Table 2.

Based on the ($\Delta ,\sigma$)-criterion, we obtained the following amino acid motifs that are statistically significant (calculated frequency) when compared to the frequency of amino acids described in Table 1

Amino acid pair	Calculated frequency (%)	Expected frequency (%)
AA	0.78	0.60
RR	0.36	0.28
NN	0.22	0.18
CC	0.04	0.02
CH	0.04	0.03
HC	0.04	0.03
QQ	0.24	0.15
EE	0.58	0.43
EK	0.48	0.39
HH	0.08	0.05
HP	0.14	0.11
KK	0.47	0.31
PP	0.31	0.22
SS	0.63	0.48
WW	0.02	0.01
YY	0.12	0.10
CM	0.03	0.04
EP	0.24	0.26
ES	0.36	0.46
HE	0.12	0.15
HK	0.11	0.14
IM	0.12	0.14
PM	0.09	0.11
WP	0.04	0.08

The first 16 residue pairs represent “positive” motifs and occur more frequently within the database than expected. The remaining eight pairs represent “negative” motifs (grayed boxes) and occur less frequently than expected. Data are taken from the Swiss-Prot database and it includes all the $\Delta$-positive and $\Delta$-negative pairs.