I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
.
14
, N
o.
3
,
J
une
2025
, pp.
2328
~
2337
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
14
.i
3
.pp
2328
-
2337
2328
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
O
p
t
i
m
i
z
i
n
g b
i
oi
n
f
or
m
at
i
c
s ap
p
l
i
c
at
i
on
s:
a n
ove
l
ap
p
r
oac
h
w
i
t
h
h
u
m
a
n
p
r
ot
e
i
n
d
at
a an
d
d
at
a m
i
n
i
n
g
t
e
c
h
n
i
q
u
e
s
P
r
e
e
t
i
T
h
ar
e
j
a, R
aj
e
n
d
e
r
S
in
gh
C
h
h
il
la
r
D
e
pa
r
t
m
e
nt
of
C
om
put
e
r
S
c
i
e
nc
e
a
nd A
ppl
i
c
a
t
i
ons
, M
a
ha
r
s
hi
D
a
ya
n
a
nd U
ni
ve
r
s
i
t
y,
R
oht
a
k
, I
ndi
a
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
J
un 17, 2024
R
e
vi
s
e
d
F
e
b 8, 2025
A
c
c
e
pt
e
d
M
a
r
15, 2025
Biomedicine
plays
a
crucia
l
role
in
medical
resea
rch,
particu
larly
in
optimizing
techniques
for
disease
prediction.
However,
selecting
ef
fective
optimization
methods
and
managing
vast
amounts
of
medical
dat
a
pose
significant
challenges.
This
study
introduces
a
novel
optimization
technique,
integrated
bioinformatic
s
optimization
model
(IBOM)
for
disease
diagnosis,
incorpora
ting
data
mining
to
efficie
ntly
store
large
datasets
for
future
analysis
.
Various
optimi
zation
algorit
hms,
such
as
whale
optim
ization
alg
orithm
(
WOA),
multi
-
verse
optimization
(
MVO),
genetic
alg
orithm
(
GA),
and
ant
colony
optimiz
ation
(
ACO),
were
compared
wi
th
the
proposed
method.
The
evaluation
focused
on
metrics
like
acc
uracy,
specificity,
sensitivity,
precision,
F
-
score,
error,
receiver
op
erating
characterist
ic
(ROC)
,
and
false
positive
rate
(FPR)
using
5
-
fold
cross
-
validation.
Results
indicated
that
the
5
-
fold
cross
-
validation
method
achieved
superior
performance
with
metrics:
98.61%
accuracy,
9
6.59%
specificity,
88.63%
sensitivity,
99.30%
precision,
92.31%
F
-
score,
1
0.80%
error,
92.61%
ROC,
and
a
3.
00%
FPR
.
This
method
was
found
to
be
the
most
effective,
achieving
an
accuracy
of
0.92
in
disease
diagnosis
co
mpared
to other optimiz
ation techn
iques.
K
e
y
w
o
r
d
s
:
5
-
f
ol
d c
r
os
s
va
li
da
ti
on
B
io
in
f
or
m
a
ti
c
s
D
a
ta
m
in
in
g
D
e
e
p l
e
a
r
ni
ng t
e
c
hni
que
s
M
a
c
hi
ne
l
e
a
r
ni
ng t
e
c
hni
qu
e
s
O
pt
im
iz
a
ti
on mode
ls
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
P
r
e
e
ti
T
ha
r
e
ja
D
e
pa
r
tm
e
nt
of
C
om
put
e
r
S
c
ie
nc
e
a
nd A
ppl
ic
a
ti
ons
,
M
a
ha
r
s
hi
D
a
ya
na
nd U
ni
ve
r
s
it
y
R
oht
a
k, H
a
r
ya
na
,
I
ndi
a
E
m
a
il
:
pr
e
e
ti
th
a
r
e
ja
10@
gm
a
il
.c
om
1.
I
N
T
R
O
D
U
C
T
I
O
N
F
or
s
e
ve
r
a
l
de
c
a
de
s
,
e
xt
r
e
m
e
va
r
ia
ti
ons
a
nd
e
nha
nc
e
m
e
n
ts
ha
ve
be
c
om
e
w
it
ne
s
s
e
s
in
th
e
bi
om
e
di
c
in
e
a
r
e
a
s
[
1]
. T
he
r
e
f
or
e
, t
he
f
r
ont
ie
r
a
nd
a
s
s
oc
ia
ti
ve
a
r
e
a
s
a
r
e
de
r
iv
e
d f
r
om
s
e
ve
r
a
l
ki
nds
of
r
e
s
e
a
r
c
h
a
nd
th
e
or
ie
s
of
c
om
pr
e
he
ns
iv
e
m
e
di
c
in
e
,
bi
ol
ogy,
a
nd
li
f
e
s
c
ie
nc
e
.
T
h
e
f
oc
us
of
th
e
t
a
s
k
i
s
to
e
m
pl
oy
e
ngi
ne
e
r
in
g
a
nd
bi
ol
ogy
m
e
th
ods
to
r
e
s
e
a
r
c
h
a
nd
r
e
s
ol
ve
is
s
ue
s
in
li
f
e
s
c
ie
nc
e
,
s
pe
c
if
ic
a
ll
y
in
m
e
di
c
a
l
f
ie
ld
s
.
M
or
e
ove
r
,
b
io
m
e
di
c
in
e
i
s
c
on
s
id
e
r
e
d
th
e
s
ig
ni
f
ic
a
nt
r
e
s
e
a
r
c
h
a
nd
or
ig
in
a
ti
on
of
bi
om
e
di
c
a
l
da
t
a
,
ge
ne
c
hi
ps
,
na
not
e
c
hnol
ogy,
im
a
gi
ng
te
c
hnol
ogy,
a
nd
ne
w
m
a
te
r
ia
l
[
2]
,
[
3]
.
T
he
da
ta
m
in
in
g
m
e
th
ods
h
a
s
b
e
e
n
ut
il
iz
e
d
in
t
he
e
xi
s
ti
ng r
e
s
e
a
r
c
h t
ha
t
ne
e
ds
t
he
l
a
r
ge
s
t
s
to
r
a
ge
de
vi
c
e
s
a
n
d hi
gh
-
c
a
pa
bi
li
ty
a
na
ly
s
is
t
ool
s
.
T
he
a
b
s
tr
a
c
t
da
ta
f
e
a
tu
r
e
s
r
e
pr
e
s
e
nt
a
ti
on
is
c
a
r
r
ie
d
out
to
obt
a
i
n
a
r
e
li
a
bl
e
a
nd
pr
e
c
is
e
pe
r
f
or
m
a
nc
e
.
N
e
ve
r
th
e
le
s
s
,
hum
a
n
a
b
s
tr
a
c
ti
on
a
nd
da
ta
a
na
ly
s
i
s
a
r
e
in
a
ppr
opr
ia
te
f
or
a
la
r
ge
a
m
ount
of
d
a
ta
w
it
h
hi
gh
di
m
e
ns
io
na
l
s
e
ve
r
a
l
num
be
r
s
of
oc
c
ur
r
e
nc
e
s
.
F
ur
th
e
r
,
th
e
da
t
a
gr
ow
th
r
a
te
is
f
a
s
te
r
c
om
pa
r
e
d
to
th
e
m
a
nua
l
a
na
ly
s
is
.
T
he
r
e
f
or
e
,
it
is
di
f
f
ic
ul
t
to
tr
a
ns
la
te
th
e
r
a
w
da
ta
in
t
o
a
n
unde
r
s
ta
nda
bl
e
w
a
y
in
or
de
r
to
pr
ovi
de
us
e
r
s
w
it
h
unde
r
s
ta
nda
bl
e
or
ig
in
a
l
da
ta
.
T
he
s
e
ga
th
e
r
e
d
da
ta
s
houl
d
be
ut
il
iz
e
d
pr
ope
r
ly
in
or
de
r
to
he
lp
i
n
th
e
c
li
ni
c
a
l
di
a
gno
s
is
a
nd
a
ll
ow
s
to
de
f
in
e
of
th
e
dr
ug
e
f
f
e
c
ts
dur
in
g
th
e
e
xpe
r
im
e
nt
a
ti
on
of
d
a
ta
;
he
nc
e
c
e
r
ta
in
ly
, i
t
is
s
ig
ni
f
ic
a
nt
t
o e
na
bl
e
t
he
a
ut
om
a
ti
c
da
t
a
a
na
ly
s
i
s
t
e
c
hni
que
f
or
e
va
lu
a
ti
ng t
he
hi
gh
-
le
ve
l
da
ta
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
O
pt
imi
z
in
g bi
oi
nf
or
m
at
ic
s
appli
c
at
io
ns
:
a nov
e
l
app
r
oac
h w
it
h
hum
an pr
ot
e
in
dat
a
…
(
P
r
e
e
ti
T
har
e
ja
)
2329
R
e
c
e
nt
ly
,
v
a
r
i
ou
s
f
ie
ld
s
[
4]
,
i
nc
l
ud
in
g
r
e
ta
il
c
om
m
uni
c
a
ti
o
n,
b
a
n
ki
ng
, a
n
d m
e
di
c
a
l
d
ia
gno
s
ti
c
s
[
5]
,
[
6]
,
w
it
h
a
p
pr
opr
ia
te
d
a
t
a
a
nd
kn
ow
le
dg
e
t
h
a
t
w
a
s
of
te
n
hi
d
de
n.
M
o
s
t
of
t
he
ti
m
e
,
pr
o
c
e
s
s
in
g
l
a
r
g
e
d
a
t
a
a
n
d
e
xt
r
a
c
ti
ng
a
ppr
opr
ia
t
e
d
a
t
a
f
r
o
m
th
e
c
o
m
pl
e
x
ta
s
k
.
T
h
e
r
e
f
or
e
,
da
t
a
m
i
ni
n
g
i
s
c
on
s
id
e
r
e
d
a
po
w
e
r
f
ul
t
ool
f
or
m
a
n
a
g
in
g
t
a
s
k
s
,
e
s
p
e
c
ia
ll
y
i
n
th
e
m
e
di
c
a
l
f
ie
ld
.
T
h
e
c
la
s
s
if
ic
a
t
io
n
of
d
a
t
a
[
7]
c
a
n
be
u
ti
l
iz
e
d
t
o
id
e
nt
if
y
th
e
r
e
s
ul
ts
of
va
r
io
us
di
s
e
a
s
e
s
in
or
d
e
r
t
o
d
e
t
e
r
m
in
e
t
he
g
e
n
e
ti
c
be
h
a
vi
or
s
.
F
ur
t
he
r
,
v
a
r
io
u
s
te
c
h
ni
qu
e
s
a
r
e
u
s
e
d
i
n
th
e
e
xi
s
ti
ng
m
e
th
od
s
to
c
l
a
s
s
if
y
a
nd
pr
e
di
c
t
t
he
c
a
n
c
e
r
p
a
t
te
r
n
s
a
nd
c
o
m
p
a
r
e
d
w
it
h t
he
t
r
e
e
c
la
s
s
if
ie
r
s
[
8]
.
A
s
ur
ve
y ha
s
be
e
n c
onduc
te
d t
o pr
ovi
de
a
n
a
s
pe
c
t
of
va
r
io
us
m
ode
ls
pr
e
s
e
nt
e
d w
it
h t
he
ut
il
iz
e
d da
ta
m
in
in
g
-
da
ta
m
in
in
g
te
c
hni
que
s
[
9]
,
[
10
]
.
T
he
r
e
a
r
e
va
r
io
us
da
ta
m
in
in
g
m
e
th
ods
[
11]
,
in
c
lu
di
ng
s
uppor
t
ve
c
to
r
m
a
c
hi
ne
(
S
V
M
)
,
c
lu
s
te
r
in
g
a
lg
or
it
hm
s
,
a
r
ti
f
ic
ia
l
in
te
ll
ig
e
nc
e
(
A
I
)
,
ge
ne
ti
c
a
lg
or
it
hm
(
G
A
)
,
ne
ur
a
l
ne
twor
k
(
N
N
)
,
de
c
is
io
n
t
r
e
e
(
D
T
)
,
a
nd
na
iv
e
B
a
ye
s
(
N
B
)
.
M
or
e
ove
r
,
va
r
io
us
s
tu
di
e
s
[
12]
ha
ve
be
e
n
c
onduc
te
d
to
e
nha
nc
e
th
e
a
c
c
ur
a
c
y
of
th
e
pr
e
di
c
ti
on
m
ode
l
[
13]
ut
il
iz
in
g
pa
r
ti
c
ul
a
r
m
e
th
ods
or
in
te
gr
a
ti
on
o
f
two e
f
f
e
c
ti
ve
m
e
th
ods
.
T
he
m
a
in
c
ont
r
ib
ut
io
n
of
th
e
pr
opos
e
d
nove
l
in
te
gr
a
te
d
bi
oi
nf
or
m
a
ti
c
s
opt
im
iz
a
ti
on
m
ode
l
(
I
B
O
M
)
is
to
a
ppl
y
v
a
r
io
us
opt
im
iz
a
ti
on
a
lg
or
it
hm
s
in
th
e
bi
oi
nf
or
m
a
ti
c
s
a
ppl
ic
a
ti
on
in
or
de
r
to
e
nha
nc
e
th
e
a
c
c
ur
a
c
y
w
it
h
th
e
a
ppr
opr
ia
te
e
xpe
r
im
e
nt
a
l
d
e
ta
il
s
a
nd
m
e
th
ods
im
pl
e
m
e
nt
e
d.
M
or
e
ove
r
,
th
e
a
na
ly
s
is
of
f
iv
e
di
f
f
e
r
e
nt
opt
im
iz
a
ti
on
a
lg
or
it
hm
s
,
na
m
e
ly
,
w
ha
le
opt
im
iz
a
ti
on
a
lg
o
r
i
th
m
(
W
O
A
)
,
G
A
,
a
nt
c
ol
ony
opt
im
iz
a
ti
on
a
lg
or
it
hm
(
A
C
O
)
,
m
ul
ti
-
ve
r
s
e
opt
im
iz
a
ti
on
a
lg
or
it
hm
(
M
V
O
)
,
a
nd
5
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on
a
lg
or
it
hm
,
ha
s
be
e
n
di
s
c
us
s
e
d.
S
e
ve
r
a
l
pe
r
f
or
m
a
nc
e
m
e
tr
ic
s
,
in
c
lu
di
ng
a
c
c
ur
a
c
y,
pr
e
c
is
io
n,
r
e
c
a
ll
,
F
-
s
c
or
e
,
r
e
c
e
iv
e
r
ope
r
a
ti
ng
c
ha
r
a
c
te
r
is
ti
c
(
R
O
C
)
, e
r
r
or
,
a
nd
f
a
ls
e
pos
it
iv
e
r
a
te
(
F
P
R
)
,
a
r
e
us
e
d
to
a
s
s
e
s
s
th
e
e
f
f
e
c
ti
ve
ne
s
s
of
th
e
s
ugge
s
te
d t
e
c
hni
que
. T
h
e
c
or
r
e
s
ponding f
in
di
ngs
a
r
e
t
he
n di
s
c
u
s
s
e
d.
T
he
c
oor
di
na
ti
on
of
th
is
a
r
ti
c
le
is
a
s
f
ol
lo
w
s
.
S
e
c
ti
on
2
of
f
e
r
s
a
s
ynops
is
of
pr
io
r
w
or
ks
on
pr
ot
e
in
-
pr
ot
e
in
in
te
r
a
c
ti
on
(
PPI
)
pr
e
di
c
ti
on
.
S
e
c
ti
on
3
de
s
c
r
ib
e
s
our
s
u
gge
s
te
d
m
e
th
od
f
or
nove
l
I
B
O
M
opt
im
iz
a
ti
on
m
ode
l
.
T
he
e
xpe
r
im
e
nt
f
in
di
ngs
a
r
e
s
how
n
in
s
e
c
ti
on
4
.
L
a
s
tl
y,
s
e
c
ti
on
5
c
onc
lu
de
s
th
e
w
or
k,
a
nd
th
e
f
ol
lo
w
in
g s
e
c
ti
on l
is
ts
r
e
f
e
r
e
nc
e
s
.
2.
L
I
T
E
R
A
T
U
R
E
R
E
V
I
E
W
H
u
a
nd
O
hue
[
14]
de
m
ons
tr
a
te
s
S
pa
ti
a
lP
P
I
,
a
te
c
hni
que
th
a
t
f
or
e
c
a
s
ts
P
P
I
s
by
a
n
a
ly
z
in
g
pr
ot
e
in
c
om
pl
e
xe
s
pr
e
di
c
te
d
by
th
e
A
lp
ha
F
ol
d
m
ul
ti
m
e
r
us
in
g
de
e
p
ne
ur
a
l
ne
twor
ks
(
D
N
N
)
.
B
y
c
onv
e
r
ti
ng
th
e
a
to
m
ic
c
oor
di
na
te
a
nd
c
om
put
in
g
th
e
a
to
m
ic
di
s
tr
ib
ut
io
n,
th
e
pr
ot
e
in
c
om
pl
e
xe
s
m
a
p
s
w
e
r
e
f
ound.
T
hi
s
m
e
th
od
us
e
s
s
ophi
s
ti
c
a
te
d
im
a
ge
pr
oc
e
s
s
in
g
te
c
hni
que
s
to
r
e
tr
ie
ve
im
por
ta
nt
th
r
e
e
-
di
m
e
ns
io
na
l
s
tr
uc
tu
r
a
l
da
ta
f
r
om
pr
ot
e
in
c
om
pl
e
xe
s
.
T
he
s
ugge
s
te
d
a
ppr
oa
c
h
pr
e
di
c
ts
P
P
I
s
w
it
h
e
nc
our
a
gi
ng
f
in
di
ngs
,
de
m
ons
tr
a
ti
ng
th
e
pos
s
ib
il
it
ie
s
of
3D
s
p
a
ti
a
l
r
e
nde
r
in
g m
e
th
ods
t
o f
ur
th
e
r
s
tr
uc
tu
r
a
l
bi
ol
ogy
r
e
s
e
a
r
c
h.
C
a
o
e
t
al
.
[
15]
pr
ovi
de
a
p
r
e
li
m
in
a
r
y
in
f
or
m
a
ti
on
f
us
io
n
-
ba
s
e
d
node
r
e
pr
e
s
e
nt
a
ti
on
te
c
hni
que
th
a
t
us
e
s
in
te
r
a
c
ti
on
a
nd
s
e
que
nc
in
g
ne
twor
k
pr
of
il
e
s
to
pr
e
s
e
nt
pr
ot
e
in
f
e
a
tu
r
e
in
f
or
m
a
ti
on.
T
o
be
m
or
e
pr
e
c
is
e
,
pr
ot
e
in
i
nt
e
r
a
c
ti
on p
r
of
il
e
a
nd p
r
ot
e
in
s
e
que
nc
e
da
ta
a
r
e
r
e
c
or
d
e
d us
in
g di
s
ta
nc
e
m
e
tr
ic
s
. A
w
e
ig
ht
e
d f
e
a
tu
r
e
s
f
us
io
n
te
c
hni
que
is
us
e
d
to
s
ta
bl
e
th
e
w
e
ig
ht
s
of
th
e
two
s
our
c
e
s
of
da
ta
w
it
h
a
w
e
ig
ht
pa
r
a
m
e
te
r
to
ge
ne
r
a
te
a
n
in
it
ia
l
in
f
or
m
a
ti
on
m
a
tr
ix
.
T
he
f
e
a
tu
r
e
s
of
pr
ot
e
in
s
a
r
e
th
e
n
r
e
pr
e
s
e
nt
e
d
by
tr
a
in
in
g
a
s
ta
c
ke
d
a
ut
oe
n
c
ode
r
(
S
A
E
)
a
r
c
hi
te
c
tu
r
e
on
th
e
or
ig
in
a
l
da
ta
f
us
io
n
m
a
tr
ix
.
F
in
a
ll
y
,
dow
ns
tr
e
a
m
pr
e
di
c
ti
on
ta
s
ks
a
r
e
pe
r
f
or
m
e
d
us
in
g
a
n
S
V
M
c
la
s
s
if
ie
r
.
U
s
in
g
a
5
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on
pr
oc
e
dur
e
,
th
e
m
e
th
od
a
tt
a
in
e
d
a
n
a
c
c
ur
a
c
y
of
97.69%
f
or
t
he
H
om
o s
a
pi
e
ns
da
ta
s
e
t,
a
ll
ow
in
g f
or
a
f
ul
l
a
s
s
e
s
s
m
e
nt
of
i
ts
pe
r
f
or
m
a
nc
e
.
G
ündüz
e
t
al
.
[
16]
pr
e
s
e
nt
e
d
G
e
nom
e
N
e
t
-
A
r
c
hi
te
c
t
is
a
ne
u
r
a
l
a
r
c
hi
te
c
tu
r
e
de
s
ig
n
pl
a
tf
or
m
th
a
t
a
ut
om
a
ti
c
a
ll
y
opt
im
iz
e
s
de
e
p
le
a
r
ni
ng
m
ode
ls
ba
s
e
d
on
ge
nom
ic
s
e
que
nc
e
da
ta
.
I
t
a
dj
us
ts
th
e
a
r
c
hi
te
c
tu
r
e
'
s
ge
ne
r
a
l
de
s
ig
n,
in
c
lu
di
ng
a
ge
nom
ic
s
-
s
pe
c
if
ic
s
e
a
r
c
h s
pa
c
e
.
I
t
a
ls
o
opt
im
iz
e
s
th
e
m
ode
l
tr
a
in
in
g
te
c
hni
que
a
s
w
e
ll
a
s
th
e
hype
r
pa
r
a
m
e
te
r
s
of
in
di
vi
dua
l
la
ye
r
s
.
I
n
c
om
pa
r
is
on
to
th
e
to
p
-
pe
r
f
or
m
in
g
de
e
p
le
a
r
ni
ng
ba
s
e
li
ne
s
,
G
e
nom
e
N
e
t
-
A
r
c
hi
te
c
t
r
e
duc
e
d
th
e
m
is
in
te
r
pr
e
ta
ti
on
r
a
te
by
19%
on
a
vi
r
a
l
c
a
te
gor
iz
a
ti
on
t
e
s
t,
r
e
qui
r
in
g 67%
l
e
s
s
t
im
e
f
or
pr
e
di
c
ti
on a
nd 83%
f
e
w
e
r
m
e
tr
ic
s
t
o r
e
a
c
h s
im
il
a
r
c
ont
ig
-
le
ve
l
a
c
c
ur
a
c
y.
D
a
ng
a
nd
V
u
[
17]
in
tr
oduc
e
xC
A
P
T
5,
a
nove
l
hybr
id
c
la
s
s
if
ie
r
th
a
t
us
e
s
th
e
T
5
-
XL
-
U
ni
R
e
f
50
pr
ot
e
in
la
r
ge
la
ngua
ge
m
ode
l
to
ge
n
e
r
a
te
r
ic
h
a
m
in
o
a
c
id
e
m
b
e
ddi
ngs
f
r
om
pr
ot
e
in
s
e
que
nc
e
s
.
T
he
he
a
r
t
of
xC
A
P
T
5
is
a
m
ul
ti
-
ke
r
ne
l
de
e
p
c
onvolut
io
na
l
n
e
ur
a
l
ne
twor
k
(
C
N
N
)
th
a
t
s
uc
c
e
s
s
f
ul
ly
c
a
pt
ur
e
s
c
om
pl
ic
a
te
d
c
ol
la
bor
a
ti
ve
in
f
or
m
a
ti
on
a
t
th
e
s
m
a
ll
a
nd
bi
g
le
ve
l
s
.
I
t
is
m
e
r
ge
d
w
it
h
th
e
X
G
B
oos
t
a
lg
or
it
hm
a
nd
c
onc
a
te
na
te
d
w
it
h
pool
in
g
f
e
a
tu
r
e
s
in
de
e
p
th
a
t
m
a
ke
s
xC
A
P
T
5
to
le
a
r
n
im
por
ta
nt
ve
c
to
r
s
w
it
h
li
tt
le
c
om
put
in
g
c
os
t.
E
xpe
r
im
e
nt
a
l
r
e
s
ul
ts
r
e
ve
a
l
th
a
t
x
C
A
P
T
5
s
ur
pa
s
s
e
s
m
a
ny
a
ppr
oa
c
he
s
in
pr
e
di
c
ti
ng
bi
na
r
y
P
P
I
,
out
s
ta
ndi
ng
a
t
c
r
os
s
-
va
li
da
ti
on on numer
ous
da
ta
s
e
t
s
.
A
hm
e
d
e
t
al
.
[
18]
bui
ld
a
nove
l
m
e
th
od
th
a
t
c
om
bi
ne
s
s
e
v
e
r
a
l
ty
pe
s
of
s
m
a
r
t
l
a
ye
r
s
a
nd
NN
.
F
oc
us
in
g
on
th
e
m
in
or
de
ta
il
s
of
s
e
que
nc
e
s
of
a
m
in
o
a
c
id
s
,
it
is
a
nt
ic
ip
a
te
d
to
de
ve
lo
p
m
or
e
a
c
c
ur
a
te
pr
e
di
c
ti
ons
r
e
ga
r
di
ng
pr
ot
e
in
s
a
nd
e
xt
r
a
c
t
c
ha
r
a
c
te
r
is
ti
c
s
.
T
he
a
im
is
to
ve
r
if
y
th
e
nove
l
a
ppr
oa
c
h
e
f
f
e
c
ti
ve
ne
s
s
by
te
s
ti
ng
it
a
t
a
br
oa
d
le
ve
l,
w
hi
c
h
in
c
r
e
a
s
e
s
th
e
c
om
pr
e
he
ns
io
n
of
bi
ol
ogy
a
nd
bi
oi
nf
or
m
a
ti
c
s
.
I
t
c
r
e
a
te
d
a
c
us
to
m
DNA
-
bi
ndi
ng
pr
ot
e
in
s
(
D
B
P
s
)
s
or
ti
ng
a
r
c
hi
te
c
tu
r
e
a
nd
im
pr
ove
d
it
to
be
e
xc
e
pt
io
na
ll
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 3, J
une
2025
:
2328
-
2337
2330
pr
e
c
is
e
a
nd
us
e
f
ul
.
T
he
f
in
di
ngs
s
ugge
s
t
th
a
t
th
is
s
tr
a
te
gy
is
e
x
tr
e
m
e
ly
e
f
f
e
c
ti
ve
in
de
te
c
ti
ng
hi
dde
n
pa
tt
e
r
ns
in
m
a
s
s
iv
e
da
ta
s
e
ts
.
Y
u
e
t
al
.
[
19]
pr
e
s
e
nt
e
d
a
nove
l
gr
a
di
e
nt
tr
e
e
boos
ti
ng
(
G
T
B
)
-
ba
s
e
d
P
P
I
pr
e
di
c
ti
on
pi
pe
li
ne
.
F
ir
s
t,
th
e
ps
e
udo
a
m
in
o
a
c
id
c
om
pos
it
io
n
(
P
s
e
A
A
C
)
,
p
s
e
udo
pos
it
io
n
-
s
pe
c
if
ic
s
c
or
in
g
m
a
tr
ix
(
P
s
e
P
S
S
M
)
,
r
e
duc
e
d
s
e
que
nc
e
a
nd
in
de
x
-
ve
c
to
r
s
(
R
S
I
V
)
,
a
nd
a
ut
oc
or
r
e
la
ti
on
de
s
c
r
ip
to
r
(
A
D
)
a
r
e
f
us
e
d
to
r
e
c
ove
r
th
e
in
it
ia
l
f
e
a
tu
r
e
ve
c
to
r
.
S
e
c
ond,
L
1
-
r
e
gul
a
r
iz
e
d
lo
gi
s
ti
c
r
e
gr
e
s
s
io
n
(
L
1
-
R
L
R
)
is
e
m
pl
oye
d
to
c
hoo
s
e
th
e
be
s
t
f
e
a
tu
r
e
s
ubs
e
t
a
nd
e
li
m
in
a
te
noi
s
e
a
nd
r
e
dunda
n
c
y.
U
lt
im
a
te
ly
,
th
e
G
T
B
-
P
P
I
m
ode
l
is
bui
lt
.
U
s
in
g
th
e
H
om
o
s
a
pi
e
n
s
da
ta
s
e
t,
G
T
B
-
P
P
I
obt
a
in
e
d
95.15%
a
c
c
ur
a
c
y,
a
c
c
or
di
ng
to
f
iv
e
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on.
F
ur
th
e
r
m
or
e
,
G
T
B
-
P
P
I
m
a
y
be
ut
il
iz
e
d
f
or
pr
e
di
c
ti
ng
in
de
pe
nde
nt
te
s
t
da
ta
s
e
ts
,
a
nd
th
e
out
c
om
e
s
de
m
ons
tr
a
te
th
a
t
it
c
a
n
gr
e
a
tl
y
in
c
r
e
a
s
e
P
P
I
pr
e
di
c
ti
on a
c
c
ur
a
c
y.
S
im
s
e
k
e
t
al
.
[
20]
de
s
c
r
ib
e
d a
hyb
r
id
-
da
ta
m
in
in
g
-
da
ta
m
in
in
g
-
ba
s
e
d t
e
c
hni
que
ha
s
be
e
n ut
il
iz
e
d t
o
di
s
ti
ngui
s
h
th
e
s
ig
ni
f
ic
a
nt
va
r
ia
bl
e
s
us
e
d
f
or
s
ur
vi
va
l
c
h
a
nge
a
nd
in
di
a
gno
s
in
g
br
e
a
s
t
c
a
nc
e
r
.
H
e
n
c
e
th
e
s
ig
ni
f
ic
a
nc
e
of
va
r
ia
bl
e
s
w
a
s
d
e
te
r
m
in
e
d
f
or
di
f
f
e
r
e
nt
pe
r
io
ds
m
e
a
s
ur
e
d
a
s
one
,
f
iv
e
,
a
nd
te
n.
F
ur
th
e
r
,
th
e
pa
r
s
im
oni
ous
m
ode
ls
a
r
e
ut
il
iz
e
d
to
pe
r
f
or
m
di
f
f
e
r
e
nt
a
na
ly
s
e
s
by
e
xe
c
ut
in
g
one
-
r
e
gr
e
s
s
io
n
a
na
ly
s
is
te
c
hni
que
s
s
u
c
h
a
s
m
e
ta
he
ur
is
ti
c
opt
im
iz
a
ti
on
te
c
hni
que
s
,
G
A
a
nd
le
a
s
t
a
b
s
ol
ut
e
s
hr
in
ka
ge
a
nd
s
e
le
c
ti
on
ope
r
a
to
r
(
L
A
S
S
O
)
.
T
he
r
e
f
or
e
,
two
w
e
ll
-
known
r
e
s
a
m
pl
in
g
s
our
c
e
s
,
s
ynt
he
ti
c
m
in
or
it
y
ove
r
-
s
a
m
pl
in
g
te
c
hni
que
(
S
M
O
T
E
)
a
nd
r
a
ndom
unde
r
-
s
a
m
pl
in
g
(
R
U
S
)
,
w
e
r
e
e
m
pl
oye
d
to
e
nha
nc
e
th
e
c
la
s
s
if
ic
a
ti
on
m
ode
l
pe
r
f
or
m
a
nc
e
.
E
ve
nt
ua
ll
y,
th
e
two
da
ta
m
in
in
g
m
ode
ls
,
in
c
lu
di
ng
lo
gi
s
ti
c
r
e
gr
e
s
s
io
n
(
L
R
)
a
nd
a
r
ti
f
ic
ia
l
ne
ur
a
l
ne
twor
ks
(
A
N
N
)
w
e
r
e
u
s
e
d
w
it
h
10
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on.
H
o
w
e
ve
r
,
s
ti
ll
,
th
e
s
e
te
c
hni
qu
e
s
a
r
e
not
a
ppl
ie
d
to
ot
he
r
c
a
nc
e
r
t
ype
s
.
V
ouga
s
e
t
al
.
[
21]
m
a
in
ly
f
oc
u
s
e
d
on
un
s
upe
r
vi
s
e
d
a
nd
s
upe
r
vi
s
e
d
te
c
hni
que
s
ut
il
iz
e
d
e
xpl
ic
it
ly
in
pr
e
di
c
ti
on
a
ppl
ic
a
ti
ons
(
dr
ug
r
e
s
pons
e
)
,
e
nh
a
nc
e
m
e
nt
of
va
r
io
us
te
c
hni
que
s
in
a
ppl
ic
a
bl
e
m
ode
l
s
,
a
nd
im
pr
ove
d
m
ode
l
pe
r
f
or
m
a
nc
e
.
F
ur
th
e
r
,
a
s
il
ic
a
-
s
c
r
e
e
ni
ng
pr
oc
e
s
s
w
a
s
a
ls
o
us
e
d
in
a
c
c
or
da
nc
e
w
it
h
a
s
s
oc
ia
ti
on
r
ul
e
-
m
in
in
g
f
or
de
f
in
in
g
ge
ne
s
.
I
nc
or
por
a
ti
ng
om
ic
s
in
f
or
m
a
ti
on
la
ye
r
s
s
uc
h
a
s
m
e
ta
bol
om
ic
s
,
in
te
r
a
c
to
m
ic
s
,
phos
pho
-
pr
ot
e
om
ic
s
,
pr
ot
e
om
ic
s
,
a
nd
m
e
ta
-
ge
n
om
ic
s
im
pr
ove
s
th
e
m
e
th
od'
s
a
ppl
ic
a
bi
li
ty
a
nd
e
nha
nc
e
s
th
e
s
il
ic
o
-
pr
oc
e
s
s
m
e
th
od.
H
ow
e
v
e
r
,
th
e
s
il
ic
o
pi
pe
li
n
e
s
im
pa
c
t
to
a
c
e
r
ta
in
le
ve
l
f
r
om
ne
ga
ti
ve
a
nd
f
a
ls
e
pos
it
iv
e
out
c
om
e
s
.
T
ha
kka
r
e
t
al
.
[
22]
p
r
e
s
e
nt
e
d
f
uz
z
y l
ogi
c
a
nd
da
ta
m
in
in
g
m
e
th
ods
a
r
e
ut
il
iz
e
d i
n di
a
be
te
s
di
a
gnos
is
;
th
e
s
e
a
r
e
us
e
d t
o l
oc
a
te
a
ppr
opr
ia
te
pa
tt
e
r
ns
i
n l
a
r
ge
da
ta
s
e
ts
us
in
g a
n i
nt
e
gr
a
ti
on of
va
r
io
us
m
a
c
hi
ne
l
e
a
r
ni
ng
(
ML
)
m
e
th
ods
,
s
ta
ti
s
ti
c
s
,
a
nd
m
a
ni
pul
a
ti
ons
.
T
he
e
xp
e
r
t
s
ys
te
m
s
li
ke
da
ta
m
in
in
g
a
nd
f
uz
z
y
lo
gi
c
a
r
e
u
s
e
d
f
or
di
f
f
e
r
e
nt
a
s
pe
c
ts
in
or
de
r
to
m
a
na
ge
th
e
unc
e
r
ta
in
ti
e
s
a
nd
f
in
d
th
e
hi
dde
n
in
f
or
m
a
ti
on.
F
ur
th
e
r
,
th
e
f
uz
z
y
e
xpe
r
t
s
ys
te
m
(
F
E
S
)
e
xa
m
in
e
d
th
e
in
f
or
m
a
ti
on
f
r
om
th
e
a
v
a
il
a
bl
e
da
ta
th
a
t
he
lp
s
to
in
di
c
a
te
li
ngui
s
ti
c
c
onc
e
pt
s
in
m
e
di
c
a
l
c
onc
e
pt
s
.
T
he
r
e
f
or
e
,
va
r
io
us
ta
s
k
s
ha
ve
b
e
e
n
pr
oc
e
s
s
e
d
w
hi
le
da
ta
s
e
t
s
e
l
e
c
ti
on
th
r
ough
pr
e
-
pr
oc
e
s
s
in
g
by
e
m
pl
oyi
ng
va
r
io
us
m
e
th
ods
,
in
c
lu
di
ng
nor
m
a
li
z
a
ti
on,
a
nd
s
ta
nd
a
r
di
z
a
ti
on.
N
e
xt
,
in
th
e
f
e
a
tu
r
e
e
xt
r
a
c
ti
on
pr
oc
e
s
s
,
a
n
e
f
f
e
c
ti
ve
m
e
th
od
in
c
lu
di
ng
f
uz
z
y
lo
gi
c
a
nd
D
M
on
di
f
f
e
r
e
nt
c
la
s
s
if
ic
a
ti
on
a
lg
or
it
hm
s
ha
s
be
e
n
u
s
e
d
to
e
nha
nc
e
pr
e
c
is
e
n
e
s
s
.
E
ve
nt
ua
ll
y,
th
e
r
a
ndom
f
or
e
s
t
(
R
F
)
m
e
th
od
pr
oc
ur
e
d
99.7%
,
a
nd
by
e
m
pl
oyi
ng
va
r
io
us
lo
gi
c
,
c
onc
e
pt
s
a
r
e
pr
oc
e
s
s
e
d
w
it
h
lo
w
c
om
pl
e
xi
ty
a
nd
hi
gh
pr
e
c
is
io
n
w
it
h
hi
gh pr
e
c
is
e
ne
s
s
of
96%
.
K
ova
lc
huk
e
t
al
.
[
23]
ut
il
iz
e
d
th
e
m
ul
ti
pl
e
c
onc
e
pt
ua
l
-
f
r
a
m
e
w
or
k
te
c
hni
que
s
to
in
te
gr
a
te
da
ta
a
na
ly
s
is
a
nd
pa
ti
e
nt
f
lo
w
.
A
n
a
s
s
o
c
ia
ti
on
of
pr
oc
e
s
s
m
in
in
g
m
e
th
ods
,
te
xt
,
a
nd
da
t
a
a
r
e
ut
il
iz
e
d
to
id
e
nt
if
y
a
nd
a
s
s
e
s
s
pa
ti
e
nt
f
lo
w
,
a
nd
c
li
ni
c
a
l
pa
th
w
a
ys
c
la
s
s
e
s
(
C
P
s
)
.
A
c
c
or
di
ngl
y,
th
is
te
c
hni
que
a
ll
ow
s
a
ut
om
a
ti
c
r
e
c
ogni
ti
on
of
pa
ti
e
nt
s
’
dyna
m
ic
s
on
a
c
e
r
ta
in
m
ic
r
o
-
le
ve
l
in
or
de
r
to
e
xe
c
ut
e
r
e
a
li
s
ti
c
s
im
ul
a
ti
on
s
a
nd
a
c
qui
r
e
m
a
c
r
o
-
le
ve
l
f
e
a
tu
r
e
s
,
in
c
lu
di
ng
que
ui
ng
pa
r
a
m
e
te
r
s
,
de
pa
r
tm
e
nt
a
l
lo
a
d,
a
nd
pa
ti
e
nt
e
xp
e
r
ie
nc
e
.
M
or
e
ove
r
,
th
e
a
ut
om
a
ti
c
c
la
s
s
if
ic
a
ti
on
a
nd
id
e
nt
if
ic
a
ti
on
of
C
P
s
ut
il
iz
a
ti
on
e
nha
nc
e
th
e
a
c
ut
e
c
or
ona
r
y
s
yndr
om
e
(
A
C
S
)
pa
ti
e
nt
di
s
c
r
e
te
-
e
ve
nt
s
im
ul
a
ti
on
pr
oc
e
s
s
.
H
ow
e
ve
r
,
s
ti
ll
,
d
a
ta
-
dr
iv
e
n
s
ol
ut
io
n
to
la
r
ge
da
ta
s
e
ts
i
s
di
f
f
ic
ul
t
dur
in
g i
m
pl
e
m
e
nt
a
ti
on.
Y
a
ng
e
t
al
.
[
24
]
de
ve
lo
pe
d
s
ta
te
-
of
-
th
e
-
a
r
t
M
L
te
c
hni
que
s
on
w
e
bs
e
r
ve
r
in
or
de
r
to
bui
ld
a
n
e
f
f
e
c
ti
ve
pr
e
di
c
ti
ve
te
c
hni
que
c
ove
r
in
g
c
r
uc
ia
l
a
bs
or
pt
io
n,
di
s
tr
ib
ut
io
n,
m
e
ta
bol
is
m
,
e
xc
r
e
ti
on,
a
nd
to
xi
c
it
y
(
A
D
M
E
T
)
f
e
a
tu
r
e
s
f
or
dr
ug
di
s
c
ove
r
y.
H
e
nc
e
,
a
dm
e
tS
A
R
-
A
D
M
E
T
th
a
t
de
s
ig
ne
d
w
it
h
m
e
di
c
in
a
l
c
he
m
i
s
ts
th
a
t
e
nha
nc
e
le
a
d
c
om
pone
nt
s
a
lo
ng
w
it
h
a
n
e
f
f
ic
ie
nt
A
D
M
E
T
pr
ope
r
ti
e
s
.
H
ow
e
ve
r
,
la
c
k
of
A
D
M
E
T
pr
ope
r
ti
e
s
i
n t
he
pr
e
di
c
ti
on on pr
a
c
ti
c
a
l
pl
a
tf
or
m
f
or
c
he
m
ic
a
l
r
e
s
e
a
r
c
h a
nd dr
ug dis
c
ove
r
y i
s
di
f
f
ic
ul
t.
3.
M
E
T
H
O
D
T
he
va
r
io
us
e
f
f
e
c
ti
ve
opt
im
iz
a
ti
on
m
ode
l
ha
s
be
e
n
ut
il
iz
e
d
in
bi
oi
nf
or
m
a
ti
c
s
a
ppl
ic
a
ti
ons
th
a
t
ha
ve
be
e
n pr
oc
e
s
s
e
d w
it
h t
he
huma
n pr
ot
e
in
da
ta
. T
hi
s
da
ta
s
e
t
ha
s
b
e
e
n pr
oc
e
s
s
e
d i
n di
f
f
e
r
e
nt
s
te
ps
, i
nc
lu
di
ng da
ta
c
ol
le
c
ti
on,
da
ta
nor
m
a
li
z
a
ti
on,
tr
a
in
in
g
a
nd
te
s
ti
ng,
a
nd
e
m
pl
oyi
ng
va
r
io
us
opt
im
iz
a
ti
on
te
c
hni
que
s
.
M
or
e
ove
r
,
th
e
m
a
in
nove
lt
y
of
th
e
w
or
k
is
th
e
in
te
gr
a
ti
on
of
f
iv
e
e
f
f
e
c
ti
ve
a
lg
or
it
hm
s
,
in
c
lu
di
ng
th
e
W
O
A
,
M
V
O
, G
A
, A
C
O
, a
nd 5
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
O
pt
imi
z
in
g bi
oi
nf
or
m
at
ic
s
appli
c
at
io
ns
:
a nov
e
l
app
r
oac
h w
it
h
hum
an pr
ot
e
in
dat
a
…
(
P
r
e
e
ti
T
har
e
ja
)
2331
3.1.
M
at
e
r
ia
ls
an
d
m
e
t
h
od
T
he
hum
a
n
pr
ot
e
in
da
ta
s
e
t
ha
s
be
e
n
ut
il
iz
e
d
in
th
e
e
xpe
r
im
e
n
ta
ti
on
pr
oc
e
s
s
.
V
a
r
io
us
opt
im
iz
a
ti
on
te
c
hni
que
s
a
r
e
ut
il
iz
e
d
a
nd
e
xe
c
ut
e
d
on
th
e
M
a
th
W
or
k
s
M
A
T
L
A
B
R
2020a
ve
r
s
io
n
1.0
to
f
in
d
th
e
opt
im
um
r
e
s
ul
ts
f
r
om
th
e
u
s
e
d
opt
im
iz
a
ti
on
te
c
hni
que
s
to
id
e
nt
if
y
t
he
e
f
f
ic
ie
nc
y
a
nd
a
c
c
ur
a
c
y
of
th
e
pr
opos
e
d
te
c
hni
que
.
T
he
r
e
f
or
e
,
th
e
pr
opos
e
d
te
c
hni
que
ha
s
be
e
n
c
om
pa
r
e
d
w
it
h
th
e
ot
he
r
te
c
hni
que
s
to
de
te
r
m
in
e
th
e
out
c
om
e
a
c
c
ur
a
c
y a
nd p
e
r
f
or
m
a
nc
e
e
va
lu
a
ti
on.
3.2.
D
at
a m
in
in
g op
t
im
iz
at
io
n
m
od
e
ls
T
he
a
na
ly
s
is
of
f
iv
e
di
f
f
e
r
e
nt
opt
im
iz
a
ti
on a
lg
or
it
hm
s
na
m
e
ly
;
W
O
A
, G
A
[
19]
, A
C
O
a
lg
or
it
hm
[
20
]
,
M
V
O
a
lg
or
it
hm
,
a
nd
5
-
f
ol
d
c
r
os
s
va
li
da
ti
on
a
lg
or
it
hm
ha
s
b
e
e
n
ut
il
iz
e
d
in
th
e
c
ur
r
e
nt
pr
opos
e
d
s
ys
t
e
m
.
T
he
be
ne
f
it
s
of
us
in
g di
f
f
e
r
e
nt
opt
im
iz
a
ti
on a
lg
or
it
hm
s
ha
ve
be
e
n f
ol
lo
w
e
d.
3.2.1.
Wh
al
e
op
t
im
iz
at
io
n
al
gor
it
h
m
T
he
W
O
A
is
c
ons
id
e
r
e
d
a
m
e
ta
-
he
ur
is
ti
c
opt
im
iz
a
ti
on
a
lg
or
i
th
m
th
a
t
he
lp
s
in
va
r
io
us
te
r
m
s
a
nd
f
in
ds
out
th
e
be
ha
vi
or
of
th
e
bubble
-
ne
t
hunt
in
g
of
hum
pba
c
k
w
ha
le
s
.
I
n
th
e
c
ur
r
e
nt
r
e
s
e
a
r
c
h,
th
is
a
lg
or
it
hm
ha
s
be
e
n
a
ppl
ie
d
to
th
e
pr
ot
e
in
da
ta
s
e
t
th
a
t
is
r
obus
t
a
nd
s
im
pl
e
,
a
nd
c
om
pl
e
te
ly
ba
s
e
d
on
th
e
s
to
c
ha
s
ti
c
-
s
w
a
r
m
-
ba
s
e
d
opt
im
iz
a
ti
on
a
lg
or
it
hm
[
25]
.
N
or
m
a
ll
y,
th
e
p
opul
a
ti
on
-
ba
s
e
d
W
O
A
ha
s
th
e
c
a
pa
bi
li
ty
to
r
e
m
ove
lo
c
a
l
opt
im
a
a
nd
pr
oc
ur
e
th
e
be
s
t
gl
oba
l
opt
im
a
l
s
ol
u
ti
on.
T
hi
s
a
lg
or
it
hm
he
lp
s
to
r
e
s
ol
ve
v
a
r
io
us
unc
ons
tr
a
in
e
d
a
nd
c
ons
tr
a
in
e
d
opt
im
iz
a
ti
on
is
s
ue
s
pr
oc
e
s
s
f
or
pr
a
c
ti
c
a
l
a
ppl
ic
a
ti
ons
in
th
e
a
bs
e
nc
e
of
s
tr
uc
tu
r
a
l
r
e
f
or
m
a
ti
on.
I
n
th
e
s
c
e
na
r
io
,
th
e
c
lu
s
te
r
in
g
is
s
ue
s
a
r
e
s
ol
ve
d
ut
il
iz
in
g
W
O
A
w
it
h
th
e
c
lu
s
te
r
in
g
c
ont
e
xt
r
e
pr
e
s
e
nt
e
d
w
it
h
k
-
c
lu
s
te
r
s
c
e
nt
e
r
s
.
H
e
n
c
e
,
e
ve
r
y
s
e
a
r
c
h
a
ge
nt
ha
s
be
e
n
c
on
s
tr
uc
te
d
a
s
X
i,
a
nd
th
e
m
a
th
e
m
a
ti
c
a
l
f
or
m
ul
a
i
s
a
s
s
how
n i
n (
1)
.
=
(
1
,
2
,
3
,
…
,
)
(
1)
H
e
r
e
,
k
is
num
be
r
o
f
c
lu
s
te
r
s
,
Z
ij
is
in
d
ic
a
te
s
th
e
jt
h
c
lu
s
te
r
c
e
nt
e
r
ve
c
to
r
th
a
t
de
not
e
s
th
e
i
th
s
e
a
r
c
h
-
a
ge
nt
in
c
lu
s
te
r
.
3.2.2.
G
e
n
e
t
ic
al
gor
it
h
m
A
G
A
is
c
ons
id
e
r
e
d
a
he
ur
is
ti
c
-
s
e
a
r
c
h
m
ode
l
us
e
d
in
va
r
i
ous
bi
oi
nf
or
m
a
ti
c
s
a
ppl
ic
a
ti
ons
a
nd
m
e
di
c
a
l
pur
pos
e
s
f
or
pr
e
di
c
ti
ng
va
r
io
us
di
s
e
a
s
e
s
.
I
t
is
ut
il
iz
e
d
to
a
na
ly
z
e
opt
im
iz
e
d
out
c
om
e
s
to
s
e
a
r
c
h
f
or
pr
obl
e
m
s
in
pr
e
di
c
ti
ng
th
e
va
r
io
us
a
c
ti
ons
ba
s
e
d
on e
vol
ut
io
na
r
y
bi
ol
ogy
a
nd
na
tu
r
a
l
s
e
l
e
c
ti
on.
B
a
s
ic
a
ll
y,
th
e
G
A
is
a
n
e
f
f
ic
ie
nt
m
ode
l
f
or
s
e
a
r
c
hi
ng
vi
a
c
om
pl
e
x
a
nd
huge
d
a
ta
s
e
ts
. T
he
r
e
f
or
e
,
th
e
y
ha
ve
th
e
c
a
pa
bi
li
ty
to
f
in
d
a
n
e
f
f
e
c
ti
ve
s
ol
ut
io
n
in
c
om
pl
e
x
e
nvi
r
onm
e
nt
s
,
e
s
pe
c
ia
ll
y
c
a
pa
bl
e
of
f
in
di
ng
a
nd
r
e
s
ol
vi
ng
c
ons
tr
a
in
e
d
a
nd
unc
ons
tr
a
in
e
d
opt
im
iz
a
ti
on
pr
obl
e
m
s
[
26]
.
T
he
f
oc
us
of
th
e
G
A
f
r
om
e
vol
ut
io
na
r
y
bi
ol
ogy
in
c
lu
de
s
r
e
c
om
bi
na
ti
on, s
e
le
c
ti
on, i
nhe
r
it
a
nc
e
, a
nd mut
a
ti
on i
n or
de
r
t
o r
e
s
ol
ve
t
he
i
s
s
ue
s
.
T
he
m
a
in
pur
pos
e
of
th
e
G
A
in
th
e
c
ur
r
e
nt
r
e
s
e
a
r
c
h
is
a
s
f
ol
lo
w
s
,
a
nd
it
c
om
pl
e
te
ly
va
r
ie
s
f
r
om
a
n
opt
im
iz
a
ti
on
a
lg
or
it
hm
,
de
r
iv
a
ti
ve
-
ba
s
e
d,
a
nd
c
la
s
s
ic
a
l
m
e
th
ods
in
two
f
or
m
s
:
F
ir
s
t,
G
A
pr
ovoking
th
e
popula
ti
on
in
e
ve
r
y
m
ove
,
w
he
r
e
in
a
tr
a
di
ti
ona
l
a
lg
or
it
hm
o
nl
y
pr
oduc
e
s
a
s
in
gl
e
poi
nt
a
t
e
ve
r
y
m
ove
.
S
e
c
ond,
G
A
pi
c
ki
ng
onl
y
th
e
s
ubs
e
que
nt
popula
ti
on
by
e
s
ti
m
a
ti
on
ut
il
iz
in
g
r
a
ndom
ge
ne
r
a
to
r
s
,
w
he
r
e
in
a
tr
a
di
ti
ona
l
te
c
hni
que
pi
c
k
s
th
e
ne
xt
poi
nt
u
s
in
g
de
te
r
m
in
is
ti
c
c
om
put
a
ti
on.
T
he
s
e
t
e
c
hni
que
s
a
r
e
c
om
pa
r
e
d
w
it
h t
he
ot
he
r
t
r
a
di
ti
ona
l
te
c
hni
que
s
[
27]
, a
nd f
in
a
ll
y, i
t
s
how
s
t
ha
t
th
e
G
A
i
s
r
obus
t.
H
ow
e
ve
r
, s
om
e
ti
m
e
s
, i
ts
br
e
a
kdowns
be
c
a
u
s
e
of
i
nput
s
a
nd noi
s
e
pr
e
s
e
n
c
e
.
3.2.
3
.
A
n
t
c
ol
on
y op
t
im
iz
a
t
io
n
al
gor
it
h
m
T
he
m
a
in
a
s
pe
c
t
of
us
in
g
ge
ne
-
e
xpr
e
s
s
io
n
da
ta
f
or
pr
e
di
c
ti
n
g
va
r
io
us
di
s
e
a
s
e
s
a
nd
pe
r
s
ona
li
z
e
d
tr
e
a
tm
e
nt
f
a
c
il
it
ie
s
i
n pr
om
is
in
g a
r
e
a
s
l
ik
e
m
e
di
c
in
e
. T
he
r
e
f
or
e
,
di
f
f
e
r
e
nt
a
lg
or
it
hm
s
a
r
e
de
ve
lo
pe
d t
o c
la
s
s
if
y
di
f
f
e
r
e
nt
di
s
e
a
s
e
s
a
c
c
or
di
ng
to
th
e
s
e
le
c
te
d
ge
n
e
e
xpr
e
s
s
io
n,
a
nd
s
ig
ni
f
ic
a
nt
ga
in
s
a
r
e
c
a
r
r
ie
d
out
in
th
e
di
s
e
a
s
e
c
la
s
s
if
ic
a
ti
on
pr
e
c
is
e
ne
s
s
[
28]
.
M
or
e
ove
r
,
th
e
c
la
s
s
if
ic
a
ti
on
a
lg
or
it
hm
s
a
r
e
de
ve
lo
pe
d
in
va
r
io
us
s
tu
di
e
s
a
nd pe
r
f
or
m
be
tt
e
r
by uti
li
z
in
g a
s
e
le
c
te
d f
e
a
tu
r
e
s
ubs
e
t
w
it
h t
he
a
va
il
a
bl
e
da
ta
.
L
e
t
us
s
uppo
s
e
th
e
r
e
a
r
e
onl
y
two
pa
th
s
w
hi
c
h a
r
e
P
1
a
nd
P
2. C
1
a
nd C
2
a
r
e
th
e
phe
r
om
one
s
f
or
th
e
pa
th
s
P
1
a
nd
P
2,
r
e
s
pe
c
ti
ve
ly
.
L
e
t
th
e
r
e
be
a
gr
a
ph
ha
vi
ng
ve
r
t
e
x
V
a
nd
e
dge
s
E
.
F
i
r
s
tl
y,
th
e
i
th
pa
th
ha
s
th
e
c
hoos
in
g pr
oba
bi
li
ty
, gi
ve
n i
n (
2)
.
=
1
+
2
⁄
;
ℎ
=
1
,
2
(
2
)
I
f
C
1
>C
2
, t
he
n pa
th
P
1
ha
s
a
hi
ghe
r
pr
oba
bi
li
ty
of
be
in
g c
hos
e
n
th
a
n t
he
pa
th
P
2
. I
f
C
1
<C
2
, P
a
th
P
2
is
t
he
be
tt
e
r
opt
io
n.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 3, J
une
2025
:
2328
-
2337
2332
T
he
r
e
tu
r
n
pa
th
is
de
te
r
m
in
e
d
by
two
f
a
c
to
r
s
:
th
e
le
ngt
h
of
th
e
pa
th
ta
ke
n
by
a
nt
a
nd
th
e
r
a
te
of
phe
r
om
one
e
va
por
a
ti
on, a
s
di
s
c
u
s
s
e
d
a
s
f
ol
lo
w
s
:
‒
T
he
phe
r
om
one
c
onc
e
nt
r
a
ti
on va
r
ie
s
w
it
h t
he
l
e
ngt
h of
t
he
pa
th
, a
s
i
ll
us
tr
a
te
d i
n (
3)
.
=
+
(
3)
W
he
r
e
L
i
is
th
e
pa
th
'
s
le
ngt
h
a
nd
K
i
s
th
e
pa
th
'
s
le
ngt
h
-
de
pe
n
de
nt
c
ons
ta
nt
.
I
f
th
e
pa
th
is
s
hor
te
r
,
th
e
phe
r
om
one
c
onc
e
nt
r
a
ti
on w
il
l
be
i
nc
r
e
a
s
e
d.
‒
In
(
4)
de
pi
c
ts
t
he
c
ha
nge
i
n c
onc
e
nt
r
a
ti
on a
s
a
f
unc
ti
on of
t
he
r
a
te
of
e
va
por
a
ti
on.
=
(
1
−
)
∗
(
4)
H
e
r
e
, pa
r
a
m
e
te
r
v r
a
nge
s
f
r
om
0 t
o 1. I
f
v i
s
hi
ghe
r
, t
he
c
onc
e
nt
r
a
ti
on w
il
l
be
l
ow
e
r
.
3.2.
4
.
M
u
lt
i
-
ve
r
s
e
op
t
im
iz
at
io
n
al
gor
i
t
h
m
T
he
m
ul
ti
obj
e
c
ti
ve
opt
im
iz
a
ti
on
a
lg
or
it
hm
(
M
O
O
)
ha
s
be
e
n
us
e
d
f
or
va
r
io
us
opt
im
iz
a
ti
on
is
s
ue
s
w
it
h
m
ul
ti
pl
e
obj
e
c
ti
ve
s
ba
s
e
d
on
th
e
c
r
it
e
r
ia
or
goa
l
s
a
nd
th
e
s
e
w
il
l
ge
ne
r
a
ll
y e
va
lu
a
te
va
r
io
us
a
s
pe
c
t
s
of
th
e
a
c
hi
e
ve
d
s
ol
ut
io
n,
he
nc
e
in
c
om
m
e
ns
ur
a
bl
e
a
nd
pa
r
ti
a
ll
y
c
onf
li
c
ts
[
29]
.
T
he
r
e
f
or
e
,
an
unc
ons
tr
a
in
e
d
M
O
O
is
s
ue
s
a
r
e
de
te
r
m
in
e
d w
it
h t
he
m
a
th
e
m
a
ti
c
a
l
e
xpr
e
s
s
io
n, a
s
i
n (
5)
.
(
)
=
(
)
=
(
1
(
)
,
2
(
)
,
…
,
(
)
)
ℎ
=
(
1
,
2
,
…
,
)
∈
(
5
)
W
he
r
e
x
n
is
di
m
e
ns
io
na
l
de
c
is
io
n s
ol
ut
io
n i
n ve
c
to
r
, X
is
d
e
c
is
i
on s
pa
c
e
,
a
nd
f
(
x)
is
o
bj
e
c
ti
ve
f
unc
ti
on.
3.2.
5
.
5
-
f
ol
d
c
r
os
s
val
id
at
io
n
al
gor
it
h
m
C
r
os
s
-
va
li
da
ti
on
is
c
ons
id
e
r
e
d
a
s
ta
ti
s
ti
c
a
l
t
e
c
hni
que
th
a
t
e
v
a
lu
a
te
s
th
e
s
ki
ll
of
M
L
te
c
hni
que
s
.
N
or
m
a
ll
y,
th
e
s
e
te
c
hni
que
s
a
r
e
e
m
pl
oye
d
in
M
L
to
c
om
pa
r
e
a
nd
pi
c
k
a
n
a
ppr
opr
ia
te
a
nd
e
f
f
e
c
ti
ve
m
ode
l
f
o
r
a
pr
ovi
de
d
pr
e
di
c
ti
ve
m
ode
ll
in
g
is
s
ue
s
due
to
th
e
s
im
pl
e
im
pl
e
m
e
nt
a
ti
on,
e
a
s
y
unde
r
s
t
a
ndi
ng,
a
nd
lo
w
e
r
bi
a
s
e
s
ti
m
a
ti
on
c
om
pa
r
e
d
to
ot
he
r
te
c
hni
que
s
.
F
ur
th
e
r
,
th
e
k
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on
is
a
pr
oc
e
s
s
ut
il
iz
e
d
to
e
va
lu
a
te
th
e
m
ode
l
on
ne
w
da
ta
.
T
he
r
e
f
or
e
,
c
om
m
on
m
e
th
ods
a
nd
t
r
ic
ks
a
r
e
ut
il
iz
e
d
in
us
in
g
a
nd
s
e
le
c
ti
ng
k
-
va
lu
e
s
f
or
t
he
da
ta
s
e
t.
L
e
t
K
:
{
1,
…,
N
}
{
1,
…,
K
}
b
e
a
n
in
de
xi
ng
f
unc
ti
on
th
a
t
s
pe
c
if
ie
s
th
e
di
vi
s
io
n
to
w
hi
c
h
r
e
por
t
I
is
a
s
s
ig
ne
d vi
a
r
a
ndomi
z
a
ti
on. L
e
t
F
(
x)
be
t
he
f
it
te
d f
unc
ti
on ob
ta
in
e
d a
f
te
r
r
e
m
ovi
ng t
he
K
th
pa
r
t
o
f
t
he
da
ta
.
T
he
C
V
e
s
ti
m
a
ti
on of
t
he
e
r
r
or
i
n pr
e
di
c
ti
on i
s
pr
ovi
de
d i
n (
6)
.
(
)
=
1
∑
(
,
(
)
1
(
6
)
H
e
r
e
, t
he
c
hoi
c
e
of
K
i
s
5.
3.
3
.
I
m
p
le
m
e
n
t
at
io
n
of
t
h
e
p
r
op
os
e
d
ap
p
r
oac
h
T
hi
s
s
e
c
ti
on
in
c
lu
de
s
a
de
ta
il
e
d
d
e
s
c
r
ip
ti
on
of
how
th
e
pr
op
os
e
d
opt
im
iz
a
ti
on
m
e
th
odol
ogy
w
a
s
im
pl
e
m
e
nt
e
d,
in
c
lu
di
ng
th
e
a
ppr
oa
c
he
s
,
to
ol
s
,
a
nd
te
c
hni
qu
e
s
us
e
d.
F
ig
ur
e
1
de
pi
c
ts
th
e
im
pl
e
m
e
nt
a
ti
on
m
e
th
od
in
s
te
ps
,
b
e
gi
nni
ng
w
it
h
da
ta
c
ol
le
c
ti
on
a
nd
pr
e
pr
oc
e
s
s
i
ng
to
e
ns
ur
e
hi
gh
-
qua
li
ty
in
put
f
or
th
e
m
ode
l.
T
he
s
ys
t
e
m
is
th
e
n
tr
a
in
e
d
a
nd
te
s
te
d
us
in
g
a
ppr
opr
ia
te
ML
te
c
hni
que
s
,
f
ol
lo
w
e
d
by
th
e
us
e
of
opt
im
iz
a
ti
on
a
lg
or
it
hm
s
s
uc
h
a
s
W
O
A
,
M
V
O
,
G
A
,
a
nd
A
C
O
,
a
s
w
e
ll
a
s
a
5
-
f
ol
d
C
V
te
c
hni
que
to
im
pr
ove
m
ode
l
pe
r
f
or
m
a
nc
e
.
F
in
a
ll
y,
th
e
be
s
t
-
opt
im
iz
e
d
r
e
s
ul
ts
a
r
e
id
e
nt
if
ie
d
by
a
s
s
e
s
s
in
g
th
e
m
ode
ls
a
ga
in
s
t
im
por
ta
nt
pe
r
f
or
m
a
nc
e
c
r
it
e
r
ia
, e
ns
ur
in
g t
ha
t
th
e
s
ugge
s
te
d
a
ppr
oa
c
h i
s
us
e
f
ul
i
n bi
oi
nf
or
m
a
ti
c
s
a
ppl
ic
a
ti
ons
.
3.
3
.1.
D
at
a c
ol
le
c
t
io
n
I
ni
ti
a
ll
y,
th
e
c
a
nc
e
r
ous
pr
ot
e
in
in
te
r
a
c
ti
on
da
ta
is
c
ol
le
c
te
d
in
th
e
c
ur
r
e
nt
r
e
s
e
a
r
c
h.
A
s
a
r
e
s
ul
t,
th
e
e
xpe
r
im
e
nt
a
l
pr
oc
e
dur
e
ta
ke
s
in
to
a
c
c
ount
th
e
da
ta
in
te
r
a
c
ti
ons
th
a
t
oc
c
ur
w
it
h
c
a
nc
e
r
pr
ot
e
in
s
.
T
he
s
e
in
te
r
a
c
ti
ons
a
r
e
r
e
f
e
r
r
e
d
to
be
m
a
li
gna
nt
P
P
I
s
.
F
ur
th
e
r
m
or
e
,
th
e
te
r
m
“
nonc
a
nc
e
r
ou
s
pr
ot
e
in
in
te
r
a
c
ti
on
”
in
th
e
pa
pe
r
r
e
f
e
r
s
to
e
it
he
r
one
or
bot
h
pr
ot
e
in
s
th
a
t
ha
ve
ye
t
to
be
r
e
c
ogni
z
e
d
in
r
e
la
ti
on
to
c
a
nc
e
r
.
H
e
r
e
w
e
pr
oc
e
s
s
e
d
w
it
h
th
e
“
hum
a
n
pr
ot
e
in
da
ta
”
da
ta
s
e
t
dow
nl
oa
de
d
f
r
om
ope
n
a
c
c
e
s
s
r
e
pos
it
or
y
F
ig
s
ha
r
e
.
T
he
pos
it
iv
e
a
s
w
e
ll
a
s
ne
ga
ti
ve
obs
e
r
va
ti
ons
w
e
r
e
t
he
n s
e
pa
r
a
te
d i
n
to
s
e
ts
f
or
bot
h
tr
a
in
in
g a
nd
te
s
ti
ng, with 80
%
of
th
e
da
ta
de
s
ig
na
te
d
f
or
tr
a
in
in
g
a
nd
th
e
r
e
m
a
in
in
g
f
or
te
s
ti
ng.
F
ur
th
e
r
,
th
e
pa
r
a
m
e
te
r
s
in
c
lu
de
d
in
th
e
hum
a
n
pr
ot
e
in
da
ta
s
e
t
s
uc
h
a
s
N
_pr
ot
e
in
_a
,
N
_pr
ot
e
in
_b,
P
_pr
ot
e
in
_a
,
a
nd
P
_pr
ot
e
in
_b,
a
nd
th
e
na
m
e
,
a
tt
r
ib
ut
e
s
, a
tt
r
ib
ut
e
t
ype
, a
nd t
he
t
ot
a
l
num
be
r
of
da
ta
ha
ve
be
e
n
l
is
te
d i
n
T
a
bl
e
1.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
O
pt
imi
z
in
g bi
oi
nf
or
m
at
ic
s
appli
c
at
io
ns
:
a nov
e
l
app
r
oac
h w
it
h
hum
an pr
ot
e
in
dat
a
…
(
P
r
e
e
ti
T
har
e
ja
)
2333
F
ig
ur
e
1. T
he
i
m
pl
e
m
e
nt
a
ti
on f
lo
w
of
t
he
pr
opos
e
d s
ys
te
m
T
a
bl
e
1
. H
um
a
n
pr
ot
e
in
da
ta
s
e
t
A
t
t
r
i
but
e
s
A
t
t
r
i
but
e
t
ype
T
ot
a
l
no. of
da
t
a
N
e
ga
t
i
ve
pr
ot
e
i
ns
P
r
ot
e
i
n ‘
a
’
4262
P
r
ot
e
i
n ‘
b’
4262
P
os
i
t
i
ve
pr
ot
e
i
ns
P
r
ot
e
i
n ‘
a
’
3899
P
r
ot
e
i
n ‘
b’
3899
O
r
i
gi
na
l
s
e
t
16,322
T
r
a
i
n s
e
t
13,058
T
e
s
t
s
e
t
3264
3.3.2.
D
at
a p
r
e
p
r
oc
e
s
s
in
g
I
n
th
is
s
ta
ge
,
w
e
a
dopt
a
pr
e
pr
oc
e
s
s
in
g
s
tr
a
te
gy
th
a
t
e
m
pl
oys
two
di
s
ti
nc
t
m
e
a
s
ur
e
s
to
d
e
te
c
t
a
nd
de
le
te
in
te
r
a
c
ti
ons
th
a
t
a
r
e
ve
r
y
li
ke
ly
to
be
f
a
k
e
.
L
a
r
ge
da
t
a
s
e
ts
a
r
e
b
e
c
om
in
g
m
or
e
pr
e
va
l
e
nt
,
a
nd
th
e
y
a
r
e
of
te
n
c
ha
ll
e
ngi
ng
to
c
om
pr
e
he
nd.
P
r
in
c
ip
a
l
c
om
pone
nt
a
na
l
ys
is
(
P
C
A
)
is
a
n
a
ppr
oa
c
h
f
or
lo
w
e
r
in
g
th
e
di
m
e
ns
io
n
of
s
uc
h
da
ta
s
e
ts
,
im
pr
ovi
ng
in
te
r
pr
e
ta
bi
li
ty
w
hi
le
m
in
im
iz
in
g
da
ta
lo
s
s
.
I
t
a
c
c
om
pl
i
s
he
s
th
i
s
by
ge
ne
r
a
ti
ng
ne
w
in
de
p
e
nde
nt
va
r
ia
bl
e
s
th
a
t
gr
a
dua
ll
y
m
a
xi
m
iz
e
va
r
ia
nc
e
.
F
in
di
ng
s
u
c
h
ne
w
va
r
ia
bl
e
s
,
known
a
s
pr
in
c
ip
a
l
c
om
pon
e
nt
s
,
i
s
e
qui
va
le
nt
to
s
ol
ve
e
ig
e
nva
lu
e
/e
ig
e
nve
c
to
r
is
s
u
e
,
a
nd
th
e
ne
w
va
r
ia
bl
e
s
a
r
e
s
pe
c
if
ie
d
by
th
e
da
ta
s
e
t
a
t
ha
nd
r
a
th
e
r
th
a
n
a
p
r
io
r
i,
m
a
ki
n
g
P
C
A
a
n
a
da
pt
a
bl
e
da
ta
a
na
ly
s
is
te
c
hni
que
.
W
e
ig
ht
e
d
K
-
m
e
a
n
s
a
nd
G
a
us
s
ia
n
m
ix
tu
r
e
m
od
e
ls
(
G
M
M
)
w
i
th
e
xpe
c
ta
ti
on
-
m
a
xi
m
iz
a
ti
on
(
E
M
)
ha
ve
be
e
n
ut
il
iz
e
d
in
th
e
r
e
s
e
a
r
c
h.
M
or
e
ove
r
,
he
r
e
w
e
do
th
e
da
ta
n
or
m
a
li
z
a
ti
on
on
th
e
in
put
da
ta
.
A
s
a
r
e
s
ul
t,
nor
m
a
li
z
a
ti
on
is
f
r
e
que
nt
ly
us
e
d
to
pr
e
pa
r
e
da
ta
f
or
ML
.
N
or
m
a
li
z
a
ti
on
s
e
e
ks
to
tr
a
ns
f
or
m
th
e
num
e
r
ic
a
l
va
lu
e
s
of
num
e
r
ic
a
l
c
ol
um
ns
in
a
da
ta
s
e
t
to
a
s
im
il
a
r
s
c
a
le
w
it
hout
di
s
to
r
ti
ng
or
lo
s
in
g
in
f
or
m
a
ti
on.
T
he
m
a
th
e
m
a
ti
c
a
l
f
or
m
ul
a
ti
on of
W
e
ig
ht
e
d K
-
m
e
a
ns
a
nd
G
M
M
ha
s
be
e
n de
f
in
e
d i
n (
7)
-
(
11)
.
W
e
ig
ht
e
d
K
-
m
e
a
ns
m
ode
l
:
(
)
=
{
−
1
|
|
−
|
|
2
}
∑
{
−
1
|
|
−
|
|
2
}
(
7)
f
or
k=
1,
…
,
K
, a
nd
(
)
=>0
and
∑
(
)
=
1
.
>
0
=
1
=
∑
(
)
∑
(
)
(
8)
x1, x2, …, xn
is
da
ta
, w
he
r
e
x ϵ
R
d,
is
w
e
ig
ht
e
d a
ve
r
a
ge
.
G
M
M
:
(
,
)
=
1
(
2
)
2
√
|
|
(
−
1
2
(
−
)
−
1
(
−
)
)
(
9)
W
he
r
e
is
m
e
a
n
, Σ
is
G
a
us
s
ia
n
c
ova
r
ia
nc
e
m
a
tr
ix
, d
is
n
um
be
r
of
f
e
a
tu
r
e
s
da
ta
s
e
t,
x
is
n
um
be
r
of
da
ta
poi
nt
s
.
G
M
M
w
it
h
EM
:
‒
E
-
s
te
p:
(
)
=
(
|
,
)
∑
(
|
,
)
, f
or
k=
1,
…,
k
(
10)
‒
M
-
s
te
p:
=
,
=
1
∑
(
)
∑
=
1
∑
(
)
(
−
=
1
=
1
)
(
−
)
(
11)
f
or
k=
1,
…,
k, n
k
=
∑
(
)
=
1
, t
he
va
lu
e
s
w
il
l
be
upda
t
e
d.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 3, J
une
2025
:
2328
-
2337
2334
3.
3
.
3
.
F
e
at
u
r
e
e
xt
r
ac
t
io
n
I
n
th
is
pha
s
e
,
th
e
f
e
a
tu
r
e
s
c
a
n
be
r
e
tr
ie
ve
d
us
in
g
th
e
noi
s
e
f
il
te
r
.
N
oi
s
e
da
ta
is
de
f
in
e
d
a
s
th
e
pr
e
s
e
nc
e
of
ove
r
s
ig
ht
s
,
dupl
ic
a
te
in
f
or
m
a
ti
on,
o
r
a
nom
a
lo
us
da
ta
in
r
e
t
r
ie
ve
d
da
ta
.
F
e
a
tu
r
e
s
e
le
c
ti
on
is
a
c
c
om
pl
is
he
d
th
r
ough
m
a
nua
l
e
va
lu
a
ti
on
of
pr
ot
e
in
da
ta
.
A
s
a
r
e
s
ul
t,
f
e
a
tu
r
e
da
ta
a
r
e
pi
c
ke
d
by
m
a
nua
ll
y
in
s
pe
c
ti
ng pr
ot
e
in
da
ta
, a
nd r
e
le
va
nt
da
ta
i
s
obt
a
in
e
d, r
e
f
or
m
a
tt
e
d, a
nd s
a
ve
d i
n
a
s
tr
uc
tu
r
e
d da
ta
ba
s
e
.
3.
3
.
4
.
T
r
ai
n
in
g an
d
t
e
s
t
in
g of
d
at
a
T
he
nor
m
a
li
z
e
d
da
ta
is
s
ys
te
m
a
ti
c
a
ll
y
di
vi
de
d
in
to
tr
a
in
in
g
a
nd
te
s
ti
ng
s
e
ts
to
f
a
c
il
it
a
te
f
ur
th
e
r
pr
oc
e
s
s
in
g
in
th
e
e
xpe
r
im
e
nt
a
ti
on
pha
s
e
.
I
n
th
e
c
ur
r
e
nt
s
c
e
na
r
io
,
80%
of
th
e
da
ta
is
a
ll
oc
a
te
d
f
or
tr
a
in
in
g,
e
ns
ur
in
g
th
a
t
th
e
m
od
e
l
le
a
r
ns
e
f
f
e
c
ti
ve
ly
f
r
om
a
s
ub
s
ta
nt
ia
l
por
ti
on
of
th
e
da
ta
s
e
t.
T
he
r
e
m
a
in
in
g
20%
is
r
e
s
e
r
ve
d
f
or
te
s
ti
ng,
a
ll
ow
in
g
f
or
a
n
unbi
a
s
e
d
e
va
lu
a
ti
on
of
th
e
m
ode
l’
s
pe
r
f
or
m
a
nc
e
a
nd
it
s
a
bi
li
ty
to
ge
ne
r
a
li
z
e
t
o uns
e
e
n da
t
a
. T
hi
s
s
pl
it
i
s
c
a
r
e
f
ul
ly
c
hos
e
n t
o m
a
in
ta
in
a
ba
la
nc
e
be
twe
e
n l
e
a
r
ni
ng e
f
f
ic
ie
nc
y a
nd
a
c
c
ur
a
te
a
s
s
e
s
s
m
e
nt
, e
n
s
ur
in
g t
he
r
obus
tn
e
s
s
of
t
he
pr
opos
e
d o
pt
im
iz
a
ti
on a
ppr
oa
c
h.
3.
3
.
5
.
O
p
t
im
iz
at
io
n
u
s
in
g t
h
e
p
r
op
os
e
d
m
od
e
l
I
n
th
is
s
te
p,
th
e
v
a
r
io
us
opt
im
iz
a
ti
on
te
c
hni
que
s
ha
ve
b
e
e
n
ut
i
li
z
e
d
a
nd
th
e
s
e
m
ode
ls
a
r
e
e
s
ti
m
a
t
e
d
by
in
de
pe
nde
nt
te
s
ti
ng
a
nd
c
r
os
s
-
va
li
da
ti
on
pr
oc
e
s
s
.
F
ur
th
e
r
,
th
e
pr
opos
e
d
opt
im
iz
a
ti
on
pr
e
di
c
ti
on
m
ode
l
is
a
tt
e
m
pt
in
g
to
a
na
ly
z
e
a
nd
f
in
d
w
it
h
th
e
s
e
le
c
te
d
f
e
a
tu
r
e
s
th
e
n
in
te
gr
a
te
d
w
it
h
th
e
c
e
r
ta
in
s
p
e
c
if
ie
d
f
e
a
tu
r
e
s
a
s
a
ne
w
f
e
a
tu
r
e
of
a
nove
l
pr
ot
ot
ype
f
or
f
ur
th
e
r
opt
im
iz
a
ti
on.
T
he
r
e
f
or
e
,
onc
e
th
e
s
e
c
onda
r
y
f
e
a
tu
r
e
s
e
le
c
ti
on
got
ove
r
,
th
e
ne
w
opt
im
a
l
m
ode
l
w
a
s
e
va
lu
a
te
d
vi
a
c
r
o
s
s
-
va
li
d
a
ti
on
a
nd
in
de
pe
nde
nt
te
s
ti
ng.
E
ve
nt
ua
ll
y,
th
e
a
s
s
oc
ia
ti
on
a
m
id
s
t
th
e
be
s
t
-
pr
e
di
c
te
d
pr
ot
e
in
s
ut
il
iz
in
g
a
n
e
f
f
ic
ie
nt
de
e
p
le
a
r
ni
ng
te
c
hni
que
s
w
il
l
be
ut
il
iz
e
d
to
f
in
d
nove
l
th
e
r
a
pe
ut
ic
ta
r
ge
ts
.
F
u
r
th
e
r
,
pow
e
r
f
u
l
m
ode
ls
pr
e
di
c
t
s
e
ve
r
a
l
dr
ug
-
a
bl
e
350
pr
ot
e
in
s
th
a
t
s
houl
d
be
de
e
pl
y
us
e
d
to
f
in
d
be
tt
e
r
th
e
r
a
pe
ut
ic
t
a
r
ge
ts
.
I
n
th
e
c
ur
r
e
nt
r
e
s
e
a
r
c
h,
w
e
ha
v
e
e
m
pl
oye
d
va
r
io
us
opt
im
iz
a
ti
on
m
e
th
ods
,
in
c
lu
di
ng
W
O
A
,
M
V
O
,
G
A
,
A
C
O
,
a
n
d
5
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on
,
to
a
c
hi
e
ve
be
tt
e
r
opt
im
um
r
e
s
ul
ts
a
nd t
he
va
r
io
us
t
ype
s
of
opt
im
iz
a
ti
on
te
c
hni
que
s
t
o
a
na
ly
z
e
th
e
a
c
c
ur
a
c
y a
nd pe
r
f
or
m
a
nc
e
of
th
e
pr
opos
e
d
s
ys
te
m
.
T
h
e
5
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on
pr
oc
ur
e
d
e
f
f
ic
ie
nt
r
e
s
ul
ts
c
om
pa
r
e
d
to
th
e
ot
he
r
opt
im
iz
a
ti
on t
e
c
hni
que
s
.
4.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
4
.1.
P
e
r
f
or
m
an
c
e
e
val
u
at
io
n
T
he
s
im
ul
a
ti
on
r
e
s
ul
ts
of
th
e
pr
opos
e
d
te
c
hni
que
s
,
w
hi
c
h
in
te
gr
a
te
of
5
-
da
ta
m
in
in
g
opt
im
iz
a
ti
on
a
lg
or
it
hm
s
-
W
O
A
,
M
V
O
,
G
A
,
A
C
O
,
a
nd
5
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on
-
ha
ve
be
e
n
e
va
lu
a
te
d
us
in
g
ke
y
p
e
r
f
or
m
a
nc
e
m
e
tr
ic
s
in
c
lu
di
ng
a
c
c
ur
a
c
y,
s
e
n
s
it
iv
it
y,
s
pe
c
if
ic
it
y,
pr
e
c
is
io
n,
a
nd
F
-
s
c
or
e
.
T
he
a
c
c
ur
a
c
y
of
th
e
s
e
opt
im
iz
a
ti
on
te
c
hni
que
s
w
a
s
a
s
s
e
s
s
e
d
in
pr
e
di
c
ti
ng
va
r
io
u
s
di
s
e
a
s
e
s
.
C
om
pa
r
a
ti
ve
a
na
ly
s
i
s
ba
s
e
d
on
pe
r
f
or
m
a
nc
e
pa
r
a
m
e
te
r
s
s
how
s
th
a
t
5
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on
a
c
hi
e
ve
d
be
tt
e
r
r
e
s
ul
ts
th
a
n
th
e
ot
he
r
te
c
hni
que
s
.
F
ur
th
e
r
m
or
e
,
th
e
e
va
lu
a
ti
on
of
5
-
f
ol
d
C
V
on
va
r
io
us
m
e
tr
ic
s
-
s
uc
h
a
s
a
c
c
ur
a
c
y,
s
p
e
c
if
ic
it
y,
s
e
n
s
it
iv
it
y,
pr
e
c
is
io
n, F
-
s
c
or
e
, e
r
r
or
r
a
te
, R
O
C
, a
nd F
P
R
-
de
m
ons
tr
a
te
d i
ts
s
upe
r
io
r
e
f
f
e
c
ti
ve
ne
s
s
.
T
he
pe
r
f
or
m
a
nc
e
of
th
e
pr
opos
e
d
opt
im
iz
a
ti
on
a
ppr
oa
c
h
w
a
s
e
va
lu
a
te
d
us
in
g
m
ul
ti
pl
e
te
c
hni
que
s
,
in
c
lu
di
ng
A
C
O
,
G
A
,
M
V
O
,
W
O
A
,
a
nd
5
-
f
ol
d
C
V
.
T
he
a
c
c
ur
a
c
y
a
c
hi
e
ve
d
w
it
h
A
C
O
,
G
A
,
M
V
O
,
a
nd
W
O
A
w
a
s
0.7841,
0.5966,
0.5455,
a
nd
0.8182,
r
e
s
pe
c
ti
ve
ly
,
w
he
r
e
a
s
5
-
f
ol
d
C
V
yi
e
ld
e
d
th
e
hi
ghe
s
t
a
c
c
ur
a
c
y
of
0.9861
(
98.61%
)
.
A
ddi
ti
ona
ll
y,
th
e
e
va
lu
a
ti
on
m
e
tr
ic
s
f
or
5
-
f
ol
d
C
V
de
m
ons
tr
a
te
d
s
upe
r
io
r
pe
r
f
or
m
a
nc
e
,
w
it
h
a
n
a
c
c
ur
a
c
y
of
98.61%
,
s
e
ns
it
iv
it
y
of
88.64%
,
s
p
e
c
if
ic
it
y
of
96.59%
,
pr
e
c
is
io
n
of
99.30%
,
F
-
s
c
or
e
of
92.31%
,
e
r
r
or
r
a
te
of
10.80%
,
R
O
C
of
92.61%
,
a
nd
a
n
F
P
R
of
3.00%
.
T
h
e
s
e
r
e
s
ul
ts
hi
ghl
ig
ht
th
e
e
f
f
e
c
ti
ve
ne
s
s
of
th
e
pr
opos
e
d
opt
im
iz
a
ti
on
m
ode
l
in
im
pr
ovi
ng
pr
e
di
c
ti
ve
pe
r
f
or
m
a
nc
e
in
bi
oi
nf
or
m
a
ti
c
s
a
ppl
ic
a
ti
ons
.
4
.2.
P
e
r
f
or
m
an
c
e
c
o
m
p
ar
is
on
A
c
om
pa
r
a
ti
ve
a
na
ly
s
is
by
c
om
p
a
r
in
g
th
e
pe
r
f
or
m
a
nc
e
of
th
e
pr
opos
e
d
a
ppr
oa
c
h
w
it
h
th
e
a
ppr
oa
c
h
pr
e
s
e
nt
e
d
in
th
e
e
xi
s
ti
ng
w
or
ks
ha
s
be
e
n
de
pi
c
te
d
in
th
e
pl
ot
di
a
gr
a
m
.
I
n
th
e
c
ur
r
e
nt
pr
opos
e
d
r
e
s
e
a
r
c
h,
th
e
di
f
f
e
r
e
nt
opt
im
iz
a
ti
on
te
c
hni
que
s
in
c
lu
di
ng
5
-
da
ta
m
in
in
g
opt
i
m
iz
a
ti
on
a
lg
or
it
hm
s
in
c
lu
de
th
e
W
O
A
,
M
V
O
,
G
A
,
A
C
O
,
a
nd
5
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on
ha
ve
e
xpe
r
im
e
nt
e
d
w
it
h
c
e
r
ta
in
pe
r
f
o
r
m
a
nc
e
e
va
lu
a
ti
on
m
e
tr
ic
s
in
c
lu
di
ng
a
c
c
ur
a
c
y,
s
e
n
s
it
iv
it
y,
s
pe
c
if
ic
it
y,
pr
e
c
is
io
n
,
a
nd
F
-
s
c
or
e
.
E
ve
nt
ua
ll
y,
F
ig
ur
e
2
de
pi
c
ts
th
a
t
5
-
f
ol
d
c
r
os
s
va
li
da
ti
on
te
c
hni
que
s
pr
oc
ur
e
d
be
tt
e
r
r
e
s
ul
ts
w
it
h
0.98
a
c
c
ur
a
c
y
c
om
pa
r
e
d
to
th
e
ot
he
r
op
ti
m
iz
a
ti
on
te
c
hni
que
s
a
nd de
not
e
d a
s
a
b
e
s
t
m
ode
l
in
di
a
gno
s
e
of
di
s
e
a
s
e
s
.
V
a
r
io
us
opt
im
iz
a
ti
on
te
c
hni
que
s
h
a
ve
be
e
n
e
xe
c
ut
e
d
to
e
va
lu
a
te
a
c
c
ur
a
c
y
in
th
e
pr
e
di
c
ti
on
of
di
s
e
a
s
e
s
.
T
h
e
s
im
ul
a
ti
on
r
e
s
ul
ts
on
di
f
f
e
r
e
nt
pe
r
f
or
m
a
nc
e
e
va
lu
a
ti
on
m
e
tr
ic
s
of
W
O
A
,
M
V
O
,
G
A
,
A
C
O
,
a
nd
5
-
f
ol
d
ha
ve
be
e
n
m
e
a
s
ur
e
d
a
nd
a
c
hi
e
ve
d
r
e
s
ul
ts
on
a
c
c
ur
a
c
y
w
it
h
0.8,
0.51,
0.6,
0.68,
a
nd
0.98.
F
in
a
ll
y,
c
om
pa
r
e
d
to
ot
he
r
opt
im
iz
a
ti
on
te
c
hni
que
s
,
5
-
f
ol
d
ha
s
be
e
n
pr
oc
ur
e
d
w
it
h
hi
ghe
r
r
e
s
ul
ts
of
0.98
on
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
O
pt
imi
z
in
g bi
oi
nf
or
m
at
ic
s
appli
c
at
io
ns
:
a nov
e
l
app
r
oac
h w
it
h
hum
an pr
ot
e
in
dat
a
…
(
P
r
e
e
ti
T
har
e
ja
)
2335
a
c
c
ur
a
c
y,
w
hi
c
h
in
di
c
a
te
s
th
e
be
tt
e
r
pr
e
di
c
ti
on
of
di
s
e
a
s
e
s
th
a
t
he
lp
s
th
e
m
e
di
c
a
l
e
xp
e
r
ts
to
id
e
nt
if
y
w
it
h
be
tt
e
r
pr
e
c
io
us
ne
s
s
a
nd ge
ne
r
a
te
t
he
po
s
s
ib
il
it
y t
o s
ta
r
t
th
e
t
r
e
a
t
m
e
nt
s
oon.
F
ig
ur
e
2.
R
e
s
ul
ts
of
t
he
pr
opos
e
d r
e
s
e
a
r
c
h c
om
pa
r
e
d w
it
h ot
he
r
opt
im
iz
a
ti
on t
e
c
hni
que
s
4
.
3
.
C
om
p
ar
is
on
w
it
h
e
xi
s
t
in
g p
r
e
d
ic
t
io
n
m
od
e
ls
T
a
bl
e
2
c
om
pa
r
e
s
th
e
pe
r
f
or
m
a
nc
e
of
our
s
ugge
s
te
d
de
e
p
le
a
r
ni
ng
opt
im
iz
e
d
m
ode
l
a
nd
S
O
T
A
m
ode
ls
s
uc
h
a
s
G
T
B
-
P
P
I
,
D
e
e
pP
P
I
,
M
I
M
I
-
N
M
B
A
C
-
R
F
,
S
pa
ti
a
l
P
P
I
,
F
F
A
N
E
,
G
e
nom
e
N
e
t
A
r
c
hi
te
c
t,
xC
A
P
T
5,
a
nd C
N
N
-
B
iL
G
.
T
hi
s
c
om
pl
e
te
e
xa
m
in
a
ti
on
of
c
la
s
s
i
f
ic
a
ti
on
pe
r
f
or
m
a
nc
e
in
c
lu
de
s
m
e
tr
ic
s
s
uc
h
a
s
s
e
ns
it
iv
it
y,
s
pe
c
if
ic
it
y,
M
a
tt
he
w
s
c
or
r
e
la
ti
on
c
oe
f
f
ic
ie
nt
(
M
C
C
)
,
a
nd
a
c
c
ur
a
c
y,
a
nd
th
e
r
e
s
ul
ts
ha
ve
be
e
n
va
li
da
te
d
us
in
g
pr
e
vi
ous
s
tu
di
e
s
'
f
in
di
ngs
.
O
ur
pr
opos
e
d
m
o
de
l
a
tt
a
in
s
a
n
e
xc
e
ll
e
nt
pr
e
c
is
io
n
of
99.29%
,
w
hi
c
h
gr
e
a
tl
y
e
xc
e
ls
w
h
e
n
c
om
pa
r
e
d
to
G
T
B
-
P
P
I
(
89.99%
)
,
D
e
e
pP
P
I
(
84.32%
)
,
F
F
A
N
E
(
98.48%
)
,
a
nd
xC
A
P
T
5
(
99.1%
)
.
T
o
be
gi
n,
pr
e
c
i
s
io
n
r
e
pr
e
s
e
nt
s
th
e
m
in
im
um
num
be
r
of
f
a
ls
e
pos
it
iv
e
s
.
T
hi
s
de
m
ons
tr
a
te
s
how
e
f
f
e
c
ti
ve
ly
our
de
e
p
le
a
r
ni
ng
a
lg
or
it
hm
c
a
n
lo
c
a
te
s
ui
t
a
bl
e
in
s
ta
nc
e
s
of
D
B
P
s
.
T
he
n,
s
pe
c
if
ic
it
y
is
c
a
lc
ul
a
te
d
a
s
th
e
pr
opor
ti
on
of
ge
nui
ne
ne
ga
ti
ve
f
or
e
c
a
s
ts
a
m
ong
a
ll
tr
ue
ne
ga
ti
ve
in
s
ta
nc
e
s
.
T
h
e
m
ode
l
w
e
pr
opos
e
e
xc
e
ls
w
it
h
a
s
pe
c
if
ic
it
y
of
96.59%
in
c
ont
r
a
s
t
to
C
N
N
-
B
iL
G
(
94.14%
)
,
S
pa
ti
a
l
P
P
I
(
79.6
%
)
,
G
T
B
-
P
P
I
(
91.15%
)
,
D
e
e
pP
P
I
(
89.44
%
)
,
a
nd
M
I
M
I
-
N
M
B
A
C
-
R
F
(
86.81%
)
.
T
hi
s
de
m
ons
tr
a
te
s
how
r
e
li
a
bl
e
our
a
ppr
oa
c
h i
s
a
t
id
e
nt
if
yi
ng ne
ga
ti
ve
c
a
s
e
s
,
w
hi
c
h i
m
pr
ove
s
t
he
g
e
ne
r
a
l
va
li
di
ty
of
t
he
c
la
s
s
if
ic
a
ti
on r
e
s
ul
ts
.
T
a
bl
e
2
.
C
om
pa
r
is
on
w
it
h ot
he
r
S
O
T
A
pr
e
di
c
to
r
s
S
O
T
A
m
e
t
hod
D
a
t
a
s
e
t
u
s
e
d
A
c
c
ur
a
c
y
R
e
f
e
r
e
nc
e
S
pa
t
i
a
l
P
P
I
M
a
m
m
a
l
i
a
n non i
nt
e
r
a
c
t
i
ng pr
ot
e
i
ns
pa
i
r
s
0.83
[
14]
F
F
A
N
E
H
om
o
s
a
pi
e
ns
0.97
[
15]
G
e
nom
e
N
e
t
a
r
c
hi
t
e
c
t
B
a
c
t
e
r
i
a
l
a
nd
vi
r
a
l
ge
nom
e
s
0.83
[
16]
xC
A
P
T
5
H
um
a
n
pa
n da
t
a
s
e
t
0.97
[
17]
C
N
N
-
B
i
L
G
a
r
c
hi
t
e
c
t
ur
e
D
N
A
bi
ndi
ng pr
ot
e
i
ns
:
a
r
a
bi
dops
i
s
a
nd ye
a
s
t
0.94
[
18]
G
T
B
-
PPI
H
om
o
s
a
pi
e
ns
0.95
[
19]
D
e
e
pP
P
I
H
om
o
s
a
pi
e
ns
0.93
[
30]
M
I
M
I
+N
M
B
A
C
+R
F
H
om
o
s
a
pi
e
ns
0.94
[
31]
P
r
opos
e
d
H
om
o
s
a
pi
e
ns
0.98
C
ur
r
e
nt
s
t
udy
5.
C
O
N
C
L
U
S
I
O
N
A
N
D
F
U
T
U
R
E
WORK
B
io
m
e
di
c
a
ls
ha
s
b
e
c
om
e
a
s
ig
ni
f
ic
a
nt
in
dus
tr
y
in
th
e
m
e
di
c
a
l
pl
a
tf
or
m
.
T
he
m
a
in
f
oc
us
of
th
e
r
e
s
e
a
r
c
h
is
to
de
ve
lo
p
e
f
f
e
c
ti
ve
opt
im
iz
a
ti
on
te
c
hni
que
s
to
r
e
s
ol
ve
m
os
t
of
th
e
m
os
t
c
om
pl
e
x
is
s
ue
s
.
N
e
ve
r
th
e
le
s
s
,
pr
e
di
c
ti
ng
th
e
e
f
f
ic
ie
nt
a
nd
a
ppr
opr
ia
te
m
e
th
od
i
s
c
ha
ll
e
ngi
ng,
a
nd
a
t
th
e
s
a
m
e
ti
m
e
,
s
to
r
in
g
a
huge
a
m
ount
of
m
e
di
c
a
l
da
ta
on
a
pl
a
tf
or
m
or
de
vi
c
e
i
s
c
om
pl
e
x
in
m
os
t
s
c
e
n
a
r
io
s
.
T
he
r
e
f
or
e
,
to
s
ol
ve
th
is
ki
nd
of
c
ha
ll
e
nge
,
r
e
s
e
a
r
c
he
r
s
ut
il
iz
e
d
v
a
r
io
us
opt
im
iz
a
ti
on
te
c
hni
que
s
s
uc
h
a
s
c
r
os
s
-
va
li
da
ti
on
in
bi
oi
nf
or
m
a
ti
c
s
a
ppl
ic
a
ti
ons
th
a
t
ha
ve
be
e
n
pr
oc
e
s
s
e
d
w
it
h
th
e
hum
a
n
pr
ot
e
in
da
t
a
.
T
hi
s
d
a
ta
s
e
t
ha
s
be
e
n
e
xe
c
ut
e
d
in
v
a
r
io
us
s
t
e
ps
,
in
c
lu
di
ng
da
ta
c
ol
le
c
ti
on,
da
ta
nor
m
a
li
z
a
ti
on,
tr
a
in
in
g
a
nd
te
s
ti
ng,
a
nd
e
m
pl
oyi
ng
va
r
io
us
opt
im
iz
a
ti
on
te
c
hni
que
s
.
I
n
th
e
c
ur
r
e
nt
r
e
s
e
a
r
c
h,
th
e
nove
lt
y
of
th
e
w
or
k
ha
s
be
e
n
done
w
it
h
th
e
in
te
gr
a
ti
on
of
f
iv
e
e
f
f
e
c
ti
ve
a
lg
or
it
hm
s
,
in
c
lu
di
ng
th
e
W
O
A
,
M
V
O
,
G
A
,
A
C
O
,
a
nd
5
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on
.
N
or
m
a
ll
y,
th
e
s
e
te
c
hni
que
s
a
r
e
de
ve
lo
pe
d
in
M
L
to
c
om
pa
r
e
,
a
nd
th
e
e
f
f
e
c
ti
ve
te
c
hni
que
w
it
h
a
n
a
ppr
opr
ia
te
m
ode
l
f
or
a
pr
ovi
de
d
pr
e
di
c
ti
ve
m
ode
ll
in
g
is
s
ue
s
s
ol
v
e
d
a
nd
m
a
de
th
e
e
nt
ir
e
pr
oc
e
s
s
w
it
h
c
e
r
ta
in
b
e
ne
f
it
s
li
ke
s
im
pl
e
im
pl
e
m
e
nt
a
ti
on,
e
a
s
y
unde
r
s
ta
ndi
ng,
a
nd
lo
w
e
r
b
ia
s
e
s
ti
m
a
ti
on
c
om
p
a
r
e
d
to
ot
he
r
te
c
hni
que
s
.
F
ur
th
e
r
,
th
e
5
-
f
ol
d
C
V
ha
s
be
e
n
ut
il
iz
e
d
in
th
is
s
tu
dy,
a
nd
th
is
m
ode
l
e
va
lu
a
te
d
th
e
ne
w
da
ta
.
I
n
th
is
s
tu
dy,
a
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 3, J
une
2025
:
2328
-
2337
2336
nove
l
I
B
O
M
opt
im
iz
a
ti
on t
e
c
hni
que
ha
s
be
e
n ut
il
iz
e
d i
n or
de
r
t
o di
a
gnos
e
di
s
e
a
s
e
s
, a
nd da
ta
m
in
in
g c
onc
e
pt
s
a
r
e
ut
il
iz
e
d
f
or
s
to
r
in
g
a
la
r
ge
a
m
ount
of
m
e
di
c
a
l
da
t
a
w
it
hout
a
ny
in
te
r
r
upt
io
n
f
or
th
e
f
ur
th
e
r
id
e
nt
if
ic
a
ti
on
pr
oc
e
s
s
.
A
t
th
e
s
a
m
e
ti
m
e
,
va
r
io
us
opt
im
iz
a
ti
on
te
c
hni
que
s
ha
v
e
be
e
n
e
xpe
r
im
e
nt
e
d
w
it
h
a
nd
c
om
pa
r
e
d
w
it
h
th
e
pr
opos
e
d
te
c
hni
que
s
.
F
in
a
ll
y,
5
-
f
o
ld
c
r
os
s
-
va
li
da
ti
on
te
c
hni
que
s
pr
oc
ur
e
d
be
tt
e
r
r
e
s
ul
ts
w
it
h
0.98
a
c
c
ur
a
c
y
c
om
pa
r
e
d
to
th
e
ot
he
r
opt
im
iz
a
ti
on
te
c
hni
que
s
a
nd
w
e
r
e
de
not
e
d
a
s
th
e
be
s
t
m
ode
l
f
or
di
a
gnos
in
g
di
s
e
a
s
e
s
.
I
n
th
e
f
ut
ur
e
,
th
e
ot
he
r
opt
im
iz
a
ti
on
te
c
hni
que
s
w
il
l
be
c
om
pa
r
e
d
w
it
h
th
e
pr
opos
e
d
te
c
hni
que
t
o de
te
r
m
in
e
t
he
a
c
c
ur
a
c
y of
pr
e
di
c
ti
ng dis
e
a
s
e
s
.
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
A
ut
hor
s
s
ta
te
no f
undi
ng i
nvol
ve
d.
A
U
T
H
O
R
C
O
N
T
R
I
B
U
T
I
O
N
S
S
T
A
T
E
M
E
N
T
T
hi
s
jo
ur
na
l
us
e
s
th
e
C
ont
r
ib
ut
or
R
ol
e
s
T
a
xonomy
(
C
R
e
di
T
)
to
r
e
c
ogni
z
e
in
di
vi
dua
l
a
ut
hor
c
ont
r
ib
ut
io
ns
, r
e
duc
e
a
ut
hor
s
hi
p di
s
put
e
s
,
a
nd f
a
c
il
it
a
te
c
ol
la
bo
r
a
ti
on.
N
am
e
o
f
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
P
r
e
e
ti
T
ha
r
e
ja
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
R
a
je
nde
r
S
in
gh
C
hhi
ll
a
r
✓
✓
✓
✓
✓
✓
✓
C
:
C
onc
e
pt
ua
l
i
z
a
t
i
on
M
:
M
e
t
hodol
ogy
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
on
Fo
:
Fo
r
m
a
l
a
na
l
ys
i
s
I
:
I
nve
s
t
i
ga
t
i
on
R
:
R
e
s
our
c
e
s
D
:
D
a
t
a
C
ur
a
t
i
on
O
:
W
r
i
t
i
ng
-
O
r
i
gi
na
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
vi
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
on
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
ni
s
t
r
a
t
i
on
Fu
:
Fu
ndi
ng a
c
qui
s
i
t
i
on
C
O
N
F
L
I
C
T
O
F
I
N
T
E
R
E
S
T
S
T
A
T
E
M
E
N
T
A
ut
hor
s
s
ta
te
no c
onf
li
c
t
of
i
nt
e
r
e
s
t.
D
A
T
A
A
V
A
I
L
A
B
I
L
I
T
Y
T
he
da
ta
th
a
t
s
uppor
t
th
e
f
in
d
in
gs
of
th
is
s
tu
dy
a
r
e
ope
nl
y
a
va
il
a
bl
e
on
F
ig
s
ha
r
e
a
t
ht
tp
s
:/
/f
ig
s
ha
r
e
.c
om
/n
dow
nl
oa
de
r
/f
il
e
s
/5
353900?
pr
iv
a
te
_l
in
k=
8
5f
d25dd6127d1bda
36e
.
F
r
om
th
e
a
va
il
a
bl
e
da
ta
s
e
ts
,
onl
y
th
e
hum
a
n
da
ta
w
a
s
ut
il
iz
e
d
f
or
th
is
r
e
s
e
a
r
c
h.
T
he
da
ta
s
e
t
c
a
n
be
a
c
c
e
s
s
e
d
a
nd
dow
nl
o
a
de
d
f
r
om
t
he
pr
ovi
de
d l
in
k.
R
E
F
E
R
E
N
C
E
S
[
1]
Y
.
Z
hua
ng
e
t
al
.
,
“
D
e
e
p
l
e
a
r
ni
ng
on
gr
a
phs
f
or
m
ul
t
i
-
om
i
c
s
c
l
a
s
s
i
f
i
c
a
t
i
on
of
C
O
P
D
,”
P
L
O
S
O
N
E
,
vol
.
18,
no.
4,
2023
,
doi
:
10.1371/
j
our
na
l
.pone
.0284563.
[
2]
Z
.
G
a
o
e
t
al
.
,
“
H
i
e
r
a
r
c
hi
c
a
l
gr
a
ph
l
e
a
r
ni
ng
f
or
pr
ot
e
i
n
–
pr
ot
e
i
n
i
nt
e
r
a
c
t
i
on,”
N
at
ur
e
C
om
m
uni
c
at
i
on
s
,
vol
.
14,
no.
1,
2023
,
doi
:
10.1038/
s
41467
-
023
-
36736
-
1.
[
3]
Z
.
H
ou,
Y
.
Y
a
ng,
Z
.
M
a
,
K
.
c
hun
W
ong,
a
nd
X
.
L
i
,
“
L
e
a
r
ni
ng
t
he
pr
ot
e
i
n
l
a
ngua
ge
of
pr
ot
e
om
e
-
w
i
de
pr
ot
e
i
n
-
pr
ot
e
i
n
bi
ndi
ng
s
i
t
e
s
vi
a
e
xpl
a
i
na
bl
e
e
ns
e
m
bl
e
d
e
e
p l
e
a
r
ni
ng,”
C
om
m
uni
c
at
i
on
s
B
i
ol
ogy
, vol
. 6, no. 1, 2023, doi
:
10.1038/
s
42003
-
023
-
04462
-
5.
[
4]
J
.
L
e
vy
e
t
al
.
,
“
A
r
t
i
f
i
c
i
a
l
i
nt
e
l
l
i
ge
nc
e
,
bi
oi
nf
o
r
m
a
t
i
c
s
,
a
nd
pa
t
hol
ogy:
e
m
e
r
gi
ng
t
r
e
nds
pa
r
t
I
—
a
n
i
n
t
r
oduc
t
i
on
t
o
m
a
c
hi
ne
l
e
a
r
ni
ng
t
e
c
hnol
ogi
e
s
,”
A
dv
anc
e
s
i
n M
ol
e
c
ul
ar
P
at
hol
ogy
, vol
. 5, no. 1, pp. e
1
–
e
24, 202
2.
[
5]
Y
.
M
a
s
oudi
-
S
obha
nz
a
d
e
h
a
nd
A
.
M
a
s
oudi
-
N
e
j
a
d,
“
S
ynt
he
t
i
c
r
e
pur
pos
i
ng
of
dr
ugs
a
ga
i
ns
t
hype
r
t
e
n
s
i
on:
a
da
t
a
m
i
ni
ng
m
e
t
ho
d
ba
s
e
d
on
a
s
s
oc
i
a
t
i
on
r
ul
e
s
a
nd
a
nove
l
di
s
c
r
e
t
e
a
l
gor
i
t
hm
,”
B
M
C
B
i
oi
nf
or
m
at
i
c
s
,
vol
.
21,
no.
1,
pp.
1
-
21,
2020,
doi
:
10.1186/
s
12859
-
020
-
03644
-
w.
[
6]
T
.
T
.
L
e
,
W
.
F
u,
a
nd
J
.
H
.
M
oor
e
,
“
S
c
a
l
i
ng
t
r
e
e
-
ba
s
e
d
a
ut
om
a
t
e
d
m
a
c
hi
ne
l
e
a
r
ni
ng
t
o
bi
om
e
di
c
a
l
bi
g
d
a
t
a
w
i
t
h
a
f
e
a
t
ur
e
s
e
t
s
e
l
e
c
t
or
,”
B
i
oi
nf
or
m
at
i
c
s
, vol
. 36, no. 1, pp. 250
–
256, 2020, doi
:
10.1093/
bi
oi
nf
or
m
a
t
i
c
s
/
bt
z
470.
[
7]
T
.
T
a
ng
e
t
al
.
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
on
pr
ot
e
i
n
–
pr
ot
e
i
n
i
nt
e
r
a
c
t
i
on
pr
e
di
c
t
i
o
n:
m
ode
l
s
,
c
ha
l
l
e
nge
s
a
nd
t
r
e
nds
,”
B
r
i
e
f
i
ngs
i
n
B
i
oi
nf
or
m
at
i
c
s
, vol
. 24, no. 2, 2023, doi
:
10.1093/
bi
b/
bba
d076.
[
8]
P
.
M
a
ur
ya
a
nd
N
.
P
.
S
i
ngh,
“
M
us
hr
oom
c
l
a
s
s
i
f
i
c
a
t
i
on
us
i
ng
f
e
a
t
ur
e
-
ba
s
e
d
m
a
c
hi
ne
l
e
a
r
ni
ng
a
ppr
oa
c
h,”
i
n
P
r
oc
e
e
di
ngs
of
3r
d
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
C
om
put
e
r
V
i
s
i
on and I
m
age
P
r
oc
e
s
s
i
ng
, 2020, pp.
197
–
206
, doi
:
10.1007/
978
-
981
-
32
-
9088
-
4_17.
[
9]
P
.
T
ha
r
e
j
a
a
nd
R
.
S
.
C
hhi
l
l
a
r
,
“
P
ow
e
r
of
de
e
p
l
e
a
r
ni
ng
m
ode
l
s
i
n
bi
oi
nf
or
m
a
t
i
c
s
,”
i
n
I
nnov
at
i
ons
i
n
D
at
a
A
nal
y
t
i
c
s
,
2023
,
pp. 535
–
542
, doi
:
10.1007/
978
-
981
-
99
-
0550
-
8_42.
[
10]
R
.
J
.
U
r
ba
now
i
c
z
,
R
.
S
.
O
l
s
on,
P
.
S
c
hm
i
t
t
,
M
.
M
e
e
ke
r
,
a
nd
J
.
H
.
M
oor
e
,
“
B
e
nc
hm
a
r
ki
ng
r
e
l
i
e
f
-
ba
s
e
d
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
m
e
t
hod
s
f
or
bi
oi
nf
o
r
m
a
t
i
c
s
da
t
a
m
i
ni
ng,”
J
our
nal
of
B
i
om
e
di
c
al
I
nf
or
m
at
i
c
s
,
vol
.
85,
pp.
168
–
188,
S
e
p.
2018,
doi
:
10.1016/
j
.j
bi
.2018.07.015.
[
11]
H
.
G
a
o,
C
.
C
h
e
n,
S
.
L
i
,
C
.
W
a
ng,
W
.
Z
hou,
a
nd
B
.
Y
u,
“
P
r
e
di
c
t
i
on
of
pr
ot
e
i
n
-
pr
ot
e
i
n
i
nt
e
r
a
c
t
i
ons
ba
s
e
d
on
e
n
s
e
m
bl
e
r
e
s
i
dua
l
c
onvol
ut
i
ona
l
ne
ur
a
l
ne
t
w
or
k,”
C
om
put
e
r
s
i
n B
i
ol
ogy
and M
e
di
c
i
ne
, vol
. 152,
2023, doi
:
10.1016/
j
.c
om
pbi
om
e
d.2022.106471.
[
12]
H
.
L
uo,
“
P
r
ot
e
om
i
c
a
nd
ge
nom
i
c
da
t
a
m
i
ni
ng
w
i
t
h
a
ppl
i
c
a
t
i
ons
i
n
pl
a
nt
s
c
i
e
nc
e
,”
P
h.D
.
t
he
s
i
s
,
D
e
pa
r
t
m
e
nt
of
B
i
oi
nf
o
r
m
a
t
i
c
s
,
W
a
ge
ni
nge
n U
ni
ve
r
s
i
t
y, W
a
ge
ni
ng
e
n, N
e
t
he
r
l
a
nds
, 2023
, doi
:
10.18174/
57976
7.
[
13]
R
.
S
yr
l
yba
e
va
a
nd
E
.
M
.
S
t
r
a
uc
h,
“
D
e
e
p
l
e
a
r
ni
ng
of
pr
ot
e
i
n
s
e
que
nc
e
de
s
i
gn
of
pr
ot
e
i
n
–
pr
ot
e
i
n
i
nt
e
r
a
c
t
i
ons
,”
B
i
oi
nf
or
m
at
i
c
s
,
vol
. 39, no. 1, 2023, doi
:
10.1093/
bi
oi
nf
or
m
a
t
i
c
s
/
bt
a
c
733.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
O
pt
imi
z
in
g bi
oi
nf
or
m
at
ic
s
appli
c
at
io
ns
:
a nov
e
l
app
r
oac
h w
it
h
hum
an pr
ot
e
in
dat
a
…
(
P
r
e
e
ti
T
har
e
ja
)
2337
[
14]
W
.
H
u
a
nd
M
.
O
hu
e
,
“
S
pa
t
i
a
l
P
P
I
:
t
hr
e
e
-
di
m
e
ns
i
ona
l
s
pa
c
e
pr
ot
e
i
n
-
pr
ot
e
i
n
i
nt
e
r
a
c
t
i
on
pr
e
di
c
t
i
on
w
i
t
h
A
l
pha
F
ol
d
m
ul
t
i
m
e
r
,”
C
om
put
at
i
onal
and St
r
uc
t
ur
al
B
i
ot
e
c
hnol
ogy
J
our
nal
, vol
. 23, pp. 1214
–
1225,
2024, doi
:
10.1016/
j
.c
s
bj
.2024.03.009.
[
15]
M
.
Y
.
C
a
o,
S
.
Z
a
i
nudi
n,
a
nd
K
.
M
.
D
a
ud,
“
P
r
ot
e
i
n
f
e
a
t
ur
e
s
f
us
i
on
us
i
ng
a
t
t
r
i
but
e
d
ne
t
w
or
k
e
m
be
ddi
ng
f
o
r
pr
e
di
c
t
i
ng
pr
ot
e
i
n
-
pr
ot
e
i
n i
nt
e
r
a
c
t
i
on,”
B
M
C
G
e
nom
i
c
s
, vol
. 25, no. 1, 2024, doi
:
10.1186/
s
12864
-
024
-
10361
-
8.
[
16]
H
.
A
.
G
ündüz
e
t
al
.
,
“
O
pt
i
m
i
z
e
d
m
ode
l
a
r
c
hi
t
e
c
t
ur
e
s
f
or
de
e
p
l
e
a
r
ni
ng
on
ge
n
om
i
c
da
t
a
,”
C
om
m
uni
c
at
i
ons
B
i
ol
ogy
,
vol
.
7,
no.
1,
2024, doi
:
10.1038/
s
42003
-
024
-
06161
-
1.
[
17]
T
.
H
.
D
a
n
g
a
nd
T
.
A
.
V
u
,
“
xC
A
P
T
5:
p
r
o
t
e
i
n
–
pr
ot
e
i
n
i
n
t
e
r
a
c
t
i
on
p
r
e
d
i
c
t
i
o
n
us
i
n
g
de
e
p
a
nd
w
i
de
m
ul
t
i
-
ke
r
ne
l
p
oo
l
i
ng
c
on
vol
ut
i
o
na
l
ne
u
r
a
l
ne
t
w
or
ks
w
i
t
h
pr
ot
e
i
n
l
a
ng
ua
g
e
m
ode
l
,
”
B
M
C
B
i
oi
nf
or
m
a
t
i
c
s
,
vol
.
25,
n
o.
1,
202
4,
do
i
:
1
0.1
18
6/
s
1
28
59
-
0
24
-
0
572
5
-
6.
[
18]
N
.
Y
.
A
hm
e
d
e
t
al
.
,
“
A
n
e
f
f
i
c
i
e
nt
de
e
p
l
e
a
r
ni
ng
a
ppr
oa
c
h
f
or
D
N
A
-
bi
ndi
ng
pr
ot
e
i
ns
c
l
a
s
s
i
f
i
c
a
t
i
on
f
r
om
pr
i
m
a
r
y
s
e
que
nc
e
s
,
”
I
nt
e
r
nat
i
onal
J
our
nal
of
C
om
put
at
i
onal
I
nt
e
l
l
i
ge
nc
e
Sy
s
t
e
m
s
, vol
. 17, no. 1, 20
24, doi
:
10.1007/
s
44196
-
024
-
00462
-
3.
[
19]
B
.
Y
u,
C
.
C
h
e
n,
H
.
Z
hou,
B
.
L
i
u,
a
nd
Q
.
M
a
,
“
G
T
B
-
P
P
I
:
pr
e
di
c
t
pr
ot
e
i
n
–
pr
ot
e
i
n
i
nt
e
r
a
c
t
i
ons
ba
s
e
d
on
L
1
-
r
e
gul
a
r
i
z
e
d
l
ogi
s
t
i
c
r
e
gr
e
s
s
i
on
a
nd
gr
a
di
e
nt
t
r
e
e
boos
t
i
ng,”
G
e
nom
i
c
s
,
P
r
ot
e
om
i
c
s
and
B
i
oi
nf
or
m
at
i
c
s
,
vol
.
18,
no.
5,
pp.
582
–
592,
2020,
doi
:
10.1016/
j
.gpb.2021.01.001.
[
20]
S
.
S
i
m
s
e
k,
U
.
K
ur
s
unc
u,
E
.
K
i
bi
s
,
M
.
A
.
A
bde
l
l
a
t
i
f
,
a
nd
A
.
D
a
g,
“
A
hybr
i
d
da
t
a
m
i
ni
ng
a
ppr
oa
c
h
f
or
i
de
nt
i
f
yi
ng
t
he
t
e
m
por
a
l
e
f
f
e
c
t
s
of
va
r
i
a
bl
e
s
a
s
s
oc
i
a
t
e
d
w
i
t
h
br
e
a
s
t
c
a
nc
e
r
s
ur
vi
va
l
,”
E
x
pe
r
t
Sy
s
t
e
m
s
w
i
t
h
A
ppl
i
c
at
i
ons
,
vol
.
139,
2020
,
doi
:
10.1016/
j
.e
s
w
a
.2019.112863.
[
21]
K
. V
ouga
s
e
t
al
.
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng a
nd
da
t
a
m
i
ni
ng f
r
a
m
e
w
or
ks
f
or
pr
e
di
c
t
i
ng dr
ug r
e
s
pons
e
i
n
c
a
nc
e
r
:
a
n
ove
r
vi
e
w
a
nd a
nove
l
i
n
s
i
l
i
c
o
s
c
r
e
e
ni
ng
pr
oc
e
s
s
ba
s
e
d
on
a
s
s
oc
i
a
t
i
on
r
ul
e
m
i
ni
ng,”
P
har
m
ac
ol
ogy
and
T
he
r
ape
ut
i
c
s
,
vol
.
203,
2019,
doi
:
10.1016/
j
.pha
r
m
t
he
r
a
.2019.107395.
[
22]
H
.
T
ha
kka
r
,
V
.
S
ha
h,
H
.
Y
a
gni
k,
a
nd
M
.
S
ha
h,
“
C
om
pa
r
a
t
i
ve
a
na
t
om
i
z
a
t
i
on
of
da
t
a
m
i
ni
ng
a
nd
f
uz
z
y
l
ogi
c
t
e
c
hni
que
s
us
e
d
i
n
di
a
be
t
e
s
pr
ognos
i
s
,
”
C
l
i
ni
c
al
e
H
e
al
t
h
, vol
. 4, pp. 12
–
23, 2021, doi
:
10.1016/
j
.c
e
h.2020.11.001.
[
23]
S
.
V
.
K
ova
l
c
huk,
A
.
A
.
F
unkne
r
,
O
.
G
.
M
e
t
s
ke
r
,
a
nd
A
.
N
.
Y
a
kovl
e
v,
“
S
i
m
ul
a
t
i
on
of
pa
t
i
e
nt
f
l
ow
i
n
m
ul
t
i
pl
e
he
a
l
t
hc
a
r
e
uni
t
s
us
i
ng
pr
oc
e
s
s
a
nd
da
t
a
m
i
ni
ng
t
e
c
hni
que
s
f
or
m
ode
l
i
de
nt
i
f
i
c
a
t
i
on,”
J
our
nal
of
B
i
om
e
di
c
al
I
nf
or
m
at
i
c
s
,
vol
.
82,
pp.
128
–
142,
2018, doi
:
10.1016/
j
.j
bi
.2018.05.004.
[
24]
H
.
Y
a
ng
e
t
al
.
,
“
A
dm
e
t
S
A
R
2.0:
W
e
b
-
s
e
r
vi
c
e
f
or
pr
e
di
c
t
i
on
a
nd
opt
i
m
i
z
a
t
i
on
of
c
he
m
i
c
a
l
A
D
M
E
T
pr
ope
r
t
i
e
s
,”
B
i
oi
nf
or
m
at
i
c
s
,
vol
. 35, no. 6, pp. 1067
–
1069, 2019, doi
:
10.1093/
bi
oi
nf
or
m
a
t
i
c
s
/
bt
y707.
[
25]
T
.
S
e
nj
yu,
A
.
Y
.
S
a
be
r
,
T
.
M
i
ya
gi
,
K
.
S
hi
m
a
buku
r
o,
N
.
U
r
a
s
a
ki
,
a
nd
T
.
F
u
na
ba
s
hi
,
“
F
a
s
t
t
e
c
hni
que
f
or
uni
t
c
om
m
i
t
m
e
nt
by
ge
ne
t
i
c
a
l
gor
i
t
hm
ba
s
e
d
on
uni
t
c
l
us
t
e
r
i
ng,”
I
E
E
P
r
oc
e
e
di
ngs
:
G
e
ne
r
at
i
on,
T
r
ans
m
i
s
s
i
on
and
D
i
s
t
r
i
but
i
on
,
vol
.
152,
no.
5,
pp. 705
–
713, 2005, doi
:
10.1049/
i
p
-
gt
d:
20045299.
[
26]
M
.
F
.
K
ha
n,
F
.
A
a
di
l
,
M
.
M
a
qs
ood,
S
.
H
.
R
.
B
ukha
r
i
,
M
.
H
us
s
a
i
n,
a
nd
Y
.
N
a
m
,
“
M
ot
h
f
l
a
m
e
c
l
u
s
t
e
r
i
ng
a
l
gor
i
t
hm
f
or
i
nt
e
r
ne
t
of
ve
hi
c
l
e
(
M
F
C
A
-
I
oV
)
,”
I
E
E
E
A
c
c
e
s
s
, vol
. 7, pp. 11613
–
11629, 2019, doi
:
10.1109/
A
C
C
E
S
S
.2018.2886420.
[
27]
S
.
M
a
ha
pa
t
r
a
a
nd
S
.
S
.
S
a
hu,
“
A
N
O
V
A
-
pa
r
t
i
c
l
e
s
w
a
r
m
opt
i
m
i
z
a
t
i
on
-
ba
s
e
d
f
e
a
t
ur
e
s
e
l
e
c
t
i
on
a
nd
gr
a
di
e
nt
boo
s
t
i
ng
m
a
c
hi
n
e
c
l
a
s
s
i
f
i
e
r
f
or
i
m
pr
ove
d
p
r
ot
e
i
n
–
pr
ot
e
i
n
i
n
t
e
r
a
c
t
i
on
pr
e
di
c
t
i
on,”
P
r
ot
e
i
ns
:
St
r
uc
t
ur
e
,
F
unc
t
i
on
and
B
i
oi
n
f
or
m
at
i
c
s
,
vol
.
90,
no.
2,
pp. 443
–
454, 2022, doi
:
10.1002/
pr
ot
.26236.
[
28]
P
.
T
ha
r
e
j
a
a
nd
R
.
S
.
C
hhi
l
l
a
r
,
“
A
de
t
a
i
l
e
d
s
ur
ve
y
on
da
t
a
m
i
ni
ng
ba
s
e
d
opt
i
m
i
z
a
t
i
on
s
c
he
m
e
s
f
or
bi
oi
nf
or
m
a
t
i
c
s
a
ppl
i
c
a
t
i
ons
,
”
E
C
S T
r
ans
ac
t
i
ons
, vol
. 107, no. 1, pp. 4689
–
4696, 2022, doi
:
10.1149/
10701.4
689e
c
s
t
.
[
29]
M
.
O
.
A
r
ow
ol
o,
M
.
O
.
A
de
bi
yi
,
A
.
A
.
A
de
bi
yi
,
a
nd
O
.
O
l
ugba
r
a
,
“
O
pt
i
m
i
z
e
d
h
ybr
i
d
i
nve
s
t
i
ga
t
i
ve
ba
s
e
d
di
m
e
ns
i
ona
l
i
t
y
r
e
duc
t
i
on
m
e
t
hods
f
or
m
a
l
a
r
i
a
ve
c
t
or
us
i
ng K
N
N
c
l
a
s
s
i
f
i
e
r
,”
J
our
nal
of
B
i
g D
at
a
, vol
. 8, no. 1, 2021, doi
:
10.1186/
s
40537
-
021
-
00415
-
z.
[
30]
X
.
D
u,
S
.
S
un,
C
.
H
u,
Y
.
Y
a
o,
Y
.
Y
a
n,
a
nd
Y
.
Z
ha
ng,
“
D
e
e
pP
P
I
:
B
oos
t
i
ng
pr
e
di
c
t
i
on
of
pr
o
t
e
i
n
-
pr
ot
e
i
n
i
nt
e
r
a
c
t
i
ons
w
i
t
h
de
e
p
ne
ur
a
l
ne
t
w
or
ks
,”
J
our
nal
of
C
he
m
i
c
al
I
nf
or
m
at
i
on
and
M
ode
l
i
ng
,
vol
.
57,
no.
6,
pp.
1499
–
1510,
2017,
doi
:
10.1021/
a
c
s
.j
c
i
m
.7b00028.
[
31]
Y
.
D
i
ng,
J
.
T
a
ng,
a
nd
F
.
G
uo,
“
P
r
e
di
c
t
i
ng
pr
ot
e
i
n
-
pr
ot
e
i
n
i
nt
e
r
a
c
t
i
ons
vi
a
m
ul
t
i
va
r
i
a
t
e
m
ut
ua
l
i
nf
or
m
a
t
i
on
of
pr
ot
e
i
n
s
e
que
nc
e
s
,
”
B
M
C
B
i
oi
nf
or
m
at
i
c
s
, vol
. 17, no. 1, 2016, doi
:
10.1186/
s
12859
-
016
-
1253
-
9.
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Preeti
Thareja
is
a
computer
science
rese
arch
scholar
at
Mahar
shi
Dayanand
University
in
Rohtak,
Haryana,
India.
Data
mining,
artificia
l
int
ell
igence,
soft
computing,
deep
learning
is
among
her
rese
arch
interests.
Over
the
last
few
ye
ars,
she
has
published
5
journal
papers,
4
confere
nce
papers,
and
1
book
chapter,
as
well
as
two
books
in
the
subjects
of Python and
soft computing
. She can be contacted a
t email: preetithareja10@gmail.com.
Rajender
Singh
Chhillar
is
a
computer
science
professor
at
Maha
rshi
Dayanand
University
in
Rohtak,
Haryana,
India.
He
was
also
the
head
of
the
Department
of
Computer
Scienc
e,
the
Chairma
n
of
a
board
of
studies,
and
a
m
ember
of
the
exec
utive
and
acad
emic
councils
.
Software
engineerin
g,
software
testing
,
software
metrics,
w
eb
metrics,
bio
metrics,
data
warehouse
and
data
mining,
computer
networking,
and
software
design
are
among
his
research
interests.
Over
the
last
several
years,
he
has
produced
over
91
journal
papers
and
65
conference
papers,
as
well
as
two
books
in
the
subjects
of
sof
tware
engineering
and
information
technology
.
He
is
a
director
of
the
CMAI
Asia
Associatio
n
in
New
Delhi,
as
wel
l
as
a
senior
member
of
the
IACSIT
in
Singapore
and
a
member
of
t
he
Computer
Society
of
India. He can be
contacted at em
ail: r.chhi
llar@
mdurohtak.
ac.in
.
Evaluation Warning : The document was created with Spire.PDF for Python.