This version introduces the MASK family of commands which allows boxing, shading etc to be applied according to the frequency of occurence of the character types at each position in the alignment. For example, it is possible to box positions where one character is seen in more than N of the sequences. It is also possible to box/shade etc the most frequently occurring character at each position. Commands exist to select which characters will be used in the calculation of frequencies and which will be excluded, thus boxing can be based upon two or more character types at a position. MASK commands also exist to show residues identical to one sequence in the set. See the section on MASK below for details.
NOTE: Although boxing according to the frequency of amino acids seen at a position is a popular method of representation it is not usually the most informative. An analysis that takes into account the physico-chemical properties of the amino acids and also relates the amino acid similarities to the overall similarity between the sequences is more helpful in identifying functionally important residues. The AMAS program (Livingstone and Barton, 1993) applies a flexible hierarchical set-based approach to this problem.