BLAST

BLAST
Name	BLAST
Developer	National Institutes of Health; National Center for Biotechnology Information
Initial release	1990s
Latest release	ongoing
Programming language	C, C++
Operating system	Unix-like, Microsoft Windows, macOS
License	public domain / open-source components

BLAST

BLAST is a widely used sequence similarity search suite developed for rapid comparison of biological sequences. It enabled researchers at institutions such as the National Institutes of Health, European Molecular Biology Laboratory, and universities including Stanford University and Harvard University to map query sequences against large repositories like GenBank, UniProt, and the Protein Data Bank. The tool influenced projects ranging from the Human Genome Project to pathogen surveillance programs at the Centers for Disease Control and Prevention.

Overview

BLAST provides heuristic algorithms to identify regions of local similarity between a query sequence and database sequences, producing alignments, scores, and significance estimates. It integrates statistical models popularized by scholars affiliated with Columbia University, University of California, Berkeley, and Cambridge University to compute E-values and bit scores, and interacts with resources such as RefSeq, Ensembl, and Swiss-Prot. Common workflows span pipelines used by researchers at Broad Institute, Wellcome Sanger Institute, and clinical labs at Mayo Clinic.

History and Development

Development began under leadership at the National Center for Biotechnology Information in the late 1980s and early 1990s, with foundational papers authored by scientists connected to University of Arizona and collaborators from Washington University in St. Louis. Early releases coincided with milestones at the Human Genome Project and technical advances at companies like Applied Biosystems and research centers such as Lawrence Berkeley National Laboratory. Over time, stewardship and extensions involved teams at Cold Spring Harbor Laboratory, University of California, San Diego, and commercial partners including Illumina and Thermo Fisher Scientific.

Key academic influences include methodology work from scholars at Princeton University, Yale University, and Massachusetts Institute of Technology, and statistical refinements from groups at Stanford University School of Medicine and University College London. Major distribution points were hosted by the National Institutes of Health and mirrored by repositories at European Bioinformatics Institute and national supercomputing centers such as Argonne National Laboratory.

Algorithm and Methodology

BLAST uses a seed-and-extend strategy: it first finds short exact or near-exact matches (seeds) between a query and database sequences and then extends those seeds to produce alignments. This approach contrasts with full dynamic programming algorithms associated with researchers at University of Cambridge and algorithmic frameworks from Carnegie Mellon University and University of Washington. Scoring matrices like BLOSUM and PAM, developed by scientists linked to University of Geneva and University of Texas, are integral for protein searches, while nucleotide scoring schemes reflect models used by groups at Johns Hopkins University.

Statistical significance of matches leverages extreme value distribution theory advanced by statisticians at Columbia University and University of California, Los Angeles, producing E-values and bit scores that guide interpretation in studies at institutions such as Yale School of Medicine and University of Michigan. Implementations optimize k-mer selection, gap penalties, and low-complexity filtering—techniques refined in collaborations with researchers at Imperial College London and University of Edinburgh.

Variants and Implementations

Multiple BLAST variants address different data types and query needs: programs developed include implementations for translated searches, gapped alignments, and domain-specific pipelines used in projects at European Molecular Biology Laboratory and National Human Genome Research Institute. Public implementations and derivatives emerged from organizations like Rosetta Commons, Open Bioinformatics Foundation, and commercial vendors including GenBank submitters and biotech firms.

Reimplementations and optimized wrappers target high-throughput environments at centers such as Lawrence Livermore National Laboratory, Oak Ridge National Laboratory, and cloud platforms managed by Amazon Web Services and Google Cloud Platform. Notable adaptations integrate with tools from Galaxy Project, Bioconductor, and workflow managers used at European Genome-phenome Archive and clinical sequencing centers like Johns Hopkins Hospital.

Applications and Use Cases

BLAST underpins annotation and discovery in numerous projects: gene annotation in consortiums like Ensembl and GENCODE; comparative genomics at Max Planck Institute; metagenomics work in initiatives such as Horizon 2020-funded studies and environmental surveys coordinated with Smithsonian Institution. Clinical applications include pathogen identification in programs at Centers for Disease Control and Prevention and antimicrobial resistance profiling in studies affiliated with World Health Organization.

Other uses span functional inference in proteomics projects deposited to Protein Data Bank, phylogenetic placement in collaborations with Royal Society-affiliated labs, and forensic genetics workflows in forensic units partnered with Federal Bureau of Investigation. Integration with education and training occurs through courses at Massachusetts Institute of Technology, University of California, Berkeley, and online platforms developed by edX and Coursera partners.

Performance, Limitations, and Accuracy

BLAST offers favorable speed-versus-sensitivity trade-offs suitable for large databases maintained by GenBank and UniProt, but heuristic shortcuts can miss weak homologies that exhaustive algorithms from groups at University of Pennsylvania or University of British Columbia might detect. Accuracy depends on scoring matrices, composition adjustments developed by teams at European Bioinformatics Institute, and proper filtering strategies used in pipelines at Wellcome Sanger Institute.

Scalability challenges arise in ultra-large datasets encountered in projects at Human Microbiome Project and national sequencing initiatives such as those coordinated by National Health Service (England), prompting alternative tools and GPUs from vendors like NVIDIA and algorithmic improvements pursued by researchers at ETH Zurich and University of Tokyo. Practical deployments balance sensitivity, runtime, and memory, and results require expert interpretation by curators at UniProt Consortium and annotation groups at RefSeq.

Category:Bioinformatics software