Search methods
Choose how a query is compared with each stored sequence, or let Automatic choose. The search record always names the method that ran.
The methods
- Automatic picks the method by the query. Protein to protein runs Smith-Waterman. DNA to DNA runs BLAST, and a query under 50 bases also gets an exact scan of every stored sequence on both strands.
- Alignment (BLAST) runs NCBI BLAST+, with up to 5,000 sequences for each query.
- Smith-Waterman aligns the query with every sequence in the set, so on DNA it finds short matches that BLAST misses.
- Global on the query aligns the whole query with the part of each sequence it fits.
- Global on both sequences aligns the whole query with the whole of each sequence.
- Exact text and Pattern find the query, or a pattern, letter for letter in every stored sequence. See Exact and pattern search.
When to use which
- A peptide or a protein: Automatic, which runs Smith-Waterman.
- A short DNA query, such as a primer or a probe: Smith-Waterman, or Automatic, which adds the exact scan under 50 bases.
- Long DNA queries, or many of them, when time matters: BLAST.
- A sequence that must appear as written: Exact text. A motif with variable positions: Pattern.
Short DNA matches
In a benchmark on Patent DNA, run when it held 28.8 million sequences, Smith-Waterman found 16 of 20 DNA fragments of 21 to 50 bases at 80% identity, where BLAST found 8.
Smith-Waterman reads every stored sequence, so a DNA search takes longer than BLAST. On the hosted service in November 2026, a Smith-Waterman query of 21 bases took 54 seconds on Patent DNA.
Example
The GLP-1 sample searched the 31 letters of GLP-1, from UniProt P01275, residues 98 to 128, in Patent proteins. Automatic ran Smith-Waterman. One of the patents it found, WO 9001540-A, published 1990-02-22, holds all 31 letters at positions 75 to 105:
Query 1 HAEGTFTSDVSSYLEGQAAKEFIAWLVKGRG 31
|||||||||||||||||||||||||||||||
Hit 75 HAEGTFTSDVSSYLEGQAAKEFIAWLVKGRG 105