BLAST QuickStart (2024)

1.1. Query and Database Sequence Formats

BLAST “query” sequences are given as character strings of single letter nucleotide or amino acid codes, preceded by a definition line, beginning with a “>” symbol and containing identifiers and descriptive information. This format is known as FASTA. BLAST databases are constructed from concatenated FASTA formatted sequences using a program called “formatdb” that produces a mixture of binary- and ascii-encoded files containing the sequences and indexing information used during the BLAST search.

1.2. Scoring of Alignments and Substitution Matrices

A BLAST alignment consists of a pair of sequences, in which every letter in one sequence is paired with, or “aligned to,” exactly one letter or a gap in the other. The alignment score is computed by assigning a value to each aligned pair of letters and then summing these values over the length of the alignment. For protein sequence alignments, scores for every possible amino acid letter pair are given in a “substitution matrix” where likely substitutions have positive values and unlikely substitutions have negative values. By default, BLAST uses the “blosum62” matrix, a member of the most commonly used series of substitution matrix (2), however, several members of the PAM (3) series are also available. For nucleotide alignments, BLAST uses a reward of +2 for aligned pairs of identical letters and a penalty of −3 for each nonidentical aligned pair. The creation of a gap in an alignment results in a negative “gap-creation” penalty, with each extension of a preexisting gap incurring a lesser penalty. For a detailed treatment of the theory of alignment scoring (seeref. 4).

1.3. Overview of the Algorithm

BLAST begins a search by indexing all character strings of a certain length within the “query” by their starting position in the query. The length of the string to index, called the “wordsize” is configurable by the user. The allowable range for the “wordsize” varies according to the BLAST program used; typical values are 3 for protein-to-protein sequence searches and 11 for nucleotide to nucleotide searches. BLAST then scans the database looking for matches between the “words” indexed in the “query” and strings found within the database sequences. For nucleotide-to-nucleotide searches, these matches must be exact; for protein-to-protein searches, the score of the match as determined using a substitution matrix, must exceed a specified threshold. When a word match is found, two nearby words in the case of protein searches, BLAST attempts to extend both forward and backward from the match to produce an alignment. BLAST will continue this extension as long as the alignment score continues to increase or until it drops by a critical amount owing to the negative scores given by mismatches. This critical amount is known as the “dropoff.” The methods BLAST uses to initiate refine alignments are given more fully in refs. 5 and 6.

1.4. Statistical Significance

The alignments found by BLAST during a search are scored, as previously described, and assigned a statistical value, called the “Expect Value.” The “Expect Value” is the number of times that an alignment as good or better than that found by BLAST would be expected to occur by chance, given the size of the database searched. An “Expect Value” threshold, set by the user, determines which alignments will be reported. A higher “Expect Value” threshold is less stringent and the BLAST default of “10” is designed to ensure that no biologically significant alignment is missed. However, “Expect Values” in the range of 0.001 to 0.0000001 are commonly used to restrict the alignments shown to those of high quality.

3.1. Available Nucleotide-Level Searches

There are two members of the BLAST suite of programs that are designed to make nucleotide-to-nucleotide alignments. The first is the original BLAST nucleotide search program known as “blastn.” The “blastn” program is a general purpose nucleotide search and alignment program that is sensitive and can be used to align tRNA or rRNA sequences as well as mRNA or genomic DNA sequences containing a mix of coding and noncoding regions. A more recently developed nucleotide-level BLAST program called MegaBLAST (7) is about 10 times faster than “blastn” but is designed to align sequences that are nearly identical, differing by only a few percent from one another. MegaBLAST allows the rapid mapping of a transcript onto a typical 3 billion base mammalian genome in seconds, and is useful for processing large batches of sequences. A refinement of MegaBLAST, known as discontiguous MegaBLAST, uses a discontiguous template to define an initial “word” in which characters in some positions, such as those in the wobble base position of codons, need not match. Discontiguous MegaBLAST allows rapid cross-species mappings involving coding regions in cases where species differences in codon usage would prevent alignments using the original MegaBLAST program.

3.2. Examples of Nucleotide BLAST Searches

4.1. Available Protein-Level Searches

The default wordsize for a blastp search is three; the default substitution matrix is the blosum62 matrix. Changing the wordsize from three to two increases the sensitivity of the search. Using a different substitution matrix can also have an effect on search sensitivity. During a “blastp” search, low-complexity regions of the query sequence are filtered to reduce the construction of spurious alignments and enhance search speed (seeNote 4).

4.2. Examples of Protein BLAST Searches

5.1. Available Translated Searches

There are three varieties of translated BLAST search; “tblastn,” “blastx,” and “tblastx.” In the first variant, “tblastn,” a protein sequence query is compared to the six-frame translations of the sequences in a nucleotide database. In the second variant, “blastx,” a nucleotide sequence query is translated in six reading frames, and the resulting six-protein sequences are compared, in turn, to those in a protein sequence database. In the third variant, “tblastx,” both the “query” and database “subject” nucleotide sequences are translated in six reading frames, after which 36 (6 × 6) protein “blastp” comparisons are made. Protein sequences are better conserved than their corresponding nucleotide sequences. Because the translated searches make their comparisons at the level of protein sequences, they are more sensitive than direct nucleotide sequence searches. A common use of the “tblastn” and “blastx” programs is to help annotate coding regions on a nucleotide sequence; they are also useful in detecting frame-shifts in these coding regions. The “tblastx” program provides a sensitive way to compare transcripts to genomic sequences without the knowledge of any protein translation, however, it is very computationally intensive. MegaBLAST can often achieve sufficient sensitivity at a much greater speed in searches between the sequences of closely related species and is preferred for batch analysis of short transcript sequences such as expressed sequence tags.

5.2. Examples of Translated BLAST Searches

6.1. Available Genome-Wide Searches

Genome BLAST services are available at NCBI for a variety of organisms including human, mouse, rat, fruit fly, and many others in a growing list. At a minimum, MegaBLAST and “blastn” searches against the complete genome are supported. These are usually offered in conjunction with “tblastn” searches against the genome, “blastp” and “blastx” searches against the proteins annotated on the genome and MegaBLAST, “blastn” and “tblastn” searches against collections of transcript sequences that have been mapped to the genome. Hits to the genome are displayed graphically within NCBI’s MapViewer to show their genomic context.

6.2. Examples of Genome-Wide BLAST Searches

7.1. Comparisons Between Two Sequences

The output of BLAST2Sequences consists of a set of the traditional pairwise alignments generated by the principal BLAST programs it uses, supplemented with a dot plot representation of these alignments. The dot plot is useful for highlighting deletions and duplications of segments between two sequences. The translated variants of BLAST2Sequences are useful for the detection of exons.

7.2.1. Problem 1

Click on the link indicated by “P” next to “Align two sequences (bl2seq).” This problem describes the comparison of two nucleotide sequences. The problem provides a genomic sequence and an mRNA (cDNA) sequence. The genomic sequence is a piece from a GenBank HTG record that contains part of the Werner’s syndrome gene WRN. This Gene contains 35 exons. The figure in the problem on the BLAST QuickStart website shows the mapping of exons to the cDNA coordinates. We will use BLAST2Sequences to determine which exon, if any, is contained in the supplied HTG sequence by comparing it against the WRN gene cDNA sequence. Access the BLAST2Sequences page by clicking on the link “Align two sequences (bl2seq).” Paste the HTG sequence in the top box under “Sequence1” and the cDNA sequence in the bottom box under “Sequence2.” Click on the “Align” button to obtain the outputof Fig. 11. The query (genomic sequence) nucleotides 116..233 are similar to “Sbjct” (cDNA sequence) nucleotides 954..1071. Compare the cDNA coordinates to the exon coordinates provided in the problem. The cDNA coordinates 954..1071 correspond to the exon number 8. Thus, the provided genomic DNA contains exon 8 of the WRN gene.

Note 1

GenBank and nr. The remaining 12 hits of the primer pair to the database sequences may represent the potential for amplification of different regions of the human genome. Alternatively, the result may stem from the redundant nature of GenBank. The default “nr” database used in this problem includes nucleotide sequences from the International Nucleotide Sequence Database Collaboration, which comprises the DNA DataBank of Japan, the European Molecular Biology Laboratory, and GenBank at NCBI (9, 10). It is redundant in nature as each laboratory can submit the nucleotide sequence that they sequenced even if an identical sequence already exists in the database. The nucleotide “nr” BLAST database is not made nonredundant as opposed to the protein “nr” BLAST database, which is made nonredundant by clustering identical proteins in one group. Thus, further analysis of the annotation on each of the entries is necessary to determine whether the primer pair will amplify a unique region of the human genomic DNA in the BRCA1 gene. The user may perform the same search against the human genome BLAST database, which is nonredundant (refer to Genomic Blast (http://www.ncbi.nlm.nih.gov/genome/seq/BlastGen/BlastGen.cgi?taxid=9606) and Note 2). Which parameters will you need to change (seeNote 2)?

Note 2

Primers and human genome BLAST. If a primer pair is used (with some N’s in between) as a query on the human genome BLAST page (NCBI home page-BLAST-Genomes-Human), the user will need to use the blastn as the program and increase thee-value to, say, 10. Select the “genome (reference only)” as the database. The primer pair finds only one hit on chromosome 17, shown in Fig. 14, aligning to two nearby regions (joined by a thin gray line). The parts of the forward and reverse primers also align to two different regions of the genome (as indicated by two separate hits not joined by a thin gray line) on chromosome X and 2. You may view the hit on the human genome by clicking on the Genome View button at the top and accessing the Map Viewer (Fig. 15).

Note 3

Alignment views. BLAST offers a variety of alignment views. For example, the pairwise option shows an alignment of the query to one database sequence at a time. There are other options such as, Query-anchored with identities, Query-anchored without identities, Flat Query-anchored with identities, and Flat Query-anchored without identities. These options show the multiple alignment-like view of the query with the database sequences. They differ in the way identities and gaps are displayed. The option, Flat Query-anchored with identities, is useful to identify the conserved regions (indicated by the dots) in the database sequences with respect to the query and the SNPs (indicated by the one letter code).

Note 4

Low-complexity sequence. The phrase “low-complexity sequence” refers to stretches of nucleotide or protein sequence that are repetitive or simple in composition (11). Extreme examples include runs of As in a nucleotide sequence such as the poly-A tails of eukaryotic mRNAs, or the poly-proline tracts found in some proteins, but the runs need not be limited to repeats of a single base or amino acid. BLAST detects and filters these runs in the “query” by default because they often lead to false starts when BLAST initiates alignments from word hits; beginning an alignment in the poly-a tail of an mRNA is not very likely to lead to a meaningful alignment between related mRNA sequences. This filtering can be turned off on the web using a checkbox, however, the resulting searches will take much longer because BLAST will have to process a great number of false starts. The results returned may also include a larger than usual number of questionable alignments. Nucleotide sequences are filtered using a program called Dust (12); protein sequences are filtered with SEG (13).

Note 5

Automatic CDD search. When a protein–protein BLAST search in ran, the query protein sequence is also searched against the conserved domains database. The presence of a conserved domain in the protein is reported on the page with the request ID before you format the page. For example, for the blastp problem 1, the query protein contains an amino transferase 1_2 conserved domain indicated by the red bar below the query line seen in Fig. 16. Click on the red bar to accessthe conserved domain database and determine the amino acid positions of the domain.

Note 6

Single nucleotide differences in blastx Problem 1. To discover the nucleotide difference in the BLASTX Problem 1, we will compare the query nucleotide sequence to the nucleotide sequence on which the protein AAL71647.1 is annotated. Click on the accession number AAL71647.1. The protein is annotated on the nucleotide entry AY077250.1 as shown in “DBSOURCE.” From the BLAST mini-course page, click on the link “Align two sequences (bl2seq)” in the panel labeled “Special.” Paste the query nucleotide sequence from the problem in the box for Sequence 1 and the accession number, AY077250, in the second box. Unclick the filter box (seeNote 4) and click the “Align” button to create the output shown in Fig. 17. The query nucleotide lacks an “A” corresponding to the nucleotide 266 in AY077250.1 causing a frame shift. There are other differences between the two nucleotide sequences (such as a nucleotide substitution or deletion of three nucleotides), which do not cause a frame shift.

Note 7

Single nucleotide differences in blastx Problem 2. To discover the nucleotide difference in the blastx problem 2, click on the accession number AAL91985.1. The protein is annotated on the nucleotide entry AF482979.1 as shown in “DBSOURCE.” From the BLAST mini-course page, click on the link “Align two sequences (bl2seq)” in the panel labeled “Special.” Paste the query nucleotide sequence from the problem in the box for Sequence 1 and the accession number, AF482979, in the second box. Unclick the filter box and click on the “Align” button to produce the alignment of Fig. 18. The querynucleotide sequence contains an extra “T” at nucleotide 565.

Note 8

Manual CDD search. A protein query can be also manually searched against the conserved domain database. The option is provided under the Protein panel at the “Search the conserved domain database (rpsblast)” link. Perform this search for the protein accession number NP_032294 from the Genomes problem 1 (Fig. 19).