Searching databases for typical protein motifs/conserved domains enables the annotation of functional units in proteins, providing insights into sequence/structure/function relationships.
Based on short parts of the query sequence program searches for similar sequences using „substitution matrix“, which defines the score of potential alignments.