algorithm that searches databases of sequences for similar matches to a query. The results are a list of sequence identifiers for the matches, an alignment of those sequences, and a score which is used to rank the hits. The score is based on how likely it is that an alignment would be produced by chance. BLAST can be used to find protein or DNA sequences, and it is also used for constructing phylogenetic trees.

Several different versions of BLAST exist, each with its own scoring matrix and search options. NCBI's webpage provides information about how to choose which BLAST version to use for a given task.

The basic BLAST algorithm is implemented in a software library that runs on a host computer, together with a database to search and other optional parameters. BLAST output is delivered in various formats, including HTML and plain text, though the default format is a table showing all of the hits with their sequences and scores.

BLAST searches are computationally intensive, because they compare every residue in the query sequence against all of the sequences in the database. To reduce the number of comparisons, BLAST looks for short, common sequence regions called "words" in the query sequence and the database sequence. Words are generally 3 to 15 letters long. If a word match is found, BLAST can then start to make local alignments by extending both forward and backward from the first match. This process is repeated until either the alignment score reaches a certain threshold, or the aligned sequences diverge too widely from the query.

Once an alignment is made, BLAST can also perform sequence conservation analysis to determine the percentage of amino acids that need to be identical in the two aligned sequences to be considered a perfect match. Using this information, other algorithms can be applied to generate a more precise alignment.

There are a number of specialized BLAST programs, such as MEGABLAST and PSI-BLAST, that can handle larger sequences more efficiently. BLAST can also be used to identify conserved domains within protein sequences, which can then be used to map annotations between species.

Power users of BLAST often have a custom set of command-line options that they use for particular tasks. These options are saved in a "strategy" file, which can be specified when running BLAST. This is useful for reproducing a specific set of results, or a search with particular settings. The BLAST+ application supports writing a strategy file in ASN.1 (Abstract Syntax Notation, a structured language similar to XML) which can be read by other applications, such as Geneious Prime.

BLAST is an algorithm that searches databases of sequences for similar matches to a query. The results are a list of sequence identifiers for the matches, an alignment of those sequences, and a score which is used to rank the hits. The score is based on how likely it is that an alignment would be produced by chance. BLAST can be used to find protein or DNA sequences, and it is also used for constructing phylogenetic trees.

Several different versions of BLAST exist, each with its own scoring matrix and search options. NCBI's webpage provides information about how to choose which BLAST version to use for a given task.

The basic BLAST algorithm is implemented in a software library that runs on a host computer, together with a database to search and other optional parameters. BLAST output is delivered in various formats, including HTML and plain text, though the default format is a table showing all of the hits with their sequences and scores.

BLAST searches are computationally intensive, because they compare every residue in the query sequence against all of the sequences in the database. To reduce the number of comparisons, BLAST looks for short, common sequence regions called "words" in the query sequence and the database sequence. Words are generally 3 to 15 letters long. If a word match is found, BLAST can then start to make local alignments by extending both forward and backward from the first match. This process is repeated until either the alignment score reaches a certain threshold, or the aligned sequences diverge too widely from the query.

Once an alignment is made, BLAST can also perform sequence conservation analysis to determine the percentage of amino acids that need to be identical in the two aligned sequences to be considered a perfect match. Using this information, other algorithms can be applied to generate a more precise alignment.

There are a number of specialized BLAST programs, such as MEGABLAST and PSI-BLAST, that can handle larger sequences more efficiently. BLAST can also be used to identify conserved domains within protein sequences, which can then be used to map annotations between species.

Power users of BLAST often have a custom set of command-line options that they use for particular tasks. These options are saved in a "strategy" file, which can be specified when running BLAST. This is useful for reproducing a specific set of results, or a search with particular settings. The BLAST+ application supports writing a strategy file in ASN.1 (Abstract Syntax Notation, a structured language similar to XML) which can be read by other applications, such as Geneious Prime.