TBLASTN

TBLASTN is a program in the BLAST (Basic Local Alignment Search Tool) family that searches a nucleotide database with a protein query by translating the database sequences in all six reading frames and aligning the protein sequence to those translations. This approach allows direct detection of coding regions in genomic or transcriptomic DNA that code for proteins related to the query, even when gene structure is unknown.

During the search, each nucleotide sequence is conceptually translated in the three forward frames and three reverse frames, producing six amino acid representations. The protein query is aligned to these translations using the standard BLAST scoring system, including substitution matrices (such as BLOSUM62) and gap penalties. Alignments are reported with the corresponding subject sequence, frame, strand, start and end positions, score, E-value, percent identity, and gaps.

Input consists of a protein sequence in FASTA format as the query and a nucleotide database (also FASTA) as the subject, such as a genome assembly. The output can be viewed in text, HTML, or tabular formats and lists significant hits, their coordinates in the nucleotide database, and alignment details. TBLASTN is widely used to annotate genomes by projecting known proteins onto DNA sequences, to discover unannotated genes, and to study evolutionary relationships across species.

Limitations include reduced sensitivity for highly diverged sequences, potential spurious matches in repetitive regions, and the difficulty of precisely delineating exon-intron boundaries from protein-to-nucleotide alignments. It is computationally intensive, especially for large genomes, and is often complemented by follow-up analyses and alternative tools such as BLASTX, BLASTN, or TBLASTX.