Table 2.

Comparison of α-helix propensity scales

Amino acid	N	Design	RAN	Pace	Agadir	Rev_C&F	Luque
G	84	1.24	0.02	1.00	1.10	1.03	0.79
A	344	−0.04	−0.05	0.00	0.00	0.00	0.00
V	257	−0.06	0.09	0.61	0.46	0.58	0.36
I	224	−0.26	0.04	0.41	0.35	0.45	0.48
L	381	−0.38	−0.21	0.21	0.19	0.12	0.15
F	127	−0.01	0.09	0.54	0.47	0.51	0.35
M	49	−0.09	0.39	0.24	0.21	0.18	0.18
W	32	0.21	0.07	0.49	0.47	0.48	0.35
C	25	0.57	0.81	0.68	0.60	0.72	0.57
S	136	0.15	−0.16	0.50	0.52	0.70	0.48
T	100	0.39	0.20	0.66	0.57	0.71	0.59
N	90	0.25	0.25	0.65	0.60	0.67	0.52
Q	107	−0.02	−0.09	0.39	0.32	0.14	0.30
H	75	−0.11	−0.30	0.53	0.47	0.55	0.62
Y	101	0.05	0.16	0.56	0.62	0.50	0.46
D	148	0.27	0.08	0.69	0.59	0.61	0.47
E	392	−0.33	−0.13	0.40	0.34	0.09	0.37
K	114	−0.18	0.02	0.21	0.06	0.23	0.11
R	419	−0.30	−0.04	0.26	0.15	0.31	0.15
Correlation coefficients for all 19 amino acids
Design		1.00	0.40	0.79	0.83	0.79	0.70
RAN		0.40	1.00	0.29	0.24	0.31	0.19

The helix propensity scales are from the following: Design, structure-based propensity scale based on computer generated sequences (this work); RAN, propensity scale based on random sequences with native amino acid composition (this work); Pace (28); AGADIR (52); Rev_C&F [revised statistical scale based on Chou and Fasman (4)]; Luque (29). N is the total number of the corresponding amino acid type in α-helices, in the pool of 100 designed sequences.