I
AE
S
I
nte
rna
t
io
na
l J
o
urna
l o
f
Art
if
icia
l
I
nte
llig
ence
(
I
J
-
AI
)
Vo
l.
7
,
No
.
2
,
J
u
n
e
201
8
,
p
p
.
95
~
1
0
4
I
SS
N:
2252
-
8938
,
DOI
: 1
0
.
1
1
5
9
1
/i
j
ai.
v
7
.
i2
.
p
p
95
-
104
95
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e.
co
m/jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JA
I
Ev
o
lutiona
ry
Com
pu
tatio
na
l Algo
rith
m
by
B
l
endin
g
o
f
PP
CA
a
nd
EP
-
Enha
nc
e
d Supervis
ed
Clas
sifier
for
M
icro
a
rray
G
ene
Ex
press
io
n
Da
ta
M
a
na
s
w
ini
P
ra
dh
a
n
P
.
G
De
p
a
rt
m
e
n
t
o
f
In
f
o
r
m
a
ti
o
n
a
n
d
C
o
m
m
u
n
ica
ti
o
n
T
e
c
h
n
o
l
o
g
y
,
F
a
k
ir
M
o
h
a
n
Un
iv
e
rsity
,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Feb
2
1
,
2
0
1
7
R
ev
i
s
ed
A
p
r
7
,
2
0
1
8
A
cc
ep
ted
Ma
y
1
9
,
2
0
1
8
In
DN
A
m
icro
a
rra
y
te
c
h
n
o
l
o
g
y
,
g
e
n
e
c
las
sif
i
c
a
ti
o
n
is
c
o
n
sid
e
re
d
to
b
e
d
if
f
icu
lt
b
e
c
a
u
se
th
e
a
tt
rib
u
tes
o
f
th
e
d
a
ta,
a
re
c
h
a
ra
c
t
e
rize
d
b
y
h
ig
h
d
im
e
n
sio
n
a
li
ty
a
n
d
sm
a
ll
sa
m
p
l
e
siz
e
.
Clas
sif
i
c
a
ti
o
n
o
f
ti
ss
u
e
s
a
m
p
les
in
su
c
h
h
ig
h
d
im
e
n
sio
n
a
l
p
r
o
b
lem
s
is
a
c
o
m
p
li
c
a
ted
tas
k
.
F
u
rth
e
rm
o
re
,
th
e
re
is
a
h
ig
h
re
d
u
n
d
a
n
c
y
in
m
icro
a
rra
y
d
a
ta
a
n
d
se
v
e
ra
l
g
e
n
e
s
c
o
m
p
rise
in
a
p
p
r
o
p
r
iate
in
f
o
rm
a
ti
o
n
f
o
r
a
c
c
u
ra
te
c
las
si
f
ica
ti
o
n
o
f
d
ise
a
se
s
o
r
p
h
e
n
o
ty
p
e
s.
Co
n
se
q
u
e
n
tl
y
,
a
n
e
ff
icie
n
t
c
las
sif
i
c
a
ti
o
n
tec
h
n
iq
u
e
is
n
e
c
e
ss
a
r
y
to
re
tri
e
v
e
th
e
g
e
n
e
in
f
o
rm
a
ti
o
n
f
ro
m
th
e
m
icro
a
rra
y
e
x
p
e
ri
m
e
n
tal
d
a
ta.
In
th
is
p
a
p
e
r,
a
c
las
si
f
ica
ti
o
n
tec
h
n
iq
u
e
is
p
ro
p
o
se
d
t
h
a
t
c
las
sif
ie
s
th
e
m
icro
a
rra
y
g
e
n
e
e
x
p
re
s
sio
n
d
a
ta
w
e
ll
.
In
th
e
p
ro
p
o
se
d
tec
h
n
iq
u
e
,
th
e
d
im
e
n
sio
n
a
li
ty
o
f
th
e
g
e
n
e
e
x
p
re
ss
io
n
d
a
tas
e
t
is
re
d
u
c
e
d
b
y
P
r
o
b
a
b
i
li
stic
P
CA
.
T
h
e
n
,
a
n
A
rti
f
ici
a
l
Ne
u
ra
l
Ne
t
w
o
rk
(
AN
N)
is
se
le
c
te
d
a
s
th
e
su
p
e
rv
ise
d
c
las
si
f
ier
a
n
d
it
is
e
n
h
a
n
c
e
d
u
si
n
g
Ev
o
lu
ti
o
n
a
ry
P
ro
g
ra
m
m
in
g
(EP
)
tec
h
n
iq
u
e
.
T
h
e
e
n
h
a
n
c
e
m
e
n
t
o
f
t
h
e
c
las
sif
ier
is
a
c
c
o
m
p
li
sh
e
d
b
y
o
p
ti
m
izin
g
th
e
d
im
e
n
sio
n
o
f
th
e
A
NN
.
T
h
e
e
n
h
a
n
c
e
d
c
las
sif
ier
is
train
e
d
u
sin
g
th
e
Ba
c
k
P
ro
p
a
g
a
ti
o
n
(
BP
)
a
lg
o
rit
h
m
a
n
d
s
o
th
e
B
P
e
rro
r
g
e
ts
m
in
i
m
ize
d
.
T
h
e
w
e
ll
-
train
e
d
A
N
N
h
a
s
th
e
c
a
p
a
c
it
y
o
f
c
las
si
fy
in
g
th
e
g
e
n
e
e
x
p
re
ss
io
n
d
a
ta
to
t
h
e
a
ss
o
c
iate
d
c
las
se
s.
T
h
e
p
ro
p
o
se
d
tec
h
n
i
q
u
e
is
e
v
a
lu
a
ted
b
y
c
las
si
f
ica
ti
o
n
p
e
rfo
rm
a
n
c
e
o
v
e
r
th
e
c
a
n
c
e
r
c
las
s
e
s,
Ac
u
te
M
y
e
lo
id
L
e
u
k
e
m
ia
(
A
M
L
)
a
n
d
Ac
u
t
e
Ly
m
p
h
o
b
las
ti
c
L
e
u
k
e
m
ia
(ALL
).
T
h
e
c
las
si
f
ic
a
ti
o
n
p
e
rf
o
rm
a
n
c
e
o
f
th
e
e
n
h
a
n
c
e
d
A
NN
c
la
ss
i
f
ier
is
c
o
m
p
a
re
d
o
v
e
r
th
e
e
x
isti
n
g
A
NN
c
la
ss
i
f
ier a
n
d
S
VM
c
las
sif
ier.
K
ey
w
o
r
d
:
A
r
ti
f
icial
Ne
u
r
al
Net
w
o
r
k
(
A
NN)
,
B
ac
k
p
r
o
p
ag
atio
n
(
B
P),
Di
m
e
n
s
io
n
alit
y
r
ed
u
cti
on
E
v
o
lu
tio
n
ar
y
p
r
o
g
r
a
m
m
i
n
g
(
E
P
)
,
M
icr
o
ar
r
ay
g
e
n
e
ex
p
r
es
s
io
n
d
ata,
P
r
o
b
ab
ilis
tic
P
C
A
(
P
P
C
A
)
,
Su
p
er
v
i
s
ed
c
lass
i
f
ier
,
Co
p
y
rig
h
t
©
2
0
1
8
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Ma
n
as
w
i
n
i P
r
ad
h
an
,
P
.
G
Dep
ar
tm
en
t o
f
I
n
f
o
r
m
atio
n
an
d
C
o
m
m
u
n
icatio
n
T
ec
h
n
o
lo
g
y
,
Fak
ir
Mo
h
a
n
U
n
iv
er
s
it
y
,
B
alaso
r
e
-
7
5
6
0
1
9
,
Od
is
h
a,
I
n
d
ia
.
E
m
ail:
m
r
s
.
m
an
a
s
w
i
n
i.p
r
ad
h
a
n
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
E
n
o
r
m
o
u
s
a
m
o
u
n
t
o
f
g
en
o
m
ic
an
d
p
r
o
teo
m
ic
d
ata
ar
e
a
v
aila
b
le
in
t
h
e
p
u
b
lic
d
o
m
ai
n
.
T
h
e
ab
ilit
y
t
o
p
r
o
ce
s
s
th
is
in
f
o
r
m
at
io
n
in
wa
y
s
th
a
t
ar
e
u
s
ef
u
l
to
h
u
m
a
n
k
in
d
is
b
ec
o
m
in
g
in
cr
ea
s
in
g
l
y
i
m
p
o
r
tan
t
[
1
]
.
T
h
e
co
m
p
u
tatio
n
al
r
ec
o
g
n
it
io
n
is
a
b
asic
s
tep
in
th
e
u
n
d
er
s
tan
d
i
n
g
o
f
a
g
e
n
o
m
e
a
n
d
it
is
o
n
e
o
f
th
e
ch
alle
n
g
es
i
n
th
e
a
n
al
y
s
i
s
o
f
n
e
w
l
y
s
e
q
u
en
ce
d
g
en
o
m
es.
Fo
r
an
al
y
zi
n
g
g
en
o
m
ic
s
eq
u
e
n
ce
s
an
d
f
o
r
i
n
ter
p
r
etin
g
g
en
e
s
,
p
r
ec
is
e
an
d
f
a
s
t
to
o
ls
ar
e
n
ec
ess
ar
y
[
2
]
.
I
n
s
u
c
h
s
it
u
ati
o
n
,
co
n
v
e
n
tio
n
al
an
d
m
o
d
er
n
s
i
g
n
al
p
r
o
ce
s
s
i
n
g
m
et
h
o
d
s
p
la
y
a
s
ig
n
i
f
ica
n
t
r
o
le
in
th
e
s
e
f
ield
s
[
1
]
.
A
r
elativ
e
l
y
n
e
w
ar
ea
i
n
b
io
-
in
f
o
r
m
a
tics
is
Gen
o
m
ic
s
i
g
n
a
l
p
r
o
ce
s
s
in
g
[
1
4
]
(
GSP
)
.
I
t
d
ea
ls
w
it
h
t
h
e
u
tili
za
tio
n
o
f
tr
ad
iti
o
n
al
d
ig
ita
l
s
i
g
n
al
p
r
o
ce
s
s
i
n
g
(
DSP
)
m
et
h
o
d
s
i
n
th
e
r
ep
r
esen
tatio
n
an
d
an
al
y
s
i
s
o
f
g
e
n
o
m
ic
d
ata.
Gen
e
i
s
a
s
e
g
m
en
t
o
f
DN
A
,
w
h
ic
h
co
n
tain
s
t
h
e
co
d
e
f
o
r
t
h
e
c
h
e
m
ical
co
m
p
o
s
itio
n
o
f
a
p
ar
ticu
lar
p
r
o
tein
.
Gen
e
s
s
er
v
e
a
s
t
h
e
p
atter
n
f
o
r
p
r
o
tein
s
a
n
d
s
o
m
e
ad
d
itio
n
al
p
r
o
d
u
cts,
a
n
d
m
R
N
A
i
s
t
h
e
m
ai
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
7
,
No
.
2
,
J
u
n
e
201
8
:
95
–
1
0
4
96
in
ter
m
ed
iar
y
th
a
t
tr
an
s
la
tes
g
en
e
i
n
f
o
r
m
atio
n
in
t
h
e
p
r
o
d
u
ctio
n
o
f
g
en
et
icall
y
e
n
co
d
ed
m
o
lec
u
les
[
4
]
.
T
h
e
s
tr
an
d
s
o
f
DN
A
m
o
lec
u
les
u
s
u
all
y
co
n
ta
in
t
h
e
g
en
o
m
ic
i
n
f
o
r
m
atio
n
r
ep
r
esen
te
d
b
y
s
eq
u
en
ce
s
o
f
n
u
cleo
tid
e
s
y
m
b
o
ls
,
s
y
m
b
o
lic
co
d
o
n
s
(
tr
ip
lets
o
f
n
u
cleo
tid
e
s
)
,
o
r
s
y
m
b
o
lic
s
eq
u
e
n
ce
s
o
f
a
m
i
n
o
ac
id
s
in
t
h
e
co
r
r
esp
o
n
d
in
g
p
o
l
y
p
ep
tid
e
ch
ain
s
[
2
]
.
Si
m
u
lta
n
eo
u
s
l
y
m
o
n
ito
r
in
g
o
f
t
h
e
e
x
p
r
ess
io
n
lev
els
o
f
te
n
s
o
f
th
o
u
s
an
d
s
o
f
g
e
n
es
u
n
d
er
d
iv
er
s
e
ex
p
er
i
m
e
n
tal
co
n
d
itio
n
s
h
as
b
ee
n
e
n
ab
led
b
y
g
e
n
e
ex
p
r
ess
io
n
m
icr
o
c
h
ip
,
w
h
ic
h
is
p
er
h
ap
s
t
h
e
m
o
s
t
r
a
p
id
ly
e
x
p
an
d
i
n
g
to
o
l
o
f
g
e
n
o
m
e
an
a
l
y
s
is
.
T
h
is
p
r
o
v
id
es
a
p
o
w
er
f
u
l
to
o
l
in
t
h
e
s
tu
d
y
o
f
co
llec
ti
v
e
g
e
n
e
r
ea
ctio
n
to
c
h
an
g
es
in
th
e
ir
en
v
ir
o
n
m
en
ts
,
a
n
d
p
r
o
v
id
es
i
n
d
icatio
n
s
ab
o
u
t
th
e
s
tr
u
ct
u
r
es o
f
t
h
e
i
n
v
o
l
v
ed
g
e
n
e
n
et
w
o
r
k
s
[
3
]
.
T
o
d
ay
,
u
s
i
n
g
m
icr
o
ar
r
ay
s
it
is
p
o
s
s
ib
le
to
s
i
m
u
lta
n
eo
u
s
l
y
m
e
asu
r
e
t
h
e
e
x
p
r
ess
io
n
le
v
els
o
f
th
o
u
s
an
d
s
o
f
g
en
e
s
,
p
o
s
s
ib
l
y
all
g
en
e
s
in
an
o
r
g
an
is
m
,
i
n
a
s
in
g
le
ex
p
e
r
i
m
en
t
[
4
]
.
Mic
r
o
a
r
r
ay
tech
n
o
l
o
g
y
h
a
s
b
ec
o
m
e
an
in
d
is
p
en
s
ab
le
to
o
l
in
th
e
m
o
n
i
to
r
in
g
o
f
g
en
o
m
e
-
w
id
e
ex
p
r
es
s
io
n
le
v
els
o
f
g
en
e
[
5
]
.
T
h
e
a
n
al
y
s
i
s
o
f
th
e
g
en
e
ex
p
r
ess
io
n
p
r
o
f
ile
s
in
v
ar
i
o
u
s
o
r
g
an
s
u
s
in
g
m
icr
o
ar
r
a
y
tech
n
o
l
o
g
i
e
s
r
ev
ea
l
ab
o
u
t
s
ep
ar
ate
g
e
n
es,
g
en
e
e
n
s
e
m
b
les,
an
d
t
h
e
m
eta
b
o
lic
w
a
y
s
u
n
d
er
l
y
i
n
g
t
h
e
s
tr
u
ctu
r
al
l
y
f
u
n
ctio
n
al
o
r
g
an
izat
i
o
n
o
f
o
r
g
an
an
d
its
p
h
y
s
io
lo
g
ical
f
u
n
ctio
n
[
6
]
.
Diag
n
o
s
tic
tas
k
ca
n
b
e
au
to
m
a
te
d
an
d
th
e
ac
cu
r
ac
y
o
f
t
h
e
co
n
v
en
t
io
n
al
d
iag
n
o
s
tic
m
et
h
o
d
s
ca
n
b
e
i
m
p
r
o
v
ed
b
y
th
e
ap
p
licatio
n
o
f
m
icr
o
ar
r
ay
tec
h
n
o
lo
g
y
.
Mic
r
o
ar
r
a
y
t
ec
h
n
o
lo
g
y
e
n
ab
les
s
i
m
u
lta
n
eo
u
s
ex
a
m
i
n
atio
n
o
f
t
h
o
u
s
a
n
d
s
o
f
g
e
n
e
ex
p
r
ess
io
n
s
[
7
]
.
E
f
f
icien
t
r
ep
r
esen
tatio
n
o
f
ce
ll
ch
ar
ac
ter
izatio
n
at
t
h
e
m
o
l
ec
u
lar
lev
el
i
s
p
o
s
s
ib
le
w
it
h
m
icr
o
ar
r
a
y
tech
n
o
lo
g
y
w
h
ic
h
s
i
m
u
lta
n
e
o
u
s
l
y
m
ea
s
u
r
es
t
h
e
ex
p
r
ess
i
o
n
lev
el
s
o
f
te
n
s
o
f
t
h
o
u
s
a
n
d
s
o
f
g
e
n
es
[
8
]
.
Gen
e
e
x
p
r
ess
io
n
an
a
l
y
s
is
[
1
0
]
[
1
8
]
th
at
u
til
izes
m
icr
o
ar
r
ay
tec
h
n
o
lo
g
y
h
a
s
a
w
id
e
r
a
n
g
e
o
f
p
o
ten
tial
f
o
r
ex
p
lo
r
in
g
th
e
b
io
lo
g
y
o
f
ce
lls
an
d
o
r
g
an
is
m
s
[
9
]
.
Mic
r
o
ar
r
ay
tec
h
n
o
lo
g
y
ass
is
t
s
i
n
t
h
e
p
r
ec
is
e
p
r
ed
ictio
n
a
n
d
d
iag
n
o
s
i
s
o
f
d
is
ea
s
e
s
.
T
h
r
ee
co
m
m
o
n
t
y
p
e
s
o
f
m
ac
h
i
n
e
lear
n
in
g
tec
h
n
iq
u
e
s
u
tili
ze
d
i
n
m
icr
o
ar
r
ay
d
ata
an
al
y
s
is
ar
e
cl
u
s
ter
i
n
g
[
1
1
]
[
1
5
]
,
class
if
ica
tio
n
[
1
2
]
[
1
6
]
,
an
d
f
ea
tu
r
e
s
elec
tio
n
[
1
3
]
[
1
7
]
:
O
f
th
ese,
class
i
f
icatio
n
p
la
y
s
a
cr
u
cial
r
o
le
in
t
h
e
f
ield
o
f
m
icr
o
ar
r
ay
t
ec
h
n
o
lo
g
y
.
Ho
w
e
v
er
,
clas
s
i
f
ic
atio
n
i
n
m
icr
o
ar
r
ay
tech
n
o
lo
g
y
i
s
co
n
s
id
er
ed
to
b
e
v
er
y
ch
al
len
g
i
n
g
b
ec
au
s
e
o
f
th
e
h
i
g
h
d
i
m
e
n
s
io
n
alit
y
a
n
d
s
m
all
s
a
m
p
le
s
ize
o
f
th
e
g
e
n
e
ex
p
r
ess
io
n
d
ata.
Nu
m
er
o
u
s
w
o
r
k
s
h
av
e
b
ee
n
ca
r
r
i
ed
o
u
t
f
o
r
th
e
ef
f
ec
ti
v
e
class
i
f
icatio
n
o
f
th
e
g
e
n
e
ex
p
r
ess
io
n
d
ata.
A
f
e
w
r
ec
en
t
w
o
r
k
s
av
ai
lab
le
in
t
h
e
liter
atu
r
e
ar
e
r
ev
ie
w
ed
in
t
h
e
f
o
llo
w
in
g
s
ec
tio
n
.
2.
RE
L
AT
E
D
WO
RK
S
So
m
e
o
f
t
h
e
r
ec
e
n
t
r
elate
d
r
esear
ch
w
o
r
k
s
ar
e
r
e
v
ie
w
ed
h
er
e.
L
iu
et
a
l.
[
1
9
]
h
a
v
e
o
f
f
er
ed
an
an
al
y
tical
m
et
h
o
d
f
o
r
ca
te
g
o
r
izin
g
t
h
e
g
en
e
ex
p
r
es
s
io
n
d
ata.
I
n
t
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
,
d
im
en
s
io
n
r
ed
u
ctio
n
h
as
b
ee
n
ac
h
iev
ed
b
y
u
t
ilizi
n
g
t
h
e
k
er
n
el
p
r
i
n
cip
al
co
m
p
o
n
en
t
a
n
al
y
s
i
s
(
KP
C
A
)
an
d
ca
te
g
o
r
izatio
n
h
as
b
ee
n
ac
h
iev
ed
b
y
u
tili
z
in
g
t
h
e
lo
g
i
s
tic
r
eg
r
es
s
io
n
(
d
i
s
cr
i
m
i
n
atio
n
)
.
KP
C
A
is
a
g
en
er
ic
n
o
n
li
n
ea
r
f
o
r
m
o
f
p
r
i
n
cip
al
co
m
p
o
n
e
n
t
an
al
y
s
is
.
F
iv
e
v
ar
ied
g
en
e
e
x
p
r
ess
io
n
d
atas
ets
r
elate
d
to
h
u
m
a
n
t
u
m
o
r
s
a
m
p
le
s
h
as
b
ee
n
c
ateg
o
r
ized
b
y
u
tili
zi
n
g
th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
.
T
h
e
h
ig
h
p
o
ten
tial
o
f
t
h
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
i
n
ca
te
g
o
r
izin
g
g
en
e
ex
p
r
es
s
io
n
d
ata
h
a
s
b
ee
n
co
n
f
ir
m
ed
b
y
co
m
p
ar
in
g
with
o
t
h
er
w
ell
-
k
n
o
w
n
cla
s
s
i
f
ic
atio
n
m
et
h
o
d
s
li
k
e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
es
a
n
d
n
eu
r
al
n
et
w
o
r
k
s
.
R
o
b
er
to
R
u
iz
et
al.
[
2
0
]
h
av
e
p
r
o
p
o
s
e
d
a
n
o
v
el
h
eu
r
i
s
tic
m
et
h
o
d
f
o
r
s
elec
ti
n
g
ap
p
r
o
p
r
iate
g
e
n
e
s
u
b
s
ets
w
h
ic
h
ca
n
b
e
u
tili
ze
d
in
t
h
e
cla
s
s
i
f
icati
o
n
tas
k
.
Stati
s
tica
l
s
ig
n
i
f
ica
n
ce
o
f
t
h
e
i
n
cl
u
s
io
n
o
f
a
g
e
n
e
to
t
h
e
f
in
al
s
u
b
s
et
f
r
o
m
a
n
o
r
d
er
ed
lis
t
is
t
h
e
cr
it
er
ia
o
n
w
h
ic
h
t
h
eir
m
et
h
o
d
is
b
ased
.
C
o
m
p
ar
is
o
n
r
esu
lt
h
a
s
p
r
o
v
ed
th
at
t
h
e
m
e
th
o
d
w
a
s
m
o
r
e
ef
f
ec
ti
v
e
an
d
ef
f
icien
t
th
a
n
o
th
er
s
u
c
h
h
eu
r
i
s
tic
m
et
h
o
d
s
.
T
h
eir
m
et
h
o
d
ex
h
ib
it
s
o
u
t
s
ta
n
d
in
g
p
er
f
o
r
m
an
ce
b
o
th
i
n
id
e
n
ti
f
i
ca
tio
n
o
f
i
m
p
o
r
ta
n
t
g
en
e
s
an
d
i
n
m
i
n
i
m
izat
io
n
o
f
co
m
p
u
tatio
n
al
co
s
t.
P
en
g
et
a
l
.
[
2
1
]
h
av
e
p
er
f
o
r
m
ed
a
co
m
p
ar
ativ
e
a
n
al
y
s
is
o
n
d
if
f
er
e
n
t
b
io
m
ar
k
er
d
is
co
v
er
y
m
eth
o
d
s
th
at
i
n
cl
u
d
es
s
i
x
f
ilter
m
et
h
o
d
s
an
d
th
r
ee
w
r
ap
p
er
m
e
th
o
d
s
.
Af
ter
t
h
i
s
,
t
h
e
y
h
a
v
e
p
r
ese
n
te
d
a
h
y
b
r
id
ap
p
r
o
ac
h
k
n
o
w
n
a
s
FR
-
W
r
ap
p
er
f
o
r
b
io
m
ar
k
er
d
is
co
v
er
y
.
T
h
e
o
b
j
ec
ti
v
e
o
f
th
eir
ap
p
r
o
ac
h
w
as
to
ac
h
iev
e
a
n
o
p
ti
m
u
m
b
alan
ce
b
et
w
ee
n
p
r
ec
is
io
n
an
d
co
m
p
u
tatio
n
co
s
t,
b
y
e
x
p
lo
itin
g
t
h
e
e
f
f
icie
n
c
y
o
f
t
h
e
f
ilt
er
m
et
h
o
d
an
d
t
h
e
ac
cu
r
ac
y
o
f
t
h
e
w
r
ap
p
er
m
et
h
o
d
.
I
n
th
e
ir
h
y
b
r
id
ap
p
r
o
ac
h
,
th
e
m
aj
o
r
ity
o
f
t
h
e
u
n
r
elate
d
g
en
e
s
h
a
v
e
b
ee
n
f
ilter
ed
o
u
t
u
tili
z
in
g
t
h
e
Fi
s
h
e
r
’
s
r
atio
m
e
th
o
d
,
w
h
ic
h
i
s
s
i
m
p
le,
ea
s
y
to
u
n
d
er
s
ta
n
d
a
n
d
i
m
p
le
m
e
n
t.
T
h
en
t
h
e
r
ed
u
n
d
an
c
y
h
as
b
ee
n
m
i
n
i
m
iz
ed
u
tili
zi
n
g
a
w
r
ap
p
er
m
et
h
o
d
.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
F
R
-
W
r
ap
p
e
r
ap
p
r
o
ac
h
h
as
b
ee
n
ap
p
r
aised
u
tili
zi
n
g
f
o
u
r
w
id
el
y
u
s
ed
m
icr
o
ar
r
ay
d
atasets
.
E
x
p
er
i
m
en
ta
l
r
esu
lt
s
h
av
e
p
r
o
v
ed
th
at
t
h
e
h
y
b
r
id
ap
p
r
o
ac
h
is
ca
p
ab
le
o
f
ac
h
iev
in
g
m
ax
i
m
u
m
r
elev
a
n
ce
w
it
h
m
in
i
m
u
m
r
ed
u
n
d
an
c
y
.
Mr
a
m
o
r
et
al.
[
2
2
]
h
av
e
p
r
o
p
o
s
ed
a
m
et
h
o
d
f
o
r
th
e
a
n
al
y
s
i
s
o
f
g
en
e
e
x
p
r
es
s
io
n
d
ata
t
h
a
t
g
i
v
es
a
n
u
n
f
ail
in
g
cla
s
s
i
f
icatio
n
m
o
d
el
an
d
g
i
v
es
u
s
ef
u
l
in
s
i
g
h
t
o
f
t
h
e
d
ata
in
t
h
e
f
o
r
m
o
f
i
n
f
o
r
m
ati
v
e
p
er
ce
p
tio
n
.
T
h
e
p
r
o
p
o
s
ed
m
e
th
o
d
is
ca
p
ab
le
o
f
f
i
n
d
in
g
s
i
m
p
le
p
er
ce
p
tio
n
s
o
f
ca
n
ce
r
g
e
n
e
ex
p
r
es
s
io
n
d
ata
s
ets
u
t
ilizi
n
g
a
v
er
y
s
m
al
l
s
u
b
s
e
t
o
f
g
e
n
es
b
y
p
r
o
j
ec
tio
n
s
co
r
in
g
a
n
d
r
an
k
in
g
h
o
w
ev
er
p
r
ese
n
ts
a
clea
r
v
is
u
al
c
lass
if
icatio
n
b
et
w
ee
n
ca
n
ce
r
t
y
p
e
s
.
T
h
e
y
h
av
e
p
r
o
p
o
s
ed
in
v
ie
w
o
f
d
ata
v
is
u
aliza
t
io
n
’
s
p
r
o
m
is
in
g
p
ar
t
in
p
en
etr
ati
v
e
d
ata
an
al
y
s
is
,
s
h
o
r
t
r
u
n
ti
m
es
a
n
d
in
ter
ac
ti
v
e
i
n
ter
f
ac
e,
t
h
at
d
at
a
v
is
u
aliza
tio
n
w
o
u
ld
en
h
a
n
c
e
o
th
er
r
ec
o
g
n
ized
tech
n
iq
u
es
i
n
ca
n
ce
r
m
icr
o
ar
r
a
y
an
al
y
s
i
s
ass
i
s
ted
b
y
e
f
f
icie
n
t
p
r
o
j
ec
tio
n
s
ea
r
ch
m
et
h
o
d
s
an
d
b
ec
o
m
e
p
ar
t
o
f
th
e
s
tan
d
ar
d
an
al
y
s
i
s
to
o
lb
o
x
.
W
o
n
g
et
a
l.
[
2
3
]
h
av
e
p
r
o
p
o
s
ed
r
eg
u
la
tio
n
-
le
v
el
m
et
h
o
d
f
o
r
s
y
m
b
o
li
zi
n
g
t
h
e
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
E
vo
lu
tio
n
a
r
y
C
o
mp
u
ta
tio
n
a
l
A
lg
o
r
ith
m
b
y
B
len
d
in
g
o
f P
P
C
A
a
n
d
E
P
-
E
n
h
a
n
ce
d
…
(
Ma
n
a
s
w
in
i P
r
a
d
h
a
n
)
97
m
icr
o
ar
r
a
y
d
ata
o
f
ca
n
ce
r
cl
ass
i
f
icatio
n
th
at
ca
n
b
e
o
p
tim
ized
u
t
ilizi
n
g
g
e
n
etic
al
g
o
r
ith
m
s
(
G
As)
.
T
h
e
p
r
o
p
o
s
ed
s
y
m
b
o
liza
tio
n
d
ec
r
e
ases
t
h
e
d
i
m
e
n
s
io
n
a
lit
y
o
f
m
i
cr
o
ar
r
ay
d
ata
to
a
g
r
ea
ter
e
x
t
en
t
co
m
p
ar
ed
w
it
h
th
e
tr
ad
itio
n
al
ex
p
r
es
s
io
n
-
le
v
el
f
ea
tu
r
e
s
.
Sev
er
al
s
ta
tis
tica
l
m
ac
h
i
n
e
-
lear
n
i
n
g
m
e
th
o
d
s
h
av
e
b
ec
o
m
e
u
s
ab
le
an
d
e
f
f
ic
ien
t
i
n
ca
n
ce
r
cla
s
s
i
f
icatio
n
b
ec
au
s
e
n
o
is
e
an
d
v
ar
iab
ilit
y
ca
n
b
e
ac
co
m
m
o
d
ated
in
th
e
p
r
o
p
o
s
ed
s
y
m
b
o
lizatio
n
.
I
t
h
a
s
b
ee
n
co
n
f
ir
m
ed
t
h
at
th
e
t
h
r
ee
r
eg
u
lati
o
n
lev
el
r
ep
r
esen
ta
tio
n
m
o
n
o
t
o
n
icall
y
co
n
v
er
g
es
to
a
s
o
lu
tio
n
b
y
e
x
p
er
i
m
e
n
tal
r
esu
l
ts
o
n
r
ea
l
-
w
o
r
ld
m
icr
o
ar
r
ay
d
ataset
s
.
T
h
is
h
as
co
n
f
ir
m
ed
t
h
e
p
r
ese
n
ce
o
f
th
r
ee
r
eg
u
latio
n
lev
e
ls
(
u
p
-
r
e
g
u
la
tio
n
,
d
o
w
n
-
r
eg
u
lat
io
n
a
n
d
n
o
n
-
s
i
g
n
i
f
ican
t
r
eg
u
latio
n
)
ass
o
ciate
d
w
it
h
ea
c
h
p
ar
ticu
lar
b
io
lo
g
ical
p
h
en
o
t
y
p
e.
I
n
ad
d
itio
n
to
im
p
r
o
v
e
m
e
n
t
to
ca
n
ce
r
class
i
f
icat
io
n
ca
p
ab
ilit
y
,
th
e
ter
n
ar
y
r
eg
u
lat
io
n
-
lev
e
l p
r
o
m
o
te
s
th
e
v
is
u
aliza
t
io
n
o
f
m
icr
o
ar
r
ay
d
a
ta.
Ah
m
ad
M.
Sar
h
an
[
7
]
h
as
d
ev
elo
p
ed
an
A
NN
a
n
d
th
e
D
is
cr
ete
C
o
s
i
n
e
T
r
an
s
f
o
r
m
(
D
C
T
)
b
ase
d
st
o
m
ac
h
ca
n
ce
r
d
etec
tio
n
s
y
s
t
e
m
.
C
las
s
i
f
icatio
n
f
ea
tu
r
e
s
ar
e
ex
tr
ac
ted
b
y
t
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
f
r
o
m
s
to
m
ac
h
m
icr
o
ar
r
a
y
s
u
til
izin
g
DC
T
.
ANN
d
o
es t
h
e
C
las
s
i
f
icatio
n
(
tu
m
o
r
o
r
n
o
-
tu
m
o
r
)
u
p
o
n
ap
p
licatio
n
o
f
t
h
e
f
ea
t
u
r
es
ex
tr
ac
ted
f
r
o
m
t
h
e
DC
T
co
ef
f
i
cien
ts
.
I
n
h
is
s
t
u
d
y
h
e
h
as
u
s
e
d
th
e
m
icr
o
ar
r
a
y
i
m
a
g
es
t
h
at
w
er
e
o
b
tain
ed
f
r
o
m
th
e
Sta
n
f
o
r
d
Me
d
ical
Data
b
ase
(
SMD)
.
T
h
e
ab
ilit
y
o
f
th
e
p
r
o
p
o
s
ed
s
y
s
te
m
to
p
r
o
d
u
ce
v
er
y
h
i
g
h
s
u
cc
es
s
r
ate
h
as
b
ee
n
co
n
f
ir
m
ed
b
y
s
i
m
u
l
atio
n
r
esu
lts
.
P
ap
ac
h
r
is
to
u
d
is
et
a
l.
[
2
4
]
h
av
e
o
f
f
er
ed
So
Fo
C
les,
a
n
in
ter
ac
tiv
e
to
o
l
th
at
h
as
m
ad
e
s
e
m
a
n
tic
f
ea
tu
r
e
f
ilter
i
n
g
a
p
o
s
s
ib
ilit
y
in
m
icr
o
ar
r
ay
cla
s
s
i
f
icat
io
n
p
r
o
b
lem
s
b
y
th
e
u
tili
za
t
io
n
o
f
ex
ter
n
al,
u
n
a
m
b
i
g
u
o
u
s
k
n
o
w
led
g
e
ac
q
u
ir
ed
f
r
o
m
th
e
Ge
n
e
On
to
lo
g
y
.
B
y
i
m
p
r
o
v
in
g
an
in
it
iall
y
cr
ea
ted
f
ea
tu
r
e
s
et
w
i
th
t
h
e
h
elp
o
f
leg
ac
y
m
et
h
o
d
s
,
g
e
n
es
th
at
ar
e
ass
o
ciate
d
w
it
h
t
h
e
s
a
m
e
b
io
lo
g
ical
p
at
h
d
u
r
in
g
th
e
m
icr
o
ar
r
a
y
e
x
p
er
im
en
t
ar
e
ex
tr
ac
ted
b
y
t
h
e
u
tili
za
tio
n
o
f
t
h
e
id
ea
o
f
s
e
m
a
n
tic
s
i
m
ilar
it
y
.
As
o
n
e
o
f
its
m
a
n
y
f
u
n
ctio
n
s
,
So
Fo
C
le
s
o
f
f
er
s
a
h
u
g
e
r
ep
o
s
ito
r
y
o
f
s
e
m
an
tic
s
i
m
ilar
it
y
m
et
h
o
d
s
f
o
r
d
er
iv
in
g
f
ea
t
u
r
e
s
et
s
an
d
m
ar
k
er
g
e
n
es.
Di
s
cu
s
s
io
n
ab
o
u
t th
e
s
tr
u
ct
u
r
e
an
d
f
u
n
ctio
n
alit
y
o
f
t
h
e
to
o
l,
an
d
it
s
ab
ilit
y
in
i
m
p
r
o
v
i
n
g
t
h
e
class
i
f
icatio
n
ac
cu
r
ac
y
h
a
s
b
ee
n
g
i
v
en
in
d
etail.
B
y
m
ea
n
s
o
f
ex
p
er
i
m
en
ta
l
ev
al
u
a
tio
n
,
th
e
i
m
p
r
o
v
ed
class
i
f
icatio
n
ac
c
u
r
ac
y
o
f
th
e
So
Fo
C
le
s
h
as
b
ee
n
d
e
m
o
n
s
tr
ated
u
til
izin
g
d
if
f
er
en
t
s
e
m
an
tic
s
i
m
ilar
it
y
co
m
p
u
tatio
n
m
et
h
o
d
s
i
n
t
w
o
r
ea
l d
atasets
Deb
n
ath
et
a
l.
[
2
5
]
h
av
e
p
r
o
p
o
s
ed
an
ev
o
l
u
tio
n
ar
y
m
et
h
o
d
th
at
i
s
ca
p
ab
le
o
f
s
elec
ti
n
g
a
s
u
b
s
et
o
f
p
o
ten
tiall
y
i
n
f
o
r
m
ati
v
e
g
e
n
e
s
th
at
ca
n
b
e
u
s
ed
in
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
SV
M)
clas
s
i
f
ier
s
.
T
h
e
p
r
o
p
o
s
ed
ev
o
lu
tio
n
ar
y
m
et
h
o
d
esti
m
ate
s
th
e
f
it
n
ess
f
u
n
ct
io
n
u
tili
zi
n
g
SVM
an
d
a
s
p
ec
if
ied
s
u
b
s
et
o
f
g
e
n
e
f
ea
tu
r
es,
a
n
d
n
e
w
s
u
b
s
ets
o
f
f
ea
t
u
r
es
w
er
e
ch
o
s
e
n
f
o
u
n
d
ed
o
n
t
h
e
f
r
eq
u
en
c
y
o
f
o
cc
u
r
r
e
n
ce
o
f
t
h
e
f
ea
t
u
r
es
in
t
h
e
ev
o
lu
tio
n
ar
y
ap
p
r
o
ac
h
an
d
am
o
u
n
t
o
f
g
e
n
er
aliza
tio
n
er
r
o
r
in
SVMs.
He
n
ce
,
th
eo
r
etica
ll
y
,
th
e
s
elec
ted
g
e
n
es
r
ef
lect
t
h
e
g
e
n
er
aliza
tio
n
p
er
f
o
r
m
a
n
ce
o
f
S
VM
cla
s
s
i
f
ier
s
to
a
ce
r
tai
n
ex
te
n
t.
C
o
m
p
a
r
is
o
n
w
i
th
s
e
v
er
al
ex
is
t
in
g
m
et
h
o
d
s
h
a
s
co
n
f
ir
m
ed
th
at
b
etter
clas
s
i
f
icatio
n
ac
cu
r
ac
y
ca
n
b
e
ac
h
iev
ed
b
y
t
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
w
it
h
f
e
w
er
n
u
m
b
er
s
o
f
s
elec
t
ed
g
en
es.
Fro
m
t
h
e
r
ev
ie
w
,
i
t
ca
n
b
e
s
ee
n
th
a
t
m
o
s
t
o
f
th
e
r
ec
en
t
w
o
r
k
s
h
a
v
e
p
er
f
o
r
m
ed
th
e
clas
s
i
f
icatio
n
u
s
in
g
s
elec
ti
v
e
g
e
n
e
e
x
p
r
ess
io
n
d
ata.
T
h
e
s
elec
ted
g
en
e
ex
p
r
ess
io
n
s
u
b
-
d
ataset
h
as
b
ee
n
o
p
ti
m
ized
an
d
clas
s
if
ied
u
s
i
n
g
tr
ad
itio
n
al
clas
s
i
f
ier
s
.
T
h
o
u
g
h
t
h
e
o
p
ti
m
izati
o
n
is
ef
f
ec
ti
v
e
th
e
u
lti
m
ate
o
b
j
ec
tiv
e
i
s
n
o
t
att
ain
ed
b
ec
au
s
e
t
h
e
ef
f
ec
t
iv
e
n
ess
o
f
clas
s
i
f
icatio
n
is
in
ad
e
q
u
ate.
Hen
ce
,
th
e
en
h
a
n
ce
m
en
t
o
f
clas
s
i
f
ier
b
ec
o
m
e
s
a
n
e
s
s
e
n
tial
p
r
e
-
r
eq
u
i
s
it
e
f
o
r
e
f
f
ec
tiv
e
clas
s
i
f
icatio
n
o
f
m
icr
o
ar
r
ay
g
en
e
ex
p
r
ess
io
n
d
ata.
I
n
th
i
s
p
ap
er
,
w
e
p
r
o
p
o
s
e
an
ef
f
ec
t
iv
e
cla
s
s
i
f
icat
io
n
tec
h
n
iq
u
e
th
at
u
s
e
s
an
e
n
h
a
n
ce
d
s
u
p
er
v
is
ed
class
i
f
ier
.
I
t
i
s
w
e
ll
k
n
o
w
n
t
h
a
t
m
icr
o
ar
r
a
y
g
e
n
e
e
x
p
r
ess
io
n
d
atasets
ar
e
c
h
ar
ac
ter
ized
b
y
h
ig
h
d
i
m
e
n
s
io
n
a
n
d
s
m
al
l
s
a
m
p
le
s
ize.
T
h
e
d
i
m
e
n
s
io
n
o
f
t
h
e
g
en
e
e
x
p
r
ess
io
n
d
ataset
is
r
ed
u
ce
d
u
s
i
n
g
P
P
C
A
.
W
ith
th
e
aid
o
f
th
e
d
i
m
en
s
io
n
al
it
y
r
ed
u
ce
d
g
e
n
e
ex
p
r
ess
io
n
d
atase
t,
th
e
A
N
N,
w
h
ich
i
s
s
elec
ted
as
s
u
p
er
v
i
s
ed
class
if
ier
in
o
u
r
w
o
r
k
,
is
e
n
h
an
ce
d
u
s
i
n
g
E
P
te
ch
n
iq
u
e.
T
h
e
en
h
a
n
ce
d
clas
s
i
f
ier
is
u
tili
ze
d
f
o
r
clas
s
i
f
icatio
n
an
d
s
o
i
t is
tr
ain
ed
u
s
i
n
g
B
P
alg
o
r
ith
m
.
T
h
e
w
el
l
-
tr
ai
n
ed
class
i
f
ier
i
s
th
e
n
s
u
b
j
ec
ted
t
o
th
e
class
i
f
icatio
n
o
f
m
icr
o
ar
r
a
y
g
en
e
ex
p
r
ess
io
n
d
ata
s
et.
T
h
e
r
est
o
f
th
e
p
ap
er
is
o
r
g
an
ized
as
f
o
ll
o
w
s
.
Sec
tio
n
3
d
etails
t
h
e
p
r
o
p
o
s
ed
class
if
ica
tio
n
tech
n
iq
u
e
w
it
h
r
eq
u
ir
ed
m
a
th
e
m
atica
l
f
o
r
m
u
latio
n
s
an
d
illu
s
tr
atio
n
s
.
Sect
io
n
4
d
is
cu
s
s
es
ab
o
u
t
t
h
e
i
m
p
le
m
en
ta
tio
n
r
es
u
lts
a
n
d
S
e
ctio
n
5
co
n
clu
d
es t
h
e
p
ap
er
.
3.
CL
AS
SI
F
I
CAT
I
O
N
T
E
CH
NIQU
E
F
O
R
M
I
CRO
ARR
AY
G
E
NE
E
XP
R
E
SS
I
O
N
D
AT
A
Her
e,
an
ef
f
icie
n
t
tech
n
iq
u
e
t
o
class
if
y
m
icr
o
ar
r
ay
g
e
n
e
ex
p
r
ess
io
n
d
ata
is
p
r
o
p
o
s
ed
.
T
h
e
p
r
o
p
o
s
ed
tech
n
iq
u
e
is
co
m
p
r
is
ed
o
f
t
h
r
ee
f
u
n
d
a
m
e
n
tal
p
r
o
ce
s
s
e
s
,
n
a
m
el
y
,
d
i
m
en
s
io
n
ali
t
y
r
ed
u
ctio
n
,
d
ev
elo
p
m
en
t
o
f
s
u
p
er
v
i
s
ed
class
if
ier
an
d
g
e
n
e
class
i
f
icatio
n
.
T
h
e
d
ev
elo
p
m
en
t
o
f
e
n
h
an
ce
d
s
u
p
er
v
i
s
ed
cl
ass
i
f
ier
is
i
llu
s
tr
ated
in
th
e
Fig
u
r
e
1
an
d
th
e
tr
ai
n
i
n
g
p
r
o
ce
s
s
is
d
ep
icted
in
Fi
g
u
r
e
2.
T
h
e
d
im
e
n
s
io
n
alit
y
r
ed
u
c
tio
n
in
v
o
l
v
es
th
e
p
r
o
ce
s
s
o
f
r
ed
u
c
i
n
g
th
e
d
i
m
e
n
s
io
n
o
f
th
e
m
icr
o
ar
r
ay
g
e
n
e
ex
p
r
ess
io
n
d
ata
u
s
i
n
g
P
P
C
A
.
I
n
th
e
s
ec
o
n
d
p
r
o
ce
s
s
,
a
s
u
p
e
r
v
is
ed
clas
s
if
ier
is
d
ev
elo
p
ed
u
s
i
n
g
f
ee
d
f
o
r
w
ar
d
A
N
N,
w
h
ich
is
en
h
a
n
ce
d
u
s
i
n
g
E
P
tech
n
iq
u
e.
I
n
t
h
e
g
e
n
e
class
if
ica
tio
n
,
t
h
e
e
n
h
a
n
ce
d
class
i
f
ier
i
s
tr
ai
n
ed
u
s
i
n
g
th
e
g
e
n
e
ex
p
r
ess
io
n
d
ata
an
d
th
en
th
e
tes
tin
g
p
r
o
ce
s
s
is
co
n
d
u
cted
.
So
,
g
iv
e
n
a
m
icr
o
ar
r
a
y
g
en
e
ex
p
r
ess
io
n
d
ata,
th
e
c
lass
if
ie
r
ef
f
ec
tiv
e
l
y
c
lass
if
ie
s
t
h
e
d
a
ta
b
y
r
ep
r
esen
ti
n
g
t
h
e
clas
s
to
w
h
ic
h
t
h
e
d
ata
b
elo
n
g
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
7
,
No
.
2
,
J
u
n
e
201
8
:
95
–
1
0
4
98
Fig
u
r
e
1
.
E
n
h
a
n
ce
m
e
n
t o
f
Fee
d
Fo
r
w
ar
d
A
NN
u
s
i
n
g
E
P
tech
n
iq
u
e
Fig
u
r
e
2
.
T
r
ain
in
g
p
r
o
ce
s
s
o
f
en
h
a
n
ce
d
s
u
p
er
v
is
ed
class
i
f
ier
u
s
i
n
g
B
P
alg
o
r
ith
m
3
.
1
.
Dim
e
ns
io
na
lity
R
ed
uct
io
n
Usi
ng
P
P
CA
L
et,
t
h
e
m
icr
o
ar
r
ay
g
en
e
e
x
p
r
ess
io
n
d
ata
b
e
1
0
,
1
0
;
g
s
jk
N
k
N
j
M
,
w
h
er
e,
s
N
r
ep
r
esen
ts
th
e
n
u
m
b
er
o
f
s
a
m
p
les
an
d
g
N
r
ep
r
esen
ts
th
e
n
u
m
b
e
r
o
f
g
en
e
s
.
T
h
e
d
i
m
e
n
s
io
n
o
f
g
en
e
d
ata
is
h
i
g
h
er
an
d
s
o
it
is
s
u
b
j
ec
ted
to
d
im
e
n
s
io
n
alit
y
r
ed
u
ctio
n
.
I
n
d
i
m
e
n
s
io
n
alit
y
r
ed
u
ctio
n
,
th
e
h
ig
h
d
i
m
en
s
io
n
a
l
g
en
e
d
ata
jk
M
is
co
n
v
er
ted
to
a
l
o
w
d
i
m
e
n
s
io
n
a
l
g
en
e
d
ata.
T
o
r
ed
u
ce
th
e
d
i
m
e
n
s
io
n
alit
y
,
w
e
u
s
e
P
P
C
A
,
w
h
ic
h
is
a
P
C
A
w
i
th
t
h
e
p
r
esen
ce
o
f
p
r
o
b
ab
ilis
tic
m
o
d
el
f
o
r
th
e
d
ata.
T
h
e
P
P
C
A
alg
o
r
ith
m
co
m
p
o
s
ed
b
y
T
ip
p
in
g
an
d
B
is
h
o
p
[
2
6
]
is
ca
p
ab
le
o
f
ca
lcu
latin
g
a
lo
w
d
i
m
en
s
io
n
a
l
r
ep
r
esen
tatio
n
u
tili
zi
n
g
a
r
ig
h
tl
y
f
o
r
m
ed
p
r
o
b
a
b
ilit
y
d
is
tr
ib
u
t
io
n
o
f
t
h
e
h
ig
h
er
d
i
m
e
n
s
io
n
al
d
ata.
T
h
e
in
s
ti
n
cti
v
e
attr
ac
tio
n
o
f
t
h
e
p
r
o
b
ab
ilis
tic
r
ep
r
esen
tatio
n
is
b
ec
au
s
e
o
f
th
e
f
ac
t
t
h
at
t
h
e
d
ef
in
it
io
n
o
f
th
e
p
r
o
b
ab
ilis
tic
m
ea
s
u
r
e
allo
w
s
co
m
p
ar
is
o
n
w
i
t
h
o
th
er
p
r
o
b
a
b
ilis
tic
tech
n
iq
u
e
s
,
at
th
e
s
a
m
e
ti
m
e
m
a
k
i
n
g
s
tatis
t
ical
test
i
n
g
ea
s
ier
an
d
p
er
m
itti
n
g
th
e
u
tili
za
t
io
n
o
f
B
ay
esia
n
m
eth
o
d
s
.
Di
m
e
n
s
io
n
a
li
t
y
r
ed
u
ctio
n
ca
n
b
e
ac
h
iev
ed
b
y
m
a
k
i
n
g
u
s
e
o
f
P
P
C
A
a
s
a
g
e
n
er
ic
Ga
u
s
s
ian
d
en
s
it
y
m
o
d
el.
Di
m
e
n
s
io
n
alit
y
r
ed
u
ctio
n
f
ac
ilit
ate
s
ef
f
icien
t
co
m
p
u
tatio
n
o
f
t
h
e
m
ax
i
m
u
m
-
li
k
eli
h
o
o
d
esti
m
ates
f
o
r
th
e
p
ar
a
m
e
ter
s
co
n
n
ec
ted
w
it
h
th
e
co
v
ar
ia
n
ce
m
atr
i
x
f
r
o
m
th
e
d
ata
p
r
in
c
i
p
al
co
m
p
o
n
e
n
t
s
.
B
y
p
er
f
o
r
m
in
g
th
e
d
i
m
e
n
s
io
n
alit
y
r
ed
u
c
tio
n
u
s
i
n
g
P
P
C
A
,
m
icr
o
ar
r
a
y
g
en
e
ex
p
r
es
s
io
n
d
ata
o
f
d
i
m
en
s
io
n
g
s
N
N
is
r
ed
u
c
ed
to
'
'
g
s
N
N
.
T
h
e
d
i
m
en
s
io
n
al
it
y
r
ed
u
ce
d
m
atr
i
x
is
g
iv
e
n
as
M
^
.
Oth
er
th
a
n
d
i
m
en
s
io
n
a
lit
y
r
ed
u
ctio
n
,
th
e
P
P
C
A
f
in
d
s
m
o
r
e
p
r
ac
tical
ad
v
an
ta
g
es s
u
c
h
as
f
in
d
i
n
g
m
i
s
s
i
n
g
d
ata,
clas
s
i
f
icatio
n
a
n
d
n
o
v
elt
y
d
etec
tio
n
[
2
6
]
.
3
.
2
.
E
nh
a
nce
m
e
nt
o
f
F
ee
d F
o
rw
a
rd
ANNs
Her
e,
an
e
n
h
an
ce
d
s
u
p
er
v
is
ed
class
i
f
ier
u
s
in
g
m
u
lti
-
la
y
er
f
ee
d
f
o
r
w
ar
d
A
N
Ns
is
d
ev
elo
p
ed
.
T
h
e
en
h
a
n
ce
m
en
t
o
f
t
h
e
n
e
u
r
al
n
et
w
o
r
k
is
ac
co
m
p
lis
h
ed
b
y
o
p
ti
m
izin
g
t
h
e
d
i
m
e
n
s
io
n
o
f
t
h
e
h
id
d
en
la
y
er
u
s
i
n
g
E
P
tech
n
iq
u
e.
E
P
is
a
s
to
ch
asti
c
o
p
ti
m
izatio
n
s
tr
ateg
y
p
r
i
m
ar
il
y
f
o
r
m
u
lated
b
y
L
a
w
r
e
n
ce
J
.
Fo
g
el
i
n
1
9
6
0
,
w
h
ic
h
is
s
i
m
ilar
to
g
en
etic
alg
o
r
ith
m
,
b
u
t
it
s
tr
ess
e
s
o
n
th
e
b
eh
av
io
r
al
lin
k
ag
e
b
e
t
w
ee
n
p
ar
en
ts
an
d
th
eir
o
f
f
s
p
r
in
g
i
n
s
tead
o
f
att
e
m
p
tin
g
to
i
m
itate
s
p
ec
if
ic
g
en
etic
o
p
er
ato
r
s
as
s
ee
n
i
n
n
at
u
r
e.
T
h
e
E
P
tech
n
iq
u
e
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
E
vo
lu
tio
n
a
r
y
C
o
mp
u
ta
tio
n
a
l
A
lg
o
r
ith
m
b
y
B
len
d
in
g
o
f P
P
C
A
a
n
d
E
P
-
E
n
h
a
n
ce
d
…
(
Ma
n
a
s
w
in
i P
r
a
d
h
a
n
)
99
is
co
m
p
r
is
ed
o
f
(
1
)
p
o
p
u
latio
n
in
i
tializatio
n
,
(
2
)
f
itn
es
s
ca
lcu
latio
n
(
3
)
s
elec
tio
n
an
d
(
4
)
m
u
tatio
n
.
T
h
e
E
P tec
h
n
iq
u
e
u
s
ed
to
en
h
an
ce
th
e
cla
s
s
i
f
ier
is
d
is
c
u
s
s
ed
b
elo
w
.
St
ep1
:
A
p
o
p
u
latio
n
s
et
a
X
;
1
N
a
0
p
is
in
itialized
,
w
h
er
e,
a
X
is
an
ar
b
itr
ar
y
i
n
te
g
er
g
en
er
ate
d
w
it
h
i
n
t
h
e
in
ter
v
al
)
1
,
0
(
H
N
an
d
p
N
is
th
e
p
o
p
u
latio
n
s
ize.
Ste
p
2
:
p
N
n
e
u
r
al
n
et
w
o
r
k
s
,
ea
ch
w
it
h
a
n
i
n
p
u
t
la
y
er
,
a
h
id
d
en
la
y
er
a
n
d
an
o
u
tp
u
t
la
y
e
r
ar
e
d
esig
n
ed
.
I
n
ev
er
y
th
a
n
eu
r
al
n
et
w
o
r
k
,
'
s
N
(
d
im
en
s
io
n
a
lit
y
r
ed
u
ce
d
)
in
p
u
t
n
eu
r
o
n
s
a
n
d
a
b
ias
n
e
u
r
o
n
,
a
X
h
i
d
d
en
n
eu
r
o
n
s
an
d
a
b
ias n
e
u
r
o
n
a
n
d
an
o
u
tp
u
t
n
eu
r
o
n
ar
e
p
r
esen
t.
Ste
p 3
:
T
h
e
d
esig
n
ed
NN
is
weig
h
ted
an
d
b
iased
r
an
d
o
m
l
y
.
T
h
e
d
ev
elo
p
ed
NN
is
s
h
o
w
n
i
n
Fi
g
u
r
e
3.
Fig
u
r
e
3
.
T
h
e
A
NN
d
ev
elo
p
ed
w
it
h
h
id
d
en
n
e
u
r
o
n
s
t
h
at
ar
e
r
ec
o
m
m
e
n
d
ed
b
y
E
P
in
d
i
v
id
u
a
ls
Ste
p 4
:
T
h
e
b
asis
f
u
n
ctio
n
a
n
d
ac
tiv
atio
n
f
u
n
ctio
n
ar
e
s
elec
t
ed
f
o
r
th
e
d
esig
n
ed
NN
as f
o
ll
o
w
s
1
0
^
'
g
N
k
jk
jk
j
M
w
y
,
1
0
'
s
N
j
(
1
)
y
e
y
g
1
1
)
(
(2
)
y
y
g
)
(
(
3
)
E
q
u
atio
n
1
is
th
e
b
asis
f
u
n
ct
io
n
(
g
iv
e
n
o
n
l
y
f
o
r
in
p
u
t
la
y
er
)
,
E
q
u
atio
n
2
an
d
E
q
u
atio
n
.
3
r
e
p
r
esen
ts
th
e
s
ig
m
o
id
a
n
d
id
en
ti
t
y
ac
ti
v
atio
n
f
u
n
ctio
n
,
w
h
ic
h
i
s
s
elec
ted
f
o
r
h
id
d
en
la
y
er
a
n
d
o
u
t
p
u
t
la
y
er
r
esp
ec
ti
v
el
y
.
I
n
E
q
u
atio
n
1
M
^
is
th
e
d
i
m
e
n
s
io
n
ali
t
y
r
ed
u
ce
d
m
icr
o
ar
r
ay
g
en
e
d
a
ta,
jk
w
is
th
e
w
ei
g
h
t
o
f
th
e
n
eu
r
o
n
s
an
d
is
th
e
b
ia
s
.
T
h
e
b
asis
f
u
n
ctio
n
g
i
v
en
i
n
E
q
u
at
io
n
1
is
co
m
m
o
n
l
y
u
s
ed
i
n
all
t
h
e
r
e
m
ain
in
g
la
y
er
s
(
h
id
d
en
an
d
o
u
tp
u
t
la
y
er
,
b
u
t
w
it
h
t
h
e
n
u
m
b
er
o
f
h
id
d
en
a
n
d
o
u
tp
u
t
n
e
u
r
o
n
s
,
r
esp
ec
ti
v
el
y
)
.
T
h
e
M
^
is
g
i
v
en
to
th
e
in
p
u
t
la
y
er
o
f
th
e
p
N
ANNs a
n
d
th
e
o
u
tp
u
t f
r
o
m
t
h
e
all
t
h
o
s
e
A
NN
s
ar
e
d
eter
m
i
n
ed
.
Ste
p 5
:
T
h
e
lear
n
in
g
er
r
o
r
is
d
eter
m
i
n
ed
f
o
r
all
th
e
p
N
n
et
w
o
r
k
s
as f
o
llo
w
s
1
0
'
'
1
s
N
b
ab
s
a
Y
D
N
E
(
4
)
w
h
er
e,
a
E
is
th
e
er
r
o
r
in
th
e
th
a
NN,
D
is
th
e
d
esire
d
o
u
tp
u
t a
n
d
ab
Y
is
t
h
e
ac
tu
al
o
u
tp
u
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
7
,
No
.
2
,
J
u
n
e
201
8
:
95
–
1
0
4
100
Ste
p
6
:
Fit
n
ess
is
d
eter
m
in
ed
f
o
r
ev
er
y
in
d
i
v
id
u
al,
w
h
ic
h
is
p
r
esen
t
in
th
e
p
o
p
u
latio
n
p
o
o
l,
u
s
i
n
g
t
h
e
f
it
n
es
s
f
u
n
ctio
n
as
f
o
llo
w
s
1
0
1
p
N
a
a
a
a
E
E
F
(
5
)
Ste
p
7
:
T
h
e
i
n
d
iv
id
u
als
w
h
ic
h
h
av
e
m
ax
i
m
u
m
f
it
n
es
s
ar
e
s
elec
ted
f
o
r
th
e
ev
o
l
u
tio
n
ar
y
p
r
o
ce
s
s
,
m
u
tatio
n
.
So
,
2
p
N
in
d
iv
id
u
als ar
e
s
elec
ted
f
r
o
m
t
h
e
p
o
p
u
latio
n
p
o
o
l a
n
d
s
u
b
j
ec
ted
to
m
u
tatio
n
.
Ste
p
8
:
I
n
m
u
tatio
n
,
n
e
w
2
p
N
in
d
iv
id
u
als
ne
w
X
ar
e
g
en
er
ated
to
f
ill
t
h
e
p
o
p
u
latio
n
p
o
o
l
an
d
th
e
g
en
er
atio
n
is
g
iv
e
n
as
f
o
llo
w
s
o
t
h
e
r
w
i
s
e
;
2
;
2
;
2
1
d
P
d
d
P
d
d
ne
w
M
N
M
if
M
N
M
if
M
X
(
6
)
I
n
E
q
u
atio
n
6
,
th
e
m
u
tatio
n
s
et
d
M
is
d
eter
m
in
ed
as
l
be
s
t
in
d
N
-
M
M
,
w
h
er
e
,
}
,
,
3
,
2
,
1
{
in
M
;
is
t
h
e
m
ed
ia
n
o
f
l
b
es
t
N
an
d
l
b
es
t
N
is
a
s
et
o
f
b
est
in
d
i
v
id
u
al
s
th
at
h
a
s
m
ax
i
m
u
m
f
itn
e
s
s
1
d
M
is
d
eter
m
i
n
ed
as
'
d
d
M
M
,
w
h
er
e,
'
d
M
is
a
s
et
o
f
r
an
d
o
m
in
te
g
er
s
t
h
at
ar
e
g
en
er
ated
w
it
h
in
t
h
e
in
ter
v
al
)
1
,
(
H
N
.
T
h
e
s
et
'
d
M
is
g
en
er
ated
in
s
u
c
h
a
w
a
y
t
h
at
it
s
ati
s
f
ie
s
th
e
f
o
llo
w
in
g
co
n
d
itio
n
s
(
i)
d
P
d
M
N
M
2
'
(
7
)
(
ii)
l
be
s
t
d
N
M
'
(
8
)
I
n
E
q
u
atio
n
6
,
2
d
M
is
t
h
e
s
et
o
f
r
an
d
o
m
ele
m
e
n
ts
w
h
ic
h
ar
e
tak
e
n
f
r
o
m
th
e
s
et
d
M
s
u
ch
t
h
at.
2
2
P
d
N
M
an
d
in
d
M
M
2
.
Ste
p
9
:
T
h
e
n
e
w
l
y
o
b
tai
n
ed
in
d
iv
id
u
als
ne
w
X
o
cc
u
p
y
t
h
e
p
o
p
u
l
atio
n
p
o
o
l
an
d
s
o
th
e
p
o
o
l
r
et
ain
s
it
s
s
ize
p
N
.
T
h
en
,
NNs
ar
e
d
ev
elo
p
ed
as
p
er
th
e
i
n
d
i
v
id
u
al
s
p
r
ese
n
t
i
n
th
e
n
e
w
p
o
p
u
latio
n
p
o
o
l
an
d
th
e
p
r
o
ce
s
s
is
iter
ativ
el
y
r
ep
ea
ted
u
n
t
il
it
r
ea
ch
es
t
h
e
m
a
x
i
m
u
m
n
u
m
b
er
o
f
iter
atio
n
1
m
a
x
I
.
On
ce
,
th
e
p
r
o
ce
s
s
is
co
m
p
leted
,
t
h
e
b
est in
d
i
v
id
u
al
is
o
b
tain
ed
f
r
o
m
t
h
e
p
o
p
u
lati
o
n
p
o
o
l b
ased
o
n
th
e
f
i
tn
e
s
s
v
alu
e.
Ste
p
1
0
:
T
h
e
o
b
tain
ed
b
est
i
n
d
iv
id
u
al
is
s
to
r
ed
an
d
th
e
p
r
o
ce
s
s
is
ag
ai
n
r
ep
ea
ted
f
r
o
m
s
tep
1
f
o
r
2
m
a
x
I
iter
atio
n
s
.
I
n
ea
ch
iter
atio
n
,
a
b
est
in
d
iv
id
u
al
is
o
b
tain
ed
an
d
s
o
2
m
a
x
I
b
est
in
d
iv
id
u
als
(
t
h
e
b
e
s
t
in
d
iv
id
u
al
r
ep
r
esen
ts
n
u
m
b
er
o
f
h
id
d
en
u
n
i
ts
,
w
h
ic
h
is
ter
m
ed
as
b
es
t
H
)
ar
e
o
b
tain
ed
af
ter
co
m
p
letio
n
o
f
all
t
h
e
iter
atio
n
s
.
Am
o
n
g
t
h
e
2
m
a
x
I
iter
atio
n
s
,
t
h
e
b
es
t
in
d
iv
id
u
al
w
h
ic
h
h
a
s
m
ax
i
m
u
m
f
r
eq
u
e
n
c
y
i.e
.
th
e
i
n
d
i
v
id
u
al,
w
h
ic
h
is
s
elec
ted
as
b
est
f
o
r
th
e
m
o
s
t
n
u
m
b
er
o
f
ti
m
es
i
s
s
e
lecte
d
as
th
e
f
i
n
al
b
est
i
n
d
iv
id
u
al.
T
h
u
s
o
b
tain
ed
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
E
vo
lu
tio
n
a
r
y
C
o
mp
u
ta
tio
n
a
l
A
lg
o
r
ith
m
b
y
B
len
d
in
g
o
f P
P
C
A
a
n
d
E
P
-
E
n
h
a
n
ce
d
…
(
Ma
n
a
s
w
in
i P
r
a
d
h
a
n
)
101
b
est
in
d
iv
id
u
al
is
s
ele
c
ted
as
th
e
d
i
m
e
n
s
io
n
o
f
th
e
h
id
d
en
la
y
er
an
d
s
o
th
e
NN
i
s
d
esig
n
ed
.
Hen
ce
,
an
en
h
an
ce
d
NN
i
s
d
ev
elo
p
ed
b
y
o
p
ti
m
izi
n
g
t
h
e
d
i
m
e
n
s
io
n
o
f
th
e
h
id
d
e
n
la
y
er
u
s
in
g
t
h
e
E
P
tech
n
iq
u
e.
3
.
3
.
Cla
s
s
if
ica
t
io
n o
f
M
icro
a
rr
a
y
G
ene
E
x
pres
s
io
n us
ing
t
he
E
nh
a
nce
d
C
la
s
s
if
ier
I
n
t
h
e
cla
s
s
i
f
icatio
n
o
f
m
icr
o
ar
r
ay
g
e
n
e
e
x
p
r
ess
io
n
d
ata,
t
wo
p
h
ases
o
f
o
p
er
atio
n
ar
e
p
er
f
o
r
m
ed
t
h
a
t
in
cl
u
d
e
tr
ain
in
g
p
h
ase
an
d
tes
tin
g
p
h
ase.
I
n
th
e
tr
ain
i
n
g
p
h
ase,
th
e
en
h
a
n
ce
d
s
u
p
er
v
is
ed
class
i
f
ier
is
tr
ain
ed
u
s
i
n
g
t
h
e
B
P
alg
o
r
ith
m
.
T
h
e
d
i
m
en
s
io
n
al
it
y
r
ed
u
ce
d
m
icr
o
ar
r
ay
g
e
n
e
e
x
p
r
ess
io
n
d
atas
et
is
u
tili
ze
d
to
tr
ai
n
th
e
NN.
3
.
3
.
1
.
T
ra
ini
ng
P
ha
s
e:
M
ini
m
i
za
t
io
n o
f
E
rr
o
r
by
B
P
a
lg
o
rit
h
m
T
h
e
tr
ain
in
g
p
h
ase
o
f
t
h
e
NN
u
s
i
n
g
B
P
alg
o
r
ith
m
i
s
d
is
cu
s
s
e
d
b
elo
w
.
1.
T
h
e
w
ei
g
h
ts
ar
e
r
a
n
d
o
m
l
y
g
e
n
er
ated
w
it
h
in
th
e
i
n
ter
v
al
1
,
0
a
n
d
ass
ig
n
ed
to
t
h
e
h
id
d
en
la
y
er
as
w
ell
as o
u
tp
u
t la
y
er
.
Fo
r
in
p
u
t la
y
e
r
,
th
e
w
ei
g
h
ts
m
ai
n
tai
n
a
co
n
s
t
an
t v
al
u
e
o
f
u
n
it
y
.
2.
T
h
e
tr
ain
in
g
g
e
n
e
d
ata
s
eq
u
en
ce
is
g
iv
e
n
to
t
h
e
NN
s
o
th
at
t
h
e
B
P
e
r
r
o
r
is
d
eter
m
i
n
ed
u
s
i
n
g
th
e
E
q
u
a
tio
n
4
.
T
h
e
b
asis
f
u
n
ct
io
n
an
d
tr
a
n
s
f
er
f
u
n
ctio
n
ar
e
s
i
m
ilar
to
t
h
at
u
s
ed
i
n
t
h
e
o
p
tim
izatio
n
(
g
i
v
e
n
in
E
q
u
atio
n
1
, Eq
u
atio
n
2
an
d
E
q
u
atio
n
3
.
3.
W
h
en
t
h
e
B
P
er
r
o
r
is
ca
lcu
late
d
,
th
e
w
ei
g
h
ts
o
f
all
th
e
n
eu
r
o
n
s
ar
e
ad
j
u
s
ted
as f
o
llo
w
s
jk
jk
jk
w
w
w
(
9
)
I
n
E
q
u
atio
n
3
,
jk
w
is
th
e
ch
an
g
e
i
n
w
e
ig
h
t
w
h
ich
ca
n
b
e
d
eter
m
in
ed
as
E
y
w
jk
jk
.
.
,
w
h
er
e,
E
is
th
e
B
P
er
r
o
r
an
d
is
th
e
lear
n
i
n
g
r
ate,
u
s
u
all
y
it r
an
g
es
f
r
o
m
0
.
2
to
0
.
5
.
4.
On
ce
th
e
w
ei
g
h
ts
ar
e
ad
j
u
s
ted
as
p
er
th
e
E
q
u
atio
n
9
,
th
e
p
r
o
ce
s
s
is
r
ep
ea
ted
f
r
o
m
s
tep
2
u
n
til
th
e
B
P
er
r
o
r
g
ets
m
in
i
m
ized
to
a
least v
alu
e.
P
r
ac
ticall
y
,
t
h
e
cr
iter
io
n
to
b
e
s
atis
f
ied
is
1
.
0
E
.
T
h
e
B
P
alg
o
r
ith
m
is
ter
m
i
n
ate
d
w
h
e
n
t
h
e
er
r
o
r
g
ets
m
i
n
i
m
iz
ed
to
a
m
in
i
m
u
m
v
al
u
e,
w
h
ich
co
n
s
tr
u
e
s
th
at
t
h
e
d
esig
n
ed
A
N
N
is
w
ell
tr
ai
n
ed
f
o
r
its
f
u
r
t
h
er
test
i
n
g
p
h
ase.
3
.
3
.
2
.
T
esting
P
ha
s
e
:
Cla
s
s
if
i
ca
t
io
n
o
f
G
iv
en
M
icro
a
rr
a
y
G
ene
S
e
qu
ence
I
n
t
h
e
tr
ai
n
i
n
g
p
h
ase,
th
e
ANN
lear
n
s
w
el
l
ab
o
u
t
th
e
tr
ai
n
in
g
g
e
n
e
s
eq
u
e
n
ce
a
n
d
th
e
cl
ass
u
n
d
er
w
h
ic
h
it
is
p
r
esen
t.
T
h
e
w
e
ll
-
t
r
ain
ed
A
NN
ca
n
cla
s
s
i
f
y
t
h
e
m
icr
o
ar
r
a
y
g
en
e
s
eq
u
en
ce
i
n
an
ef
f
ec
ti
v
e
m
a
n
n
er
.
Giv
e
n
a
test
s
eq
u
e
n
ce
,
th
e
d
i
m
en
s
io
n
al
it
y
r
ed
u
ctio
n
is
p
er
f
o
r
m
ed
u
s
i
n
g
t
h
e
P
P
C
A
.
T
h
e
d
im
e
n
s
io
n
a
lit
y
r
ed
u
ce
d
g
en
e
s
eq
u
en
ce
is
g
iv
en
as
i
n
p
u
t
to
t
h
e
w
ell
-
tr
ai
n
e
d
en
h
a
n
ce
d
s
u
p
er
v
is
ed
clas
s
i
f
ier
.
T
h
e
class
i
f
ier
ef
f
ec
tiv
e
l
y
c
lass
if
ies
t
h
e
g
e
n
e
s
eq
u
e
n
ce
b
y
d
eter
m
i
n
in
g
th
e
cla
s
s
to
w
h
ic
h
it
b
elo
n
g
s
.
T
h
e
s
u
p
er
v
is
ed
class
i
f
ier
is
d
esig
n
ed
w
i
th
th
e
in
ten
tio
n
o
f
clas
s
if
y
i
n
g
th
e
m
icr
o
ar
r
a
y
g
e
n
e
s
eq
u
en
ce
an
d
h
er
eb
y
,
it is
ac
co
m
p
lis
h
ed
w
ell.
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
p
r
o
p
o
s
ed
class
if
icat
io
n
te
ch
n
iq
u
e
is
i
m
p
le
m
e
n
ted
i
n
th
e
MA
T
L
A
B
p
lat
f
o
r
m
(
v
er
s
io
n
7
.
8
)
an
d
it
is
e
v
alu
a
ted
u
s
in
g
th
e
m
icr
o
ar
r
ay
g
e
n
e
e
x
p
r
ess
io
n
d
ata
o
f
h
u
m
a
n
ac
u
te
le
u
k
e
m
i
as.
T
h
e
s
tan
d
ar
d
leu
k
e
m
ia
d
ataset
f
o
r
tr
ain
in
g
an
d
test
in
g
i
s
o
b
tain
ed
f
r
o
m
[
2
7
]
.
T
h
e
tr
ain
in
g
le
u
k
e
m
ia
d
ataset
is
o
f
d
i
m
en
s
io
n
7192
g
N
an
d
38
s
N
.
T
h
is
h
ig
h
d
i
m
e
n
s
io
n
al
tr
ain
i
n
g
d
ataset
is
s
u
b
j
e
cted
to
d
im
en
s
io
n
ali
t
y
r
ed
u
ctio
n
u
s
in
g
P
P
C
A
an
d
s
o
a
d
ataset
o
f
d
im
en
s
io
n
30
g
N
an
d
38
s
N
is
o
b
tain
ed
.
I
n
d
ev
elo
p
in
g
th
e
en
h
an
ce
d
s
u
p
er
v
i
s
ed
class
if
ier
,
t
h
e
d
i
m
en
s
io
n
a
lit
y
r
ed
u
ce
d
m
icr
o
ar
r
a
y
g
en
e
d
atase
t
is
u
s
ed
to
f
in
d
th
e
o
p
tim
a
l
d
i
m
en
s
io
n
o
f
t
h
e
h
i
d
d
en
la
y
er
.
T
h
e
e
n
h
a
n
ce
m
en
t
o
f
t
h
e
ANN
i
s
p
er
f
o
r
m
ed
w
it
h
th
e
p
ar
a
m
e
tr
ic
v
alu
e
s
g
i
v
e
n
in
t
h
e
T
ab
le
1
.
I
n
ea
ch
iter
atio
n
,
th
e
er
r
o
r
g
ets
m
i
n
i
m
ized
an
d
th
e
o
p
tim
al
v
al
u
e
f
o
r
th
e
d
i
m
en
s
io
n
o
f
t
h
e
h
id
d
en
la
y
er
is
f
o
u
n
d
.
W
h
ile
en
h
a
n
ci
n
g
ANN,
th
e
er
r
o
r
,
w
h
ic
h
is
d
eter
m
i
n
ed
f
o
r
d
if
f
er
e
n
t
iter
atio
n
s
an
d
th
e
ca
lcu
lated
f
it
n
es
s
,
w
h
ile
en
h
a
n
ci
n
g
AN
N,
ar
e
d
ep
icted
in
t
h
e
F
ig
u
r
e
4
.
T
h
e
tr
ain
i
n
g
o
f
en
h
a
n
ce
d
A
NN
cla
s
s
i
f
ier
i
s
i
m
p
le
m
e
n
ted
u
s
i
n
g
th
e
Neu
r
al
Net
w
o
r
k
T
o
o
lb
o
x
in
M
A
T
L
AB
.
T
h
e
er
r
o
r
v
er
s
u
s
ep
o
ch
s
,
w
h
ic
h
is
o
b
tain
ed
i
n
t
h
e
tr
ain
i
n
g
o
f
A
NN
u
s
i
n
g
B
P
,
is
ill
u
s
tr
ated
in
Fig
u
r
e
5.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
7
,
No
.
2
,
J
u
n
e
201
8
:
95
–
1
0
4
102
T
ab
le
1
.
E
P
p
a
r
am
eter
s
u
s
ed
i
n
th
e
e
n
h
a
n
ce
m
e
n
t o
f
A
NN
S
.
N
o
EP Par
a
me
t
e
r
s
V
a
l
u
e
s
1
p
N
10
2
H
N
20
3
D
0
.
2
5
(
f
o
r
A
LL
)
0
.
7
5
(
f
o
r
A
M
L
)
4
1
m
a
x
I
50
5
2
m
a
x
I
1
0
0
(
a)
(
b
)
Fig
u
r
e
4
.
E
P
p
e
r
f
o
r
m
a
n
ce
i
n
en
h
a
n
ce
m
en
t o
f
A
NN
cla
s
s
i
f
ier
: (
a)
E
r
r
o
r
v
er
s
u
s
N
u
m
b
er
o
f
it
er
atio
n
s
an
d
(
b
)
Fit
n
e
s
s
v
er
s
u
s
n
u
m
b
er
o
f
iter
a
tio
n
s
.
On
ce
t
h
e
en
h
a
n
ce
d
s
u
p
er
v
i
s
ed
class
i
f
ier
is
d
ev
elo
p
ed
an
d
tr
ai
n
ed
w
ell,
th
e
clas
s
i
f
i
ca
tio
n
i
s
p
er
f
o
r
m
ed
b
y
p
r
o
v
id
in
g
th
e
m
icr
o
ar
r
ay
g
en
e
ex
p
r
es
s
io
n
test
d
ataset.
T
h
e
clas
s
i
f
ier
d
etec
ts
th
e
t
y
p
e
o
f
ca
n
ce
r
f
r
o
m
th
e
d
atase
t
w
ith
a
g
o
o
d
ac
cu
r
ac
y
.
T
h
e
s
ig
n
i
f
ican
ce
o
f
th
e
en
h
an
ce
d
A
NN
cla
s
s
i
f
ier
is
d
e
m
o
n
s
tr
ated
b
y
co
m
p
ar
i
n
g
it
s
clas
s
if
icatio
n
p
e
r
f
o
r
m
an
ce
w
i
th
t
h
at
o
f
t
h
e
ex
i
s
tin
g
A
NN
cla
s
s
i
f
ier
.
T
h
e
co
m
p
ar
i
s
o
n
r
es
u
lts
ar
e
p
r
o
v
id
ed
in
th
e
T
ab
le
2
.
M
o
r
e
o
v
er
,
th
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
c
lass
i
f
ier
is
al
s
o
co
m
p
ar
ed
w
it
h
th
e
ex
i
s
ti
n
g
SVM
class
i
f
ier
an
d
th
e
r
esu
lts
ar
e
g
i
v
en
i
n
th
e
T
ab
le
3
.
T
h
e
co
m
p
ar
is
o
n
r
esu
lt
s
g
i
v
e
n
in
th
e
T
ab
le
d
em
o
n
s
tr
at
e
th
a
t
th
e
cla
s
s
i
f
icatio
n
ac
c
u
r
ac
y
o
f
th
e
e
n
h
an
ce
d
c
lass
if
ier
w
it
h
o
p
ti
m
ized
h
id
d
en
la
y
er
d
i
m
e
n
s
io
n
is
g
o
o
d
,
an
d
9
0
%
m
o
r
e
t
h
an
th
a
t
o
f
t
h
e
ANN
clas
s
i
f
ier
w
it
h
ar
b
itra
r
y
h
i
d
d
en
la
y
er
d
i
m
e
n
s
io
n
.
Fro
m
T
ab
le
2
an
d
T
a
b
le
3
r
esu
lt
s
,
it
ca
n
b
e
s
ee
n
th
at
t
h
e
p
r
o
p
o
s
ed
tech
n
iq
u
e
h
as
g
o
o
d
class
i
f
icatio
n
ac
c
u
r
ac
y
an
d
le
s
s
er
r
o
r
r
ate
w
h
e
n
co
m
p
ar
ed
w
it
h
t
h
e
SVM
cla
s
s
if
ier
.
T
h
e
r
esu
lt
s
s
h
o
w
t
h
at
t
h
e
en
h
a
n
ce
d
s
u
p
er
v
is
ed
cla
s
s
i
f
ie
r
p
er
f
o
r
m
s
w
ell
i
n
class
i
f
y
in
g
t
h
e
m
icr
o
ar
r
ay
g
en
e
ex
p
r
ess
io
n
d
ataset.
Fig
u
r
e
5
.
P
er
f
o
r
m
a
n
c
e
o
f
B
P
in
tr
ain
i
n
g
th
e
e
n
h
an
ce
d
A
NN
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
E
vo
lu
tio
n
a
r
y
C
o
mp
u
ta
tio
n
a
l
A
lg
o
r
ith
m
b
y
B
len
d
in
g
o
f P
P
C
A
a
n
d
E
P
-
E
n
h
a
n
ce
d
…
(
Ma
n
a
s
w
in
i P
r
a
d
h
a
n
)
103
T
ab
le
2
.
C
o
m
p
ar
is
o
n
b
et
w
ee
n
en
h
a
n
ce
d
A
NN
clas
s
i
f
ier
an
d
ex
is
t
in
g
A
NN
cla
s
s
i
f
ier
C
a
n
c
e
r
c
l
a
ss
En
h
a
n
c
e
d
A
N
N
c
l
a
ssi
f
i
e
r
H
b
e
s
t
=4
Ex
i
st
i
n
g
A
N
N
c
l
a
ssi
f
i
e
r
C
l
a
ssi
f
i
c
a
t
i
o
n
a
c
c
u
r
a
c
y
(
i
n
%)
Er
r
o
r
r
a
t
e
(
i
n
%)
H
=
2
H
=
6
H
=
8
C
l
a
ssi
f
i
c
a
t
i
o
n
a
c
c
u
r
a
c
y
(
i
n
%)
Er
r
o
r
r
a
t
e
(
i
n
%)
C
l
a
ssi
f
i
c
a
t
i
o
n
a
c
c
u
r
a
c
y
(
i
n
%)
Er
r
o
r
r
a
t
e
(
i
n
%)
C
l
a
ssi
f
i
c
a
t
i
o
n
a
c
c
u
r
a
c
y
(
i
n
%)
Er
r
o
r
r
a
t
e
(
i
n
%)
A
LL
9
2
.
5
9
2
6
7
.
4
0
7
4
7
0
.
3
7
0
4
2
9
.
6
2
9
6
5
5
.
5
5
5
6
4
4
.
4
4
4
4
8
8
.
8
8
8
9
1
1
.
1
1
1
1
A
M
L
9
0
.
9
0
9
1
9
.
0
9
0
9
8
3
.
1
8
4
5
1
6
.
8
1
5
5
6
3
.
7
3
1
4
3
6
.
2
6
8
6
8
1
.
9
9
8
3
1
8
.
0
0
1
7
T
ab
le
3
.
C
o
m
p
ar
is
o
n
b
et
w
ee
n
en
h
a
n
ce
d
A
NN
clas
s
i
f
ier
an
d
ex
is
t
in
g
SVM
cla
s
s
i
f
ier
C
a
n
c
e
r
c
l
a
ss
En
h
a
n
c
e
d
A
N
N
c
l
a
ssi
f
i
e
r
Ex
i
st
i
n
g
S
V
M
c
l
a
ss
i
f
i
e
r
C
l
a
ssi
f
i
c
a
t
i
o
n
a
c
c
u
r
a
c
y
(
i
n
%)
Er
r
o
r
r
a
t
e
(
i
n
%)
C
l
a
ssi
f
i
c
a
t
i
o
n
a
c
c
u
r
a
c
y
(
i
n
%)
Er
r
o
r
r
a
t
e
(
i
n
%)
A
LL
9
2
.
5
9
2
6
7
.
4
0
7
4
7
0
.
3
7
0
4
2
9
.
6
2
9
6
A
M
L
9
0
.
9
0
9
1
9
.
0
9
0
9
7
2
.
7
2
7
3
2
7
.
2
7
2
7
5.
CO
NCLU
SI
O
N
I
n
th
is
p
ap
er
,
w
e
h
a
v
e
p
r
o
p
o
s
ed
an
e
f
f
icie
n
t
cla
s
s
i
f
icatio
n
tech
n
iq
u
e
w
it
h
an
en
h
a
n
ce
d
s
u
p
er
v
i
s
ed
class
i
f
ier
u
s
i
n
g
A
N
N.
T
h
e
p
r
o
p
o
s
ed
tech
n
iq
u
e
h
a
s
b
ee
n
d
e
m
o
n
s
tr
ated
b
y
p
er
f
o
r
m
i
n
g
t
h
e
cla
s
s
i
f
icatio
n
o
f
A
M
L
a
n
d
AL
L
ca
n
ce
r
s
.
T
h
e
i
m
p
le
m
e
n
tat
io
n
r
es
u
lts
h
a
v
e
s
h
o
w
n
th
at
t
h
e
clas
s
i
f
icatio
n
o
f
t
h
e
ca
n
ce
r
is
p
er
f
o
r
m
ed
w
it
h
g
o
o
d
class
if
ic
atio
n
r
ate.
T
h
e
b
etter
clas
s
if
ic
atio
n
p
er
f
o
r
m
an
ce
is
ac
h
iev
ed
m
ai
n
l
y
b
ec
a
u
s
e
o
f
th
e
e
n
h
a
n
ce
m
en
t
o
f
t
h
e
A
N
N.
T
h
e
en
h
a
n
ce
m
e
n
t
is
p
er
f
o
r
m
ed
w
it
h
t
h
e
i
n
ten
tio
n
o
f
f
in
d
i
n
g
t
h
e
d
i
m
e
n
s
io
n
o
f
th
e
h
id
d
en
la
y
er
s
u
c
h
t
h
at
th
e
er
r
o
r
is
m
i
n
i
m
ized
.
U
s
in
g
th
e
E
P
,
an
o
p
ti
m
al
d
i
m
e
n
s
io
n
f
o
r
h
id
d
en
la
y
er
h
a
s
b
ee
n
id
en
ti
f
ied
.
T
h
e
tr
ai
n
i
n
g
o
f
A
N
N
u
s
i
n
g
B
P
h
as
r
ed
u
ce
d
th
e
B
P
er
r
o
r
to
a
co
n
s
id
er
ab
le
a
m
o
u
n
t.
T
h
e
co
m
p
ar
is
o
n
r
es
u
lt
s
f
o
r
ex
i
s
ti
n
g
A
NN
clas
s
i
f
ier
a
n
d
SVM
clas
s
i
f
ier
h
as
d
em
o
n
s
tr
ated
t
h
at
t
h
e
class
i
f
icatio
n
ac
cu
r
ac
y
is
m
o
r
e
in
th
e
en
h
an
ce
d
A
NN
clas
s
i
f
ier
r
ath
er
th
a
n
th
e
o
th
er
class
if
ier
.
Hen
ce
,
it
ca
n
b
e
co
n
clu
d
ed
th
at
t
h
e
p
r
o
p
o
s
e
d
class
if
icatio
n
tec
h
n
iq
u
e
is
m
o
r
e
e
f
f
ec
ti
v
e
in
cla
s
s
i
f
y
in
g
t
h
e
m
icr
o
a
r
r
a
y
g
en
e
ex
p
r
ess
io
n
d
ata
f
o
r
ca
n
ce
r
s
w
i
th
r
e
m
ar
k
ab
le
class
i
f
icat
io
n
ac
cu
r
ac
y
.
RE
F
E
R
E
NC
E
S
[1
]
.
V
a
id
y
a
n
a
th
a
n
a
n
d
By
u
n
g
-
Ju
n
Yo
o
n
,
"
T
h
e
ro
le
o
f
sig
n
a
l
p
ro
c
e
ss
in
g
c
o
n
c
e
p
ts
in
g
e
n
o
m
ic
s
a
n
d
p
ro
teo
m
ic
s"
,
J
o
u
rn
a
l
o
f
t
h
e
Fra
n
k
li
n
I
n
stit
u
te
,
V
o
l
.
3
4
1
,
No
.
2
,
p
p
.
1
1
1
-
1
3
5
,
M
a
rc
h
2
0
0
4
.
[2
]
.
A
n
ib
a
l
Ro
d
rig
u
e
z
F
u
e
n
tes
,
Ju
a
n
V
.
L
o
re
n
z
o
G
in
o
ri
a
n
d
Rica
rd
o
G
ra
u
A
b
a
lo
,
“
A
Ne
w
P
re
d
icto
r
o
f
Co
d
i
n
g
Re
g
io
n
s
in
G
e
n
o
m
i
c
S
e
q
u
e
n
c
e
s u
sin
g
a
C
o
m
b
in
a
ti
o
n
o
f
Di
ff
e
re
n
t
A
p
p
ro
a
c
h
e
s”
,
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
Bi
o
lo
g
ic
a
l
a
n
d
L
if
e
S
c
ien
c
e
s
,
Vo
l.
3
,
N
o
.
2
,
p
p
.
1
0
6
-
1
1
0
,
2
0
0
7
.
[3
]
.
Yin
g
X
u
,
Vic
to
r
Olm
a
n
a
n
d
Do
n
g
X
u
,
"
M
in
im
u
m
S
p
a
n
n
i
n
g
T
re
e
s
f
o
r
G
e
n
e
Ex
p
re
ss
io
n
Da
ta
Clu
ste
rin
g
"
,
Ge
n
o
me
In
fo
rm
a
t
ics
,
Vo
l.
1
2
,
p
p
.
2
4
–
3
3
,
2
0
0
1
.
[4
]
.
A
n
a
n
d
h
a
v
a
ll
i
G
a
u
th
a
m
a
n
,
"
A
n
a
ly
sis
o
f
DN
A
M
icro
a
rra
y
Da
t
a
u
sin
g
A
ss
o
c
iatio
n
Ru
les
:
A
S
e
lec
ti
v
e
S
t
u
d
y
"
,
W
o
rld
Aca
d
e
my
o
f
S
c
ien
c
e
,
En
g
i
n
e
e
rin
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
Vo
l.
4
2
,
p
p
.
1
2
-
1
6
,
2
0
0
8
.
[5
]
.
Ch
in
tan
u
K.
S
a
rm
a
h
,
S
a
n
d
h
y
a
S
a
m
a
ra
sin
g
h
e
,
Do
n
Ku
las
iri
a
n
d
Da
n
iel
Ca
tch
p
o
o
le,
"
A
S
i
m
p
le
Affy
m
e
tri
x
Ra
ti
o
-
tran
sf
o
r
m
a
ti
o
n
M
e
t
h
o
d
Yie
ld
s
Co
m
p
a
ra
b
le
Ex
p
re
ss
io
n
L
e
v
e
l
Qu
a
n
ti
f
ica
ti
o
n
s
w
it
h
CDN
A
Da
ta"
,
W
o
rld
Aca
d
e
my
o
f
S
c
ien
c
e
,
En
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
Vo
l.
6
1
,
p
p
.
7
8
-
8
3
,
2
0
1
0
.
[6
]
.
Kh
lo
p
o
v
a
,
G
laz
k
o
a
n
d
G
l
a
z
k
o
,
“
Diff
e
r
e
n
ti
a
ti
o
n
o
f
G
e
n
e
E
x
p
re
ss
i
o
n
P
r
o
f
il
e
s
Da
ta
f
o
r
L
i
v
e
r
a
n
d
K
id
n
e
y
o
f
P
ig
s”
,
W
o
rld
Aca
d
e
my
o
f
S
c
ien
c
e
,
En
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
lo
g
y
,
V
o
l
.
5
5
,
p
p
.
2
6
7
-
2
7
0
,
2
0
0
9
.
[7
]
.
A
h
m
a
d
M
.
S
a
rh
a
n
,
"
Ca
n
c
e
r
Cl
a
ss
if
i
c
a
ti
o
n
Ba
se
d
on
M
icro
a
rra
y
G
e
n
e
E
x
p
re
ss
io
n
Da
ta
U
sin
g
d
c
t
a
n
d
A
n
n
"
,
J
o
u
rn
a
l
o
f
T
h
e
o
re
ti
c
a
l
a
n
d
A
p
p
l
ied
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
,
Vo
l.
6
,
No
.
2
,
p
p
.
2
0
7
-
2
1
6
,
2
0
0
9
.
[8
]
.
Hu
il
in
X
i
o
n
g
,
Ya
Z
h
a
n
g
a
n
d
X
ue
-
W
e
n
Ch
e
n
,
"
Da
ta
-
De
p
e
n
d
e
n
t
Ke
rn
e
l
M
a
c
h
i
n
e
s
f
o
r
M
i
c
ro
a
rra
y
Da
ta
Clas
sif
ic
a
ti
o
n
"
,
IEE
E/
ACM
T
ra
n
sa
c
ti
o
n
s
o
n
Co
m
p
u
t
a
ti
o
n
a
l
B
io
l
o
g
y
a
n
d
Bi
o
in
fo
rm
a
t
ics
(
T
CBB
)
,
Vo
l.
4
,
N
o
.
4
,
p
p
.
5
8
3
-
5
9
5
,
Oc
to
b
e
r
2
0
0
7
.
[9
]
.
Ja
v
ier
He
rre
ro
,
Ju
a
n
M
.
V
a
q
u
e
ri
z
a
s,
F
a
ti
m
a
A
l
-
S
h
a
h
ro
u
r,
L
u
c
ıa
Co
n
d
e
,
A
lv
a
ro
M
a
teo
s,
Ja
v
ier
S
a
n
to
y
o
Ra
m
o
n
Dıa
z
-
Uria
rte
a
n
d
Jo
a
q
u
ın
Do
p
a
z
o
,
"
Ne
w
Ch
a
ll
e
n
g
e
s
in
Ge
n
e
Ex
p
re
ss
io
n
Da
ta
A
n
a
l
y
sis
a
n
d
th
e
Ex
ten
d
e
d
G
EP
A
S
"
,
Nu
c
leic
Acid
s R
e
se
a
rc
h
,
Vo
l.
3
2
,
p
p
.
4
8
5
–
4
9
1
,
2
0
0
4
.
[1
0
]
.
S
v
e
ta
Ka
b
a
n
o
v
a
,
P
e
tra
Kle
in
b
o
n
g
a
rd
,
Je
n
s
V
o
lk
m
e
r
,
Birg
it
A
n
d
ré
e
,
M
a
lt
e
Ke
l
m
a
n
d
T
h
o
m
a
s
W
.
Ja
x
,
"
Ge
n
e
Ex
p
re
ss
io
n
A
n
a
ly
sis
o
f
Hu
m
a
n
Re
d
Blo
o
d
Ce
ll
s"
,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
M
e
d
ica
l
S
c
ien
c
e
s
,
V
o
l.
6
,
No
.
4
,
p
p
.
1
5
6
-
1
5
9
,
2
0
0
9
.
[1
1
]
.
S
u
n
g
w
o
o
Kw
o
n
a
n
d
Ch
o
n
g
h
u
n
Ha
n
,
"
H
y
b
rid
Clu
ste
ri
n
g
M
e
th
o
d
f
o
r
DN
A
M
icro
a
rra
y
D
a
ta
A
n
a
ly
sis"
,
Ge
n
o
me
In
fo
rm
a
t
ics
,
Vo
l.
1
3
,
p
p
,
2
5
8
-
2
5
9
,
2
0
0
2
.
[1
2
]
.
S
u
n
g
w
o
o
Kw
o
n
,
y
o
u
n
g
-
Hw
a
n
C
h
u
,
He
u
i
-
S
e
o
k
y
i
a
n
d
Ch
o
n
g
h
u
n
Ha
n
,
"
DN
A
M
icro
a
rra
y
Da
ta
A
n
a
ly
sis
f
o
r
Ca
n
c
e
r
Clas
sif
ic
a
ti
o
n
Ba
se
d
o
n
S
tep
w
ise
Disc
ri
m
in
a
n
t
A
n
a
ly
sis
a
n
d
Ba
y
e
sia
n
De
c
isio
n
T
h
e
o
ry
"
,
Ge
n
o
me
In
fo
rm
a
t
ics
,
V
o
l
.
1
2
,
p
p
.
2
5
2
-
2
5
4
,
2
0
0
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
7
,
No
.
2
,
J
u
n
e
201
8
:
95
–
1
0
4
104
[1
3
]
.
S
u
p
o
j
He
n
g
p
ra
p
ro
h
m
a
n
d
P
ra
b
h
a
s
Ch
o
n
g
stit
v
a
tan
a
,
"
F
e
a
tu
re
S
e
l
e
c
ti
o
n
By
Weig
h
ted
-
S
n
r
f
o
r
Ca
n
c
e
r
M
icro
a
rra
y
Da
ta
Clas
sif
i
c
a
ti
o
n
"
,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
In
n
o
v
a
t
ive
Co
mp
u
t
in
g
,
In
fo
rm
a
t
io
n
a
n
d
Co
n
tro
l
,
Vo
l.
4
,
No
.
1
2
(
A
),
p
p
.
4
6
2
7
-
4
6
3
5
,
De
c
e
m
b
e
r
2
0
0
9
.
[1
4
]
.
A
n
a
sta
s
sio
u
,
"
G
e
n
o
m
i
c
S
ig
n
a
l
P
r
o
c
e
ss
in
g
,
"
IEE
E
S
i
g
n
a
l
Pro
c
e
ss
in
g
M
a
g
a
zin
e
,
Vo
l.
1
8
,
p
p
.
8
-
2
0
,
2
0
0
1
.
[1
5
]
.
Ho
S
u
n
S
h
o
n
,
S
u
n
sh
in
Kim
,
Ch
u
n
g
S
e
i
Rh
e
e
a
n
d
Ke
u
m
Ho
R
y
u
,
"
Clu
ste
rin
g
A
p
p
ro
a
c
h
Us
in
g
M
c
l
A
l
g
o
rit
h
m
f
o
r
A
n
a
l
y
z
in
g
M
icro
a
rra
y
Da
ta"
,
In
te
rn
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Bi
o
e
lec
tro
ma
g
n
e
t
ism
,
Vo
l.
9
,
N
o
.
2
,
P
p
.
6
5
-
6
6
,
2
0
0
7
.
[1
6
]
.
Hie
u
T
ru
n
g
Hu
y
n
h
,
Ju
n
g
-
Ja
Kim
a
n
d
Yo
n
g
g
w
a
n
W
o
n
,
"
Clas
sif
ic
a
ti
o
n
S
tu
d
y
o
n
DN
A
M
icro
a
rra
y
w
it
h
F
e
e
d
f
o
rwa
rd
Ne
u
ra
l
Ne
t
w
o
rk
T
r
a
in
e
d
b
y
S
in
g
u
lar
V
a
lu
e
De
c
o
m
p
o
siti
o
n
"
,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Bi
o
-
S
c
ien
c
e
a
n
d
b
io
-
T
e
c
h
n
o
lo
g
y
,
V
o
l.
1
,
n
o
.
1
,
p
p
.
1
7
-
2
4
,
De
c
e
m
b
e
r
2
0
0
9
.
[1
7
]
.
Ch
a
n
g
ji
n
g
S
h
a
n
g
a
n
d
Qia
n
g
S
h
e
n
,
"
A
id
in
g
Clas
sif
ic
a
ti
o
n
o
f
Ge
n
e
Ex
p
re
ss
io
n
Da
ta
w
it
h
F
e
a
tu
re
S
e
lec
ti
o
n
:
A
Co
m
p
a
ra
ti
v
e
S
tu
d
y
"
,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
c
o
mp
u
ta
ti
o
n
a
l
In
tell
ig
e
n
c
e
Res
e
a
rc
h
,
V
o
l
.
1
,
No
.
1
,
p
p
.
6
8
-
7
6
,
2
0
0
5
[1
8
]
.
Ch
e
n
-
Hs
in
C
h
e
n
,
He
n
ry
Ho
rn
g
-
S
h
in
g
L
u
,
Ch
e
n
-
T
u
o
L
iao
,
Ch
u
n
-
h
o
u
h
C
h
e
n
,
Ue
n
g
-
Ch
e
n
g
Ya
n
g
a
n
d
Yu
n
-
S
h
ie
n
L
e
e
,
"
Ge
n
e
Ex
p
re
ss
io
n
A
n
a
l
y
sis
Re
f
in
in
g
S
y
st
e
m
(
G
E
A
RS
)
v
ia
S
tatisti
c
a
l
A
p
p
ro
a
c
h
:
A
P
re
li
m
in
a
ry
R
e
p
o
rt"
,
Ge
n
o
me
In
f
o
rm
a
ti
c
s
,
V
o
l
.
1
4
,
p
p
.
3
1
6
-
3
1
7
,
2
0
0
3
.
[1
9
]
.
Zh
e
n
q
iu
L
iu
,
De
c
h
a
n
g
Ch
e
n
a
n
d
Ha
li
m
a
Be
n
s
m
a
il
,
"
Ge
n
e
Ex
p
re
ss
io
n
Da
ta
Clas
sif
ica
ti
o
n
w
it
h
K
e
rn
e
l
P
ri
n
c
ip
a
l
Co
m
p
o
n
e
n
t
A
n
a
ly
sis
"
,
J
Bi
o
me
d
Bi
o
tec
h
n
o
l
,
Vo
l.
2
0
0
5
,
No
.
2
,
p
p
.
1
5
5
–
1
5
9
,
2
0
0
5
.
[2
0
]
.
Ro
b
e
rto
Ru
iz,
Jo
se
C.
Riq
u
e
lm
e
a
n
d
Je
su
s
S
.
A
g
u
il
a
r
-
Ru
iz,
"
In
c
re
m
e
n
tal
W
ra
p
p
e
r
-
Ba
se
d
Ge
n
e
S
e
lec
ti
o
n
f
ro
m
M
icro
a
rra
y
D
a
ta f
o
r
Ca
n
c
e
r
Clas
s
if
ica
ti
o
n
,
"
Pa
tt
e
rn
Rec
o
g
n
i
ti
o
n
,
Vo
l.
3
9
,
No
.
1
2
,
p
p
.
2
3
8
3
-
2
3
9
2
,
2
0
0
6
.
[2
1
]
.
Ya
n
x
io
n
g
P
e
n
g
,
W
e
n
y
u
a
n
L
i
a
n
d
Yin
g
L
iu
,
"
A
H
y
b
rid
A
p
p
ro
a
c
h
f
o
r
Bio
m
a
rk
e
r
Disc
o
v
e
r
y
f
ro
m
M
icro
a
rra
y
G
e
n
e
Ex
p
re
ss
io
n
Da
ta f
o
r
Ca
n
c
e
r
Clas
sif
ica
ti
o
n
”
,
Ca
n
c
e
r In
f
o
rm
.
,
Vo
l.
2
,
p
p
.
3
0
1
-
3
1
1
,
F
e
b
ru
a
ry
2
0
0
7
.
[2
2
]
.
M
in
c
a
M
ra
m
o
r,
G
re
g
o
r
L
e
b
a
n
,
Ja
n
e
z
De
m
a
r
a
n
d
Bla
Zu
p
a
n
,
"
V
isu
a
li
z
a
ti
o
n
-
Ba
se
d
Ca
n
c
e
r
M
icro
a
rra
y
Da
ta
Clas
sif
ic
a
ti
o
n
A
n
a
l
y
sis"
,
Bi
o
in
fo
r
ma
ti
c
s
,
V
o
l.
2
3
,
N
o
.
1
6
,
p
p
.
2
1
4
7
-
2
1
5
4
,
J
u
n
e
2
0
0
7
.
[2
3
]
.
Ha
u
-
S
a
n
W
o
n
g
a
n
d
Ho
n
g
-
Qia
n
g
W
a
n
g
,
"
Co
n
stru
c
ti
n
g
t
h
e
G
e
n
e
Re
g
u
latio
n
-
L
e
v
e
l
Re
p
re
se
n
tatio
n
o
f
M
icro
a
rra
y
Da
ta f
o
r
Ca
n
c
e
r
Clas
si
f
ica
ti
o
n
"
,
J
o
u
rn
a
l
o
f
Bi
o
me
d
ica
l
In
f
o
rm
a
ti
c
s
,
Vo
l.
4
1
,
N
o
.
1
,
p
p
.
9
5
-
1
0
5
,
F
e
b
ru
a
ry
2
0
0
8
.
[2
4
]
.
G
e
o
rg
io
s
P
a
p
a
c
h
risto
u
d
is,
S
o
ti
ri
s
Dip
laris
a
n
d
P
e
ricle
s
A
.
M
it
k
a
s,
"
S
o
F
o
Cles
:
F
e
a
tu
re
f
il
terin
g
f
o
r
m
icro
a
rra
y
c
las
si
f
ica
ti
o
n
b
a
se
d
o
n
G
e
n
e
On
t
o
lo
g
y
"
,
J
o
u
rn
a
l
o
f
Bi
o
me
d
ica
l
In
f
o
rm
a
ti
c
s
,
V
o
l
.
4
3
,
N
o
.
1
,
p
p
.
1
-
1
4
,
F
e
b
ru
a
ry
2
0
1
0
.
[2
5
]
.
Ra
m
e
s
wa
r
De
b
n
a
th
a
n
d
T
a
k
io
Ku
rit
a
,
"
A
n
e
v
o
lu
ti
o
n
a
ry
a
p
p
ro
a
c
h
f
o
r
g
e
n
e
se
lec
ti
o
n
a
n
d
c
las
si
f
ica
ti
o
n
o
f
m
icro
a
rra
y
d
a
ta b
a
se
d
o
n
S
VM
e
r
ro
r
-
b
o
u
n
d
t
h
e
o
ries
"
,
Bi
o
S
y
ste
ms
,
V
o
l
.
1
0
0
,
No
.
1
,
p
p
.
3
9
-
4
6
,
A
p
ril
2
0
1
0
.
[2
6
]
.
M
.
E.
T
ip
p
in
g
a
n
d
C.
M
.
Bish
o
p
,
“
P
ro
b
a
b
il
ist
ic
P
ri
n
c
ip
a
l
Co
m
p
o
n
e
n
t
A
n
a
l
y
sis”
,
J
o
u
rn
a
l
o
f
th
e
R
o
y
a
l
S
t
a
ti
stica
l
S
o
c
iety
,
S
e
rie
s B
,
Vo
l.
2
1
,
No
.
3
,
p
.
p
.
6
1
1
–
6
2
2
,
1
9
9
9
.
[2
7
]
.
ALL
/
A
M
L
d
a
tas
e
ts
f
ro
m
h
tt
p
:/
/w
ww
.
b
ro
a
d
in
stit
u
te.o
rg
/ca
n
c
e
r/so
f
tw
a
r
e
/g
e
n
e
p
a
tt
e
rn
/d
a
tas
e
ts/
Evaluation Warning : The document was created with Spire.PDF for Python.