tblastx

Translated nucleotide to nucleotide search

TLDR

Search query against database

$ tblastx -query [sequence.fasta] -db [database] -out [results.txt]

Search with specific output format

$ tblastx -query [sequence.fasta] -db [database] -outfmt [6]

Search against local FASTA file

$ tblastx -query [query.fasta] -subject [subject.fasta]

Set e-value threshold

$ tblastx -query [sequence.fasta] -db [database] -evalue [1e-5]

Use multiple threads

$ tblastx -query [sequence.fasta] -db [database] -num_threads [4]

Limit number of alignments

$ tblastx -query [sequence.fasta] -db [database] -max_target_seqs [10]

Display help

$ tblastx -help

tblastx searches a translated nucleotide query against a translated nucleotide database. Both query and database sequences are translated in all six reading frames (three forward, three reverse), and protein-level comparisons are performed.This program is useful for finding distant protein-coding relationships between nucleotide sequences, especially when working with unannotated genomic or EST data. It is computationally intensive due to the multiple translations.Unlike other BLAST programs, tblastx only supports ungapped alignments. For sequences with insertions or deletions, consider using blastx or tblastn instead.Part of the NCBI BLAST+ suite, tblastx requires properly formatted BLAST databases created with makeblastdb.

PARAMETERS

-query file

Input FASTA file with nucleotide query sequence(s).

-db database

BLAST database name to search.

-subject file

Subject sequence file for pairwise comparison.

-out file

Output file (default: stdout).

-outfmt format

Output format (0=pairwise, 6=tabular, 7=tabular with comments, etc.).

-evalue value

Expectation value threshold (default: 10).

-max_target_seqs n

Maximum number of aligned sequences to keep.

-num_threads n

Number of CPU threads to use.

-query_gencode code

Genetic code for query translation.

-db_gencode code

Genetic code for database translation.

-strand strand

Query strand(s) to search: both, plus, minus.

-soft_masking bool

Apply soft masking to query.

-seg value

Filter query with SEG (yes, no, or parameters).

-matrix name

Scoring matrix (default: BLOSUM62).

-word_size n

Word size for initial match.

-h, -help

Display help information.

INSTALL

sudo apt install ncbi-blast+

CAVEATS

Only ungapped alignments are supported. Very computationally intensive due to six-frame translations on both query and database. Requires BLAST+ installation and formatted databases. Memory usage scales with database and query sizes.

HISTORY

tblastx is part of BLAST (Basic Local Alignment Search Tool), originally developed at NCBI in 1990 by Stephen Altschul and colleagues. The BLAST+ suite, rewritten in C++, replaced the original C implementation. It remains a fundamental tool in bioinformatics for sequence similarity searching.