Information Technology Reference
In-Depth Information
Table 4.
Class II ISC Method results.
No. of
iterations
LOO
Non-cross Validation
n
a
Epitope
SEP
b
q
2 c
NC
d
SEE
e
r
2
DRB1*0101
90
13
0.567
0.808
8
0.075
0.994
Human
DRB1*0401 185
7
0.701 0.716
4
0.174
0.967
DRB1*0701
84
11
0.562
0.649
7
0.051
0.999
I-A
b
44
7
0.459
0.850
6
0.089
0.994
I-A
d
145
14
0.534
0.898
6
0.136
0.993
I-A
k
55
4
0.816
0.790
6
0.180
0.990
Mouse
I-A
s
81
17
0.588
0.783
6
0.177
0.980
I-E
d
69
8
0.557
0.732
6
0.096
0.992
I-E
k
52 8 0.418 0.925 6 0.106 0.995
a
Number of epitopes;
b
Standard Error of Prediction;
c
Obtained after Leave-One-Out Cross-
Validation;
d
Number of components;
e
Standard Error of Estimate.
Table 5.
Class II ISC-Additive Method: nonanchor residues related with strong and
weak binding for amino acids only.
Allele
P1
P2
P3
P4
P5
P6
P7
P8
P9
Favored binding*
DRB1*0101
YF
IY
T
P
L
A
N
P
S
DRB1*0401
VW
VY
L
A
A
S
PV
L
N
DRB1*0701
Y
P
V
L
T
A
V
N
SV
I-A
b
P
K
A
L
LT
M
R
A
M
AC
FT
I-A
d
CTW
AGM
T
CMSW
QLV
L
AIY
GIL
I-A
k
T
T
G
C
GS
F
EY
Q
C
I-A
s
F
C
NL
F
A
I
I
G
H
I-E
d
M
Q
W
W
S
A
W
R
C
I-E
k
I
A
Y
R
R
Q
L
T
A
Disfavored binding*
DRB1*0101
F
N
K
G
D
K
A
E
D
DRB1*0401
T
T
CEHS
AFI
R
R
GNQ
DI
S
DRB1*0701
GI
AV
C
N
N
QS
ST
DE
AR
I-A
b
DLR
TV
DFM
ADLN
SW
D
FLT
DLQ
A
I-A
d
G
M
N
E
PY
P
V
R
A
I-A
k
L
N
L
E
S
K
H
E
AQ
I-A
s
L
L
A
C
R
R
L
A
Y
I-E
d
Q
A
F
T
S
F
A
R
K
I-E
k
A
N
G
V
V
G
R
S
Y
* A cut-off value of > +/- 0.4 is applied to favored and disfavored binding amino acids, with
residues outside the limit in italics.
