Improve hot region prediction by analyzing different machine learning algorithms

BMC Bioinformatics

Table 1 Distribution of data in SKEMPI 2.0

Amino acid	Non-hot spots	Hot spots	All residues	Ratio of hot spots	property of side chain
SER	143	21	164	0.128	Hydroxyl-containing
CYS	6	1	7	0.143	Sulfur-containing
GLN	107	29	136	0.213	Amid
THR	114	31	144	0.214	Hydroxyl-containing
PRO	42	17	59	0.288	Cyclic
ASN	101	41	142	0.289	Amid
GLY	48	20	68	0.294	Aliphatic
VAL	70	30	100	0.3	Aliphatic
GLU	154	68	222	0.306	Acid
HIS	60	28	88	0.318	Basic aromatic
MET	23	11	34	0.326	Sulfur-containing
LYS	131	64	195	0.328	Basic
ARG	146	80	226	0.354	Basic
ASP	101	41	142	0.409	Acid
LEU	48	41	89	0.419	Aliphatic
ILE	72	52	124	0.461	Aliphatic
PHE	51	55	106	0.519	Aromatic
TYR	75	104	179	0.581	Aromatic
TRP	21	50	71	0.704	Aromatic
All	1513	813	2623	0.35	none

ISSN: 1471-2105