Table 1.

Artificial data set.

Artificial data set A (4-mers: P2-P1-P1'-P2')	Artificial data set B (6-mers: P4-P3-P2-P1-P1'-P2')
P1' ∈ {A,I,L,M,F,V}	P4 ∈ {A,G,I,L,M,F,T,V}
P2' ∉ {P}	P3 ∈ {A,G,I,L,F,T,V,W}
	P2 ∈ {P}
	P1 ∈ {R}
	P1' ∉ {D,E}
	P2' ∉ {D,E}

Rules for the two artificial data sets used. occur.

Any amino acid could occupy the two first positions in artificial data set A (the generated peptides were longer than the actual rule). One letter amino acid abbreviations are used. The sign ∈ means "in" and the sign ∉ means "not in". The rules are connected with the Boolean AND operator, which means that all position rules must be true for cleavage to