Pattern
Amino acids descriptive for a protein family, domain or function
The word pattern has many meanings. Within the context of protein sequence and structure bioinformatics a pattern is normally a description of a series of amino acids that are representative for a protein family, for a domain, or for a certain function.
More general, a pattern can be used to describe the conserved bits of a sequence. lets look at the small alignment:
ALRDFASTHDDF SMTAEALTHDSI ECDQAAGTHEAS GSSMLATTHETA
in which we see the conserved pattern A-x-T-H. But we also see the pattern A-x-T-H-[DE] ([DE] means D or E at this position).
Another example: The pattern Asn - not a Pro - a Ser or a Thr - not a Pro is a potential N-glycosylation site. So if you find the sequence ....Asn-Ala-Ser-Val.... you might have found an N-glycosylation site. This pattern can be described as:
For these purposes Prosite uses a pattern language that has gotten widely accepted. In this language the glysosylation pattern becomes:
N - {P} - [ST] - {P}
in which the curly brackets mean 'not what is listed between' and the square brackets mean 'one of those listed between'.
The term pattern and motif are both in use.
The Prosite database holds very many of these patterns.