BLAST

BLAST
Name	BLAST
Developer	National Center for Biotechnology Information
Released	1990
Genre	Bioinformatics tool
License	Public domain

BLAST. The Basic Local Alignment Search Tool is a fundamental algorithm in the field of bioinformatics used for comparing primary biological sequence information, such as the amino acid sequences of proteins or the nucleotide sequences of DNA and RNA. Developed at the National Institutes of Health, it enables researchers to rapidly search sequence databases for regions of local similarity, which can infer functional and evolutionary relationships between sequences. Its speed, sensitivity, and accessibility have made it an indispensable resource for molecular biologists, geneticists, and researchers across the life sciences.

Introduction to BLAST

BLAST is designed to address the critical need for fast and sensitive database similarity searches, a cornerstone of modern genomics and proteomics. It operates by comparing a query sequence against a vast library of sequences housed in repositories like GenBank, the Protein Data Bank, and the European Molecular Biology Laboratory database. The core principle involves finding short, high-scoring segment pairs, which are then extended to generate alignments. This methodology allows for the identification of homologous genes, prediction of protein function, and discovery of novel sequence motifs, underpinning research from phylogenetics to structural biology.

History of BLAST

The original BLAST algorithm was conceived and published in 1990 by Stephen Altschul, Warren Gish, Webb Miller, Eugene Myers, and David J. Lipman at the National Center for Biotechnology Information. This work, a landmark paper in the Journal of Molecular Biology, revolutionized sequence analysis by providing a heuristic method that was orders of magnitude faster than the rigorous Smith-Waterman algorithm. Subsequent developments led to specialized variants, including PSI-BLAST for more sensitive protein searches and BLASTN for nucleotide queries. The algorithm's continuous refinement and integration into platforms like the Ensembl genome database project have maintained its central role for over three decades.

BLAST Algorithm

The BLAST algorithm employs a heuristic approach to approximate the optimal local alignment without performing an exhaustive search. It begins by compiling a list of high-scoring words, or *k*-mers, from the query sequence based on a substitution matrix like BLOSUM or PAM. These words are then scanned against a pre-indexed database to find exact matches, termed *seeds*. Each seed is extended in both directions to generate an alignment, a process halted when the cumulative score drops below a threshold. This efficient strategy, balancing sensitivity and speed, is mathematically formalized in the Karlin-Altschul statistics which assess the significance of the results.

Applications of BLAST

BLAST has a vast array of applications in biological research and biotechnology. It is routinely used for annotating genes in newly sequenced genomes from organisms ranging from Escherichia coli to Homo sapiens. In medical diagnostics, it helps identify pathogenic strains by comparing sequences to databases like the Influenza Research Database. It facilitates the design of polymerase chain reaction primers and molecular cloning experiments. Furthermore, BLAST is instrumental in metagenomics studies of environments like the human microbiome and in evolutionary studies to construct phylogenetic trees based on sequence homology.

BLAST Parameters and Options

Users can tailor a BLAST search through numerous parameters to optimize for specific goals. Key settings include the choice of the **E-value** threshold, which filters results based on statistical significance, and the **word size**, which affects speed and sensitivity. The selection of a scoring matrix, such as BLOSUM62 for proteins, defines the penalties for mismatches and gaps. Other important options involve filtering for low-complexity regions, adjusting gap costs, and limiting the search to specific organisms or taxonomic groups within the NCBI Taxonomy Database.

BLAST Output and Interpretation

A standard BLAST report provides a ranked list of database matches, each with critical metrics for biological interpretation. The output includes a graphical overview, a list of significant alignments with scores like **bit score** and **E-value**, and detailed pairwise alignments showing sequence identity. A low E-value indicates a high-probability match, suggesting potential homology. Researchers use this data to infer gene function, predict protein domains, and identify conserved regions critical for structure, as referenced in resources like the Conserved Domain Database. Proper interpretation requires understanding these statistical measures within the biological context of the query.

Category:Bioinformatics Category:Computational biology Category:Bioinformatics algorithms