Local Scoring Matrices
Local scoring matrices are vital in the process of alignment. The PAM matrix is an example. PAM stands for "Point Accepted Mutations" or "Percent of Accepted Mutations." It gives the probability that a given amino acid will remain the same or be replaced by another during evolution. The matrix takes into account the frequency with which each amino acid occurs in proteins. The PAM-1 matrix represents an evolutionary progression in which 1% of the amino acids have changed. The most widely used PAM matrix is the PAM-250.
BLOSUM, "Blocks Substitution Matrix," a set of matrices was developed by Steve Henikoff and is considered to be more accurate by PAM. Below is a BLOSUM matrix.