I
nd
o
ne
s
ia
n
J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
3
7
,
No
.
1
,
J
an
u
ar
y
20
2
5
,
p
p
.
58
0
~
5
89
I
SS
N:
2
502
-
4
7
52
,
DOI
: 1
0
.
1
1
5
9
1
/ijee
cs
.v
3
7
.
i
1
.
pp
58
0
-
5
8
9
580
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs
.
ia
esco
r
e.
co
m
EM
SPLA f
o
r
a
cc
ura
te
f
ea
ture
mo
lecula
r e
x
tract
io
n
from
pro
tein
-
lig
a
nd i
nt
eractio
ns
Srini
d
hi K
ulk
a
rni V
1
,2
,
G
a
nes
h Dha
n
da
pa
ni
2
,
K
uree
ck
a
l V
.
Ra
m
esh
3
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
J
y
o
t
h
y
I
n
st
i
t
u
t
e
o
f
T
e
c
h
n
o
l
o
g
y
,
B
a
n
g
a
l
o
r
e
,
K
a
r
n
a
t
a
k
a
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
Ja
i
n
(
D
e
e
m
e
d
-
to
-
b
e
U
n
i
v
e
r
s
i
t
y
)
,
K
a
r
n
a
t
a
k
a
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
B
i
o
t
e
c
h
n
o
l
o
g
y
,
Ja
i
n
(
D
e
e
me
d
-
to
-
b
e
U
n
i
v
e
r
si
t
y
)
,
K
a
r
n
a
t
a
k
a
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
May
31
,
2
0
2
4
R
ev
is
ed
Sep
11
,
2
0
2
4
Acc
ep
ted
Sep
29
,
2
0
2
4
P
ro
tein
-
li
g
a
n
d
i
n
tera
c
ti
o
n
s
a
re
fu
n
d
a
m
e
n
tal
in
v
a
rio
u
s
b
i
o
l
o
g
ica
l
a
n
d
m
e
d
ica
l
field
s,
i
n
flu
e
n
c
i
n
g
d
r
u
g
d
isc
o
v
e
ry
a
n
d
th
e
ra
p
e
u
ti
c
d
e
v
e
lo
p
m
e
n
t.
I
n
re
c
e
n
t
y
e
a
rs,
d
e
e
p
lea
rn
in
g
(D
L)
h
a
s
re
v
o
l
u
ti
o
n
ize
d
th
e
st
u
d
y
o
f
t
h
e
se
in
tera
c
ti
o
n
s,
b
u
t
si
g
n
if
ica
n
t
c
h
a
l
len
g
e
s
re
m
a
in
in
a
c
c
u
ra
tely
re
p
re
se
n
ti
n
g
m
o
lec
u
lar
stru
c
tu
re
s
fo
r
DL
m
o
d
e
ls.
Trad
it
i
o
n
a
l
fe
a
tu
riza
ti
o
n
tec
h
n
iq
u
e
s
o
ften
d
e
p
e
n
d
o
n
h
a
n
d
c
ra
fted
f
e
a
tu
re
s,
re
q
u
iri
n
g
e
x
p
e
rt
k
n
o
wl
e
d
g
e
a
n
d
p
o
ten
ti
a
ll
y
m
issi
n
g
c
ru
c
ial
m
o
lec
u
lar
a
sp
e
c
ts.
Th
is
wo
r
k
a
d
d
re
ss
e
s
th
e
se
c
h
a
ll
e
n
g
e
s
b
y
d
e
v
e
lo
p
in
g
a
n
d
e
v
a
lu
a
ti
n
g
a
n
o
v
e
l
p
r
o
tein
-
li
g
a
n
d
fe
a
tu
re
e
x
trac
ti
o
n
sy
ste
m
u
si
n
g
a
n
e
n
h
a
n
c
e
d
m
o
lec
u
lar
sim
il
a
rit
y
p
r
o
t
e
in
-
li
g
a
n
d
a
li
g
n
e
r
(EM
S
P
LA).
T
h
e
p
rima
ry
o
b
jec
ti
v
e
is
to
lev
e
ra
g
e
E
M
S
P
LA
fo
r
sim
il
a
rit
y
m
a
tch
in
g
i
n
p
ro
tei
n
-
li
g
a
n
d
i
n
tera
c
ti
o
n
s,
i
m
p
ro
v
i
n
g
p
re
d
ictiv
e
m
o
d
e
l
a
c
c
u
ra
c
y
.
Th
e
m
e
th
o
d
o
lo
g
y
c
o
m
b
in
e
s
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
two
rk
s
(CNN
)
fo
r
l
o
c
a
l
fe
a
tu
re
e
x
trac
ti
o
n
with
a
n
a
tt
e
n
ti
o
n
m
o
d
u
le
to
c
a
p
tu
re
lo
n
g
-
d
istan
c
e
d
e
p
e
n
d
e
n
c
ies
,
e
n
h
a
n
c
i
n
g
b
in
d
in
g
site
p
re
d
icti
o
n
s.
Us
in
g
t
h
e
P
DBb
in
d
v
.
2
0
2
0
d
a
tas
e
t,
th
e
EM
S
P
LA
m
o
d
e
l
d
e
m
o
n
stra
ted
su
p
e
rio
r
p
e
rfo
rm
a
n
c
e
with
a
r
o
o
t
m
e
a
n
sq
u
a
re
e
rro
r
(RM
S
E)
o
f
0
.
6
7
,
s
u
rp
a
ss
in
g
c
u
rre
n
t
sta
te
-
of
-
th
e
-
a
rt
m
o
d
e
ls.
Th
e
se
fi
n
d
in
g
s
h
ig
h
li
g
h
t
th
e
sy
ste
m
’
s
p
o
ten
ti
a
l
f
o
r
e
fficie
n
t
d
e
p
l
o
y
m
e
n
t
a
n
d
sc
a
lab
il
it
y
,
p
o
siti
o
n
in
g
it
a
s
a
p
o
we
rfu
l
t
o
o
l
in
c
o
m
p
u
tati
o
n
a
l
b
io
l
o
g
y
a
n
d
d
ru
g
d
isc
o
v
e
ry
,
u
lt
ima
tely
a
d
v
a
n
c
in
g
o
u
r
u
n
d
e
rsta
n
d
i
n
g
o
f
p
ro
tein
-
li
g
a
n
d
in
tera
c
ti
o
n
s.
K
ey
w
o
r
d
s
:
Alig
n
er
C
NN
Dee
p
lear
n
in
g
PDB
b
in
d
Pro
tein
-
L
ig
an
d
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Srin
id
h
i K
u
lk
ar
n
i
V
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
,
J
ain
(
Dee
m
ed
-
to
-
b
e
Un
iv
er
s
ity
)
B
an
g
alo
r
e,
Kar
n
atak
a
,
I
n
d
ia
E
m
ail: su
d
h
asu
r
en
0
5
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
A
p
r
o
tein
-
lig
a
n
d
i
n
ter
ac
tio
n
i
n
v
o
lv
es
th
e
b
in
d
in
g
o
f
a
lig
a
n
d
,
wh
ich
is
ty
p
ically
a
s
m
all
m
o
lecu
le,
to
a
s
p
ec
if
ic
s
ite
o
n
a
p
r
o
tein
[
1
]
.
Pro
tein
s
ar
e
lar
g
e
,
co
m
p
lex
m
o
lecu
les
th
at
p
lay
m
a
n
y
cr
itical
r
o
les
in
th
e
b
o
d
y
,
in
clu
d
i
n
g
ac
tin
g
as
en
zy
m
es,
s
ig
n
alin
g
m
o
lecu
les,
an
d
s
tr
u
ctu
r
al
c
o
m
p
o
n
en
ts
[
2
]
.
L
ig
an
d
s
ca
n
b
e
v
ar
io
u
s
s
u
b
s
tan
ce
s
,
in
clu
d
in
g
d
r
u
g
s
,
h
o
r
m
o
n
es,
o
r
n
eu
r
o
tr
a
n
s
m
itter
s
[
3
]
.
T
h
e
in
ter
ac
tio
n
b
etwe
en
a
p
r
o
tein
an
d
its
lig
an
d
is
f
u
n
d
am
en
tal
t
o
m
an
y
b
io
lo
g
ical
p
r
o
ce
s
s
es,
as
it
ca
n
alter
th
e
p
r
o
tein
’
s
s
tr
u
ctu
r
e
an
d
f
u
n
ctio
n
,
o
f
ten
tr
ig
g
er
in
g
a
b
io
lo
g
ical
r
esp
o
n
s
e
[
4
]
.
Pro
tein
-
lig
an
d
in
ter
ac
t
io
n
s
ar
e
c
r
u
cial
in
v
ar
i
o
u
s
b
io
lo
g
ical
an
d
m
ed
ical
f
ield
s
.
T
h
ey
ca
n
r
e
g
u
late
p
h
y
s
io
lo
g
ical
p
r
o
ce
s
s
es
b
y
ac
tiv
atin
g
o
r
in
h
ib
itin
g
t
h
e
p
r
o
tein
’
s
f
u
n
ctio
n
[
5
]
.
Fo
r
i
n
s
tan
ce
,
in
en
zy
m
at
ic
r
ea
ctio
n
s
,
a
s
u
b
s
tr
ate
(
lig
a
n
d
)
b
in
d
s
to
an
en
z
y
m
e
(
p
r
o
tein
)
to
ca
taly
ze
a
b
io
ch
em
ical
r
ea
ctio
n
[
6
]
.
I
n
t
h
e
p
h
ar
m
ac
eu
tical
in
d
u
s
tr
y
,
u
n
d
er
s
tan
d
in
g
th
ese
i
n
ter
ac
tio
n
s
is
v
ital
f
o
r
d
r
u
g
d
is
co
v
er
y
an
d
d
ev
elo
p
m
en
t.
Dr
u
g
s
ar
e
d
esig
n
ed
to
tar
g
et
s
p
ec
if
ic
p
r
o
tein
s
,
m
o
d
u
latin
g
t
h
eir
ac
tiv
ity
to
tr
ea
t
d
is
ea
s
es
[
7
]
.
B
y
b
in
d
in
g
to
th
eir
tar
g
et
p
r
o
tein
s
,
th
ese
d
r
u
g
s
ca
n
in
h
ib
it
o
r
s
tim
u
late
th
e
p
r
o
tein
’
s
f
u
n
ctio
n
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
MS
P
LA
fo
r
a
cc
u
r
a
te
fea
tu
r
e
mo
lecu
la
r
ex
tr
a
ctio
n
fr
o
m
p
r
o
tein
-
lig
a
n
d
i
n
tera
ctio
n
s
(
S
r
in
i
d
h
i Ku
lka
r
n
i V
)
581
th
er
eb
y
alter
in
g
th
e
d
is
ea
s
e
p
r
o
ce
s
s
.
T
h
e
s
p
ec
if
icity
an
d
af
f
in
ity
o
f
th
e
lig
an
d
f
o
r
its
p
r
o
tein
tar
g
et
ar
e
k
ey
f
ac
to
r
s
in
d
eter
m
i
n
in
g
th
e
ef
f
i
ca
cy
an
d
s
af
ety
o
f
a
d
r
u
g
[
8
]
.
I
n
r
ec
e
n
t
y
ea
r
s
,
d
ee
p
lear
n
in
g
(
DL
)
h
as
r
ev
o
l
u
tio
n
ized
t
h
e
s
tu
d
y
o
f
p
r
o
tein
-
lig
an
d
in
ter
a
ctio
n
s
[
9
]
.
DL
,
a
s
u
b
s
et
o
f
m
ac
h
in
e
lear
n
in
g
(
ML
)
,
u
s
es
n
eu
r
al
n
etw
o
r
k
s
with
m
an
y
lay
er
s
(
h
en
c
e
“
d
ee
p
”
)
to
m
o
d
e
l
co
m
p
lex
p
atter
n
s
in
d
ata.
I
n
th
e
p
r
o
tein
-
li
g
an
d
i
n
ter
ac
tio
n
s
,
DL
m
o
d
els
ca
n
p
r
ed
ict
b
in
d
in
g
af
f
in
ities
,
id
en
tif
y
p
o
ten
tial
b
in
d
i
n
g
s
ites
,
an
d
e
v
en
d
esig
n
n
o
v
el
lig
a
n
d
s
[
1
0
]
.
T
h
ese
m
o
d
els
ar
e
tr
ain
ed
o
n
lar
g
e
d
atasets
o
f
k
n
o
wn
p
r
o
tein
-
lig
a
n
d
co
m
p
l
ex
es,
lear
n
in
g
to
r
ec
o
g
n
ize
s
u
b
tle
p
atter
n
s
th
at
g
o
v
e
r
n
b
i
n
d
in
g
in
te
r
ac
tio
n
s
.
Fo
r
ex
am
p
le,
co
n
v
o
lu
tio
n
al
-
n
eu
r
al
-
n
etwo
r
k
s
(
C
NNs)
[
1
1
]
h
av
e
b
ee
n
u
s
ed
to
an
aly
z
e
3
D
s
tr
u
ctu
r
es
o
f
p
r
o
tein
-
lig
a
n
d
co
m
p
lex
es,
ca
p
tu
r
in
g
s
p
atial
f
ea
tu
r
es
th
at
a
r
e
cr
itical
f
o
r
b
i
n
d
in
g
.
R
ec
u
r
r
e
n
t
-
n
eu
r
al
-
n
etwo
r
k
s
(
R
NNs)
[
1
2
]
,
an
o
th
er
ty
p
e
o
f
DL
ar
c
h
i
tectu
r
e,
h
av
e
b
e
en
ap
p
lied
to
s
eq
u
e
n
tial
d
at
a,
s
u
ch
as
p
r
o
tein
s
eq
u
en
ce
s
,
to
p
r
ed
ict
th
eir
in
t
er
ac
tio
n
s
with
lig
an
d
s
.
On
e
s
ig
n
if
ican
t
c
h
allen
g
e
i
n
lev
e
r
ag
in
g
DL
f
o
r
p
r
o
tein
-
lig
an
d
i
n
ter
ac
tio
n
s
is
th
e
n
ee
d
to
ac
cu
r
ately
r
e
p
r
esen
t
m
o
l
ec
u
lar
s
tr
u
ctu
r
es
in
a
f
o
r
m
th
at
th
e
m
o
d
els
ca
n
p
r
o
ce
s
s
,
th
is
is
wh
e
r
e
f
ea
tu
r
i
za
tio
n
co
m
es
in
[
1
3
]
.
A
f
ea
t
u
r
izer
c
o
n
v
e
r
ts
th
e
co
m
p
le
x
,
m
u
ltid
im
en
s
io
n
al
in
f
o
r
m
atio
n
o
f
m
o
lecu
lar
s
tr
u
ctu
r
es
in
to
f
ea
tu
r
e
v
ec
to
r
s
,
wh
ich
ar
e
n
u
m
e
r
ical
r
e
p
r
esen
tatio
n
s
th
at
ca
p
tu
r
e
th
e
ess
en
tial
ch
ar
ac
t
er
is
tics
o
f
th
e
m
o
lecu
les
[
1
4
]
.
Desig
n
in
g
a
n
ef
f
ec
tiv
e
f
ea
tu
r
ize
r
is
cr
itical
b
ec
au
s
e
th
e
q
u
ality
o
f
th
e
f
ea
tu
r
e
v
ec
to
r
s
d
ir
ec
tl
y
im
p
ac
ts
th
e
p
e
r
f
o
r
m
an
ce
o
f
th
e
DL
m
o
d
els.
T
h
ese
f
ea
t
u
r
es
m
ig
h
t
in
cl
u
d
e
in
f
o
r
m
atio
n
ab
o
u
t
t
h
e
m
o
lecu
lar
g
eo
m
etr
y
,
elec
tr
o
n
ic
p
r
o
p
er
ties
,
an
d
p
h
y
s
ico
c
h
em
ical
c
h
ar
ac
ter
is
tics
o
f
th
e
lig
an
d
an
d
p
r
o
tein
.
T
r
ad
itio
n
al
f
ea
tu
r
izatio
n
tec
h
n
iq
u
es
o
f
ten
r
ely
o
n
h
a
n
d
cr
af
ted
f
ea
tu
r
es,
wh
ic
h
r
e
q
u
ir
e
ex
p
e
r
t
k
n
o
wled
g
e
an
d
m
ay
n
o
t
ca
p
tu
r
e
all
r
elev
an
t
asp
ec
ts
o
f
th
e
m
o
lecu
les
[
1
5
]
.
Ho
wev
e
r
,
DL
m
o
d
els
ca
n
b
e
u
s
ed
to
lear
n
f
ea
tu
r
es
d
i
r
ec
tly
f
r
o
m
r
aw
d
ata,
a
p
r
o
ce
s
s
k
n
o
wn
as
d
ee
p
r
ep
r
esen
tatio
n
lear
n
in
g
[
1
6
]
.
T
h
is
ap
p
r
o
ac
h
ca
n
u
n
co
v
e
r
m
o
r
e
in
tr
icate
a
n
d
in
f
o
r
m
ativ
e
r
ep
r
esen
tatio
n
s
th
at
m
ig
h
t
b
e
m
is
s
ed
b
y
m
an
u
al
f
ea
tu
r
izatio
n
.
As
a
r
esu
lt,
d
esig
n
in
g
a
r
o
b
u
s
t
f
ea
tu
r
izer
o
r
lev
e
r
ag
in
g
DL
to
au
t
o
m
atica
lly
ex
tr
ac
t
f
ea
t
u
r
es
is
ess
en
tial
f
o
r
u
n
d
er
s
tan
d
i
n
g
o
f
p
r
o
tei
n
-
lig
a
n
d
in
te
r
ac
tio
n
s
an
d
im
p
r
o
v
in
g
th
e
ac
cu
r
ac
y
o
f
p
r
e
d
ictiv
e
m
o
d
els
in
d
r
u
g
d
is
co
v
er
y
a
n
d
o
t
h
er
ap
p
licatio
n
s
.
T
h
e
co
n
t
r
ib
u
tio
n
o
f
t
h
is
wo
r
k
is
as
:
−
Desig
n
in
g
a
DL
f
ea
tu
r
izer
m
o
d
el
f
o
r
e
x
tr
ac
tio
n
o
f
m
o
lecu
lar
tar
g
et
in
t
o
f
ea
t
u
r
e
v
ec
t
o
r
.
−
Pro
p
o
s
in
g
an
e
n
h
an
ce
d
m
o
l
ec
u
lar
s
im
ilar
ity
p
r
o
tein
-
lig
a
n
d
alig
n
er
(
E
MSPLA
)
f
o
r
en
h
an
cin
g
th
e
s
im
ilar
ity
m
atch
in
g
.
−
Usi
n
g
C
NN
to
ex
tr
ac
t lo
ca
l f
e
atu
r
es f
r
o
m
n
eig
h
b
o
r
in
g
r
esid
u
es f
r
o
m
th
e
p
r
o
tei
n
s
eq
u
en
ce
s
.
−
C
o
m
p
ar
ed
th
e
r
o
o
t m
ea
n
s
q
u
a
r
e
er
r
o
r
(
R
MSE
)
s
co
r
e
with
cu
r
r
en
t state
-
of
-
ar
t w
o
r
k
s
.
T
h
is
wo
r
k
is
o
r
g
an
ized
in
th
e
f
o
llo
win
g
m
a
n
n
er
.
I
n
s
ec
tio
n
2
,
ex
is
tin
g
DL
m
o
d
els
o
f
p
r
o
t
ein
-
lig
an
d
wo
r
k
s
ar
e
d
is
cu
s
s
ed
.
I
n
s
ec
ti
o
n
3
,
t
h
e
p
r
o
p
o
s
ed
DL
a
p
p
r
o
ac
h
is
d
is
cu
s
s
ed
.
I
n
s
ec
tio
n
4
,
th
e
r
esu
lts
f
o
r
p
r
o
p
o
s
ed
E
MSPLA
ar
e
d
is
cu
s
s
ed
.
Fin
ally
,
in
s
ec
tio
n
5
,
th
e
c
o
n
clu
s
io
n
o
f
th
e
wo
r
k
is
g
iv
en
.
2.
L
I
T
E
R
AT
U
RE
SU
RVE
Y
I
n
th
is
s
ec
tio
n
,
v
ar
i
o
u
s
p
r
o
tein
-
lig
an
d
a
p
p
r
o
ac
h
es
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
a
r
e
d
is
cu
s
s
ed
.
W
an
g
et
a
l.
[
1
7
]
,
p
r
esen
ted
a
m
o
d
el
ca
lle
d
m
u
lti
-
ch
a
n
n
el
s
u
b
-
s
tr
u
ctu
r
e
g
r
ap
h
-
g
ated
r
ec
u
r
r
en
t
-
u
n
it
(
MCS
G
-
G
R
U)
wh
ich
co
n
s
is
ted
o
f
v
ar
io
u
s
n
eu
r
al
n
et
wo
r
k
s
h
av
i
n
g
d
if
f
er
e
n
t
atten
ti
o
n
m
o
d
u
les
wh
ich
wer
e
a
p
p
lie
d
o
n
th
e
m
o
lecu
les
s
u
b
-
s
tr
u
ctu
r
e
f
o
r
p
r
ed
ictin
g
an
d
lear
n
in
g
d
if
f
er
e
n
t
p
r
o
p
e
r
ties
o
f
m
o
lecu
les.
First,
th
e
f
ea
tu
r
es
f
r
o
m
th
e
m
o
lecu
lar
in
f
o
r
m
atio
n
wer
e
ex
tr
ac
ted
at
m
o
lecu
le
an
d
n
o
d
e
lev
el
f
o
r
g
ettin
g
co
a
r
s
e
an
d
f
in
e
-
g
r
ain
e
d
in
f
o
r
m
atio
n
.
Fu
r
th
er
,
th
is
w
o
r
k
u
tili
ze
d
b
i
-
d
ir
ec
tio
n
al
G
R
Us
f
o
r
ex
tr
ac
tio
n
o
f
f
ea
tu
r
es
f
o
r
cr
ea
tin
g
a
r
ep
r
esen
tatio
n
o
f
a
g
i
v
en
m
o
l
ec
u
le.
I
n
th
is
wo
r
k
,
f
o
r
ev
alu
a
tio
n
,
E
SOL,
PDB
b
in
d
,
f
r
ee
s
o
lv
atio
n
(
Fre
eSo
lv
)
,
an
d
L
ip
o
p
h
ilicity
d
atasets
we
r
e
u
s
ed
an
d
R
MSE
was
u
s
ed
f
o
r
ev
alu
atio
n
.
Fin
d
in
g
s
s
h
o
wed
th
at
th
e
E
SOL
d
ataset
ac
h
iev
ed
R
MSE
o
f
0
.
6
5
3
,
L
ip
o
p
h
ilicity
d
ataset
ac
h
iev
ed
R
MSE
o
f
0
.
6
5
3
,
Fre
eS
o
lv
d
ataset
ac
h
iev
e
d
R
MSE
o
f
0
.
9
4
a
n
d
PDB
b
in
d
d
ataset
ac
h
iev
e
d
R
MSE
o
f
1
.
2
7
.
Fu
r
th
er
,
Den
g
et
a
l.
[
1
8
]
,
p
r
esen
ted
a
n
ar
ch
itectu
r
e
w
h
ich
c
o
n
s
is
ted
o
f
g
r
ap
h
-
n
eu
r
al
n
etwo
r
k
(
GNN)
an
d
XGBo
o
s
t
(
XGB)
ca
ll
ed
as
XGr
ap
h
B
o
o
s
t
f
o
r
ex
tr
ac
tio
n
o
f
f
ea
tu
r
es
f
r
o
m
th
e
m
o
lecu
les.
T
h
e
GNN
was
u
s
ed
as
f
ea
tu
r
e
ex
tr
ac
t
o
r
wh
er
e
th
eu
p
r
o
p
o
s
ed
th
r
ee
d
if
f
e
r
en
t
n
e
u
r
al
n
etwo
r
k
s
,
i.e
.
,
g
r
a
p
h
-
c
o
n
v
o
lu
tio
n
-
n
e
two
r
k
(
GC
N)
,
d
ir
ec
ted
-
m
ess
ag
e
-
p
ass
in
g
n
eu
r
al
n
etwo
r
k
(
DM
PNN)
an
d
g
ated
-
g
r
ap
h
n
eu
r
al
n
etwo
r
k
(
GGNN
)
.
Als
o
,
th
e
GNN
was
u
s
ed
as
p
r
ed
ictiv
e
m
o
d
el
f
o
r
p
r
ed
ictin
g
th
e
m
o
lecu
le
ch
a
r
ac
ter
is
tics
an
d
XGB
was
u
s
ed
as
class
if
icatio
n
alg
o
r
i
th
m
.
R
esu
lts
wer
e
ev
alu
ated
u
s
in
g
R
MSE
f
o
r
p
r
ed
ictio
n
(
r
e
g
r
ess
io
n
)
an
d
a
r
ea
-
u
n
d
er
-
cu
r
v
e
r
ec
eiv
er
-
o
p
e
r
atin
g
-
ch
ar
ac
ter
is
tic
(
AUC
-
R
O
C
)
f
o
r
class
if
y
in
g
t
h
e
m
o
lecu
les
(
class
if
icatio
n
)
.
Fo
r
ev
alu
atio
n
1
0
d
ata
s
ets
wer
e
u
s
ed
wh
er
e
o
n
ly
th
r
ee
d
atasets
,
i.e
.
,
L
ip
o
p
h
ilicity
,
E
SOL
an
d
Fre
eSo
lv
we
r
e
u
s
ed
f
o
r
p
r
ed
ictio
n
(
to
ev
alu
ated
th
e
f
ea
tu
r
e
ex
tr
ac
tio
n
)
.
B
y
u
tili
zin
g
th
e
G
C
N,
GGNN
,
an
d
DM
PNN,
th
e
R
MSE
s
co
r
e
ac
h
iev
ed
b
y
th
e
E
SOL
d
ataset
was
1
.
4
7
0
,
1
.
0
2
3
6
an
d
0
.
3
2
9
r
esp
e
ctiv
ely
.
Fu
r
th
er
,
f
o
r
Fre
eSo
lv
d
ataset,
th
e
GC
N,
GGNN
an
d
DM
PNN
ac
h
iev
ed
3
.
4
9
9
,
1
.
7
2
5
an
d
0
.
2
8
7
r
esp
ec
tiv
ely
.
Fin
ally
,
w
h
en
e
v
alu
atin
g
u
s
in
g
L
ip
o
p
h
ilicity
,
it
ac
h
iev
ed
R
MSE
o
f
1
.
9
1
8
,
1
.
0
0
5
a
n
d
0
.
4
5
3
f
o
r
GC
N,
GGNN
,
an
d
DM
PNN
r
esp
ec
tiv
ely
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
7
,
No
.
1
,
J
an
u
ar
y
20
2
5
:
58
0
-
5
8
9
582
L
i
an
d
J
ian
g
[
1
9
]
,
p
r
o
p
o
s
ed
a
DL
ar
ch
itectu
r
e
ca
lled
m
o
lecu
lar
b
id
ir
ec
tio
n
al
-
en
co
d
e
r
-
r
ep
r
esen
tatio
n
s
f
r
o
m
tr
an
s
f
o
r
m
er
s
(
Mo
l
-
B
E
R
T
)
wh
ich
co
m
b
in
ed
th
e
r
ep
r
esen
tatio
n
o
f
ef
f
icien
t
m
o
lecu
lar
s
tr
u
ctu
r
es
an
d
was
tr
ain
ed
u
tili
zin
g
B
E
R
T
f
o
r
p
r
ed
ictio
n
o
f
m
o
lecu
lar
ch
ar
ac
te
r
is
t
ics.
T
h
e
B
E
R
T
m
o
d
el
was
u
s
ed
as
f
ea
tu
r
e
ex
tr
ac
to
r
a
n
d
f
o
r
g
e
n
er
atin
g
m
o
lecu
le
s
s
im
p
lifie
d
m
o
lecu
lar
-
in
p
u
t
lin
e
-
en
tr
y
-
s
y
s
tem
(
SMI
L
E
S).
Fu
r
th
er
af
ter
e
x
tr
ac
tio
n
f
r
o
m
th
e
B
E
R
T
m
o
d
el
th
e
m
o
d
el
was
f
in
e
tu
n
ed
f
o
r
p
r
e
d
ictio
n
.
Fo
u
r
d
atasets
wer
e
u
s
ed
f
o
r
ev
alu
at
i
o
n
,
i.e
.
,
C
lin
T
o
x
,
T
o
x
2
1
,
SID
E
R
an
d
B
B
B
P.
T
h
e
Mo
l
-
B
E
R
T
ac
h
iev
ed
R
MSE
o
f
0
.
9
2
3
f
o
r
C
lin
to
x
,
0
.
8
3
9
f
o
r
T
o
x
2
1
,
0
.
8
7
5
f
o
r
B
B
B
P
an
d
0
.
6
9
5
f
o
r
SID
E
R
.
Yan
g
et
a
l.
[
2
0
]
,
p
r
esen
ted
a
GNN
h
av
in
g
2
7
GC
N
f
o
r
ex
tr
ac
tin
g
g
lo
b
al
a
n
d
lo
ca
l
s
tr
u
ct
u
r
es
o
f
m
o
lecu
les
(
i.e
.
,
p
r
o
tein
-
lig
an
d
co
m
p
o
u
n
d
)
,
ca
lled
as
m
o
lecu
lar
-
g
r
ap
h
d
r
u
g
-
tar
g
et
af
f
in
ity
(M
-
Gr
a
p
h
DT
A)
.
T
h
e
also
p
r
o
p
o
s
ed
a
g
r
ad
ien
t
-
weig
h
ted
af
f
in
ity
-
ac
tiv
atio
n
-
m
ap
p
i
n
g
(
Gr
ad
-
AAM
)
ap
p
r
o
ac
h
f
o
r
an
a
ly
zin
g
th
e
ch
em
ical
p
r
o
p
e
r
ties
o
f
m
o
lecu
les.
T
o
tal
o
f
s
ev
en
d
atasets
wer
e
u
s
ed
f
o
r
ev
alu
atio
n
o
f
p
r
o
p
o
s
ed
a
p
p
r
o
ac
h
es.
T
h
e
M
-
Gr
a
p
h
DT
A
a
ch
iev
ed
an
av
e
r
ag
e
R
MSE
o
f
0
.
6
9
5
f
o
r
all
s
ev
en
d
atasets
.
W
en
et
a
l.
[
2
1
]
,
p
r
esen
ted
a
m
o
lecu
la
r
-
p
r
o
p
er
ty
p
r
ed
ictiv
e
(
MPP)
ap
p
r
o
ac
h
.
T
h
is
wo
r
k
u
tili
ze
d
B
E
R
T
m
o
d
el
to
ex
tr
ac
t
im
p
o
r
tan
t
f
ea
tu
r
es
f
r
o
m
m
o
lecu
la
r
f
i
n
g
er
tip
s
,
h
en
ce
was
ca
lled
as
f
in
g
er
p
r
in
t
-
B
E
R
T
(
FP
-
B
E
R
T
)
.
T
h
e
f
ea
tu
r
es
ex
tr
ac
ted
f
r
o
m
th
e
FP
-
B
E
R
T
wer
e
th
en
u
s
ed
as
in
p
u
t
f
o
r
th
e
C
NN
lay
er
f
o
r
f
u
r
th
er
ex
tr
ac
tin
g
d
if
f
er
en
t
f
ea
t
u
r
es.
E
v
alu
atio
n
was
d
o
n
e
u
s
in
g
Fr
ee
So
lc,
E
SOL,
HI
V,
B
B
B
P,
C
E
P
an
d
M
alar
ia
d
atasets
.
T
h
e
FP
-
B
E
R
T
-
C
NN
a
ch
iev
ed
R
MSE
o
f
1
.
5
2
3
f
o
r
Fre
eSo
lv
,
1
.
2
2
f
o
r
Ma
lar
ia,
1
.
0
5
f
o
r
C
E
P,
0
.
6
0
2
f
o
r
L
ip
o
p
h
ilicity
,
an
d
0
.
5
5
2
f
o
r
E
SOL.
Fu
r
th
er
,
Gu
et
a
l
.
[
2
2
]
,
p
r
esen
ted
a
s
u
p
er
-
ed
g
e
g
r
ap
h
co
n
v
o
lu
tio
n
-
b
ased
s
u
p
er
v
is
ed
atten
tio
n
-
b
ased
DT
A
(
SEGSA
-
DT
A)
w
h
ich
ex
tr
ac
ted
e
d
g
e
an
d
n
o
d
e
d
ata
f
r
o
m
m
o
lecu
le
s
u
s
in
g
v
ar
io
u
s
atten
tio
n
m
o
d
u
les
to
lear
n
th
e
d
is
tr
ib
u
tio
n
o
f
p
r
o
tein
-
lig
an
d
s
in
ter
ac
tio
n
s
.
Fo
r
ev
alu
atio
n
,
P
DB
b
in
d
was u
s
ed
wh
er
e
SEGSA
-
DT
A
ac
h
iev
ed
R
MSE
o
f
1
.
3
1
9
.
Xu
et
a
l.
[
2
3
]
,
p
r
esen
ted
an
ar
ch
ite
ctu
r
e
wh
ich
co
m
b
in
ed
g
r
ap
h
-
atten
tio
n
-
n
etwo
r
k
(
GAT
)
with
s
elf
-
atten
tio
n
lo
n
g
-
s
h
o
r
t
ter
m
-
m
e
m
o
r
y
(
SA
-
L
STM
)
f
o
r
e
x
tr
ac
tio
n
o
f
f
ea
t
u
r
es
f
r
o
m
g
r
a
p
h
an
d
s
eq
u
en
ce
s
o
f
m
o
lecu
les.
I
n
th
is
wo
r
k
,
S
MI
L
E
S
wer
e
in
itially
u
s
ed
,
th
en
GAT
was
u
s
ed
f
o
r
f
ea
tu
r
e
ex
t
r
ac
tio
n
.
T
h
e
atten
tio
n
lay
er
s
o
f
GAT
an
d
SA
-
L
STM
wer
e
b
o
th
co
m
b
in
ed
to
u
n
d
er
s
tan
d
th
e
c
o
m
p
lete
m
o
lecu
le.
T
h
e
ev
al
u
atio
n
was
d
o
n
e
u
s
in
g
v
a
r
io
u
s
d
atasets
to
ev
alu
ate
th
eir
wo
r
k
.
T
h
e
SA
-
L
STM
G
AT
ac
h
iev
ed
R
MSE
o
f
0
.
7
0
9
f
o
r
L
ip
o
p
h
ilicity
,
0
.
8
8
5
f
o
r
E
SOL
a
n
d
1
.
2
1
1
f
o
r
Fr
ee
So
lv
d
atasets
r
esp
ec
tiv
ely
.
Pecin
a
et
a
l.
[
2
4
]
,
p
r
o
p
o
s
ed
a
s
co
r
i
n
g
a
p
p
r
o
ac
h
ca
lled
as
s
em
i
-
em
p
ir
ical
q
u
an
tu
m
-
m
ec
h
an
ical
s
co
r
in
g
(
SQM2
.
2
0
)
to
s
o
lv
e
th
e
is
s
u
e
o
f
b
in
d
in
g
.
Fu
r
th
er
,
to
ev
alu
ate
its
wo
r
k
,
th
e
PL
-
R
E
X
d
ataset
wh
ich
co
n
s
is
ts
o
f
v
ar
io
u
s
cr
y
s
tal
-
lik
e
s
tr
u
ctu
r
es
h
av
in
g
1
0
p
r
o
tein
t
ar
g
ets
was
co
n
s
id
er
ed
.
T
h
e
S
QM
2
.
2
0
ac
h
iev
ed
R
MSE
o
f
0
.
6
9
f
o
r
PL
-
R
E
X
an
d
s
o
lv
ed
th
e
is
s
u
e
o
f
co
s
t
in
d
en
s
ity
-
f
u
n
ctio
n
al
-
th
eo
r
y
(
DFT)
.
Fin
ally
,
[
2
5
]
p
r
esen
ted
a
clea
n
ed
v
er
s
io
n
o
f
PDB
b
in
d
d
ataset
(
co
n
s
is
tin
g
o
f
p
r
o
tein
-
lig
an
d
s
)
wh
ic
h
was f
r
ee
o
f
d
ata
lea
k
ag
e
an
d
co
u
ld
b
e
u
s
ed
f
o
r
tr
ain
in
g
an
d
test
in
g
.
T
h
e
d
ataset
co
n
s
i
s
ted
o
f
lo
n
g
s
eq
u
en
ce
s
o
f
m
o
lecu
le
s
an
d
h
av
i
n
g
s
im
ilar
s
tr
u
ctu
r
al
s
im
ilar
ity
.
T
h
e
d
ataset
was
th
en
u
s
ed
f
o
r
ev
alu
atin
g
th
e
ex
is
tin
g
s
c
o
r
in
g
m
o
d
els,
i.e
.
,
r
an
d
o
m
-
f
o
r
s
t
s
co
r
in
(
R
FS
)
,
in
ter
ac
tio
n
-
g
r
a
p
h
-
n
etwo
r
k
(
I
G
N)
,
au
to
-
d
o
ck
v
in
a
(
ADV)
an
d
d
eep
-
DT
A.
T
h
e
y
also
p
r
esen
ted
a
n
o
v
el
s
co
r
i
n
g
ca
lle
d
lin
ea
r
-
p
o
ten
tial
(LP
-
PDB
b
in
d
)
.
Fin
d
in
g
s
s
h
o
w
th
at
th
e
L
P
-
PD
B
b
in
d
s
h
o
wed
b
etter
r
esu
lts
in
co
m
p
ar
is
o
n
with
e
x
is
tin
g
ap
p
r
o
ac
h
es.
Fro
m
th
e
ab
o
v
e
liter
atu
r
e
s
u
r
v
ey
,
it
ca
n
b
e
s
ee
n
t
h
at
u
s
in
g
th
e
Fre
eSo
lv
,
E
SOL
an
d
L
ip
o
p
h
ilicity
d
ataset,
v
ar
io
u
s
m
o
d
els
h
av
e
ac
h
iev
ed
b
etter
r
esu
lts
.
B
u
t
s
till
v
er
y
less
wo
r
k
h
as
b
ee
n
co
n
s
id
er
ed
o
n
PDB
B
in
d
d
ataset
to
ac
h
iev
e
le
s
s
R
MSE
.
Fu
r
th
er
,
it c
an
also
b
e
s
ee
n
th
at
m
o
s
t o
f
th
e
wo
r
k
ar
e
u
tili
zin
g
B
E
R
T
m
o
d
els
f
o
r
f
ea
tu
r
e
ex
t
r
ac
tio
n
an
d
s
im
ilar
ity
s
co
r
in
g
a
p
p
r
o
ac
h
es
f
o
r
e
v
alu
at
in
g
t
h
e
s
im
ilar
ity
am
o
n
g
th
e
m
o
lecu
les
o
f
p
r
o
tein
-
lig
an
d
s
an
d
its
o
u
tp
u
t
g
iv
e
n
to
C
NN
f
o
r
b
etter
o
u
tco
m
es.
Hen
ce
,
in
th
is
wo
r
k
,
we
p
r
esen
t a
n
ap
p
r
o
ac
h
wh
ich
u
ti
lizes B
E
R
T
m
o
d
el
with
a
n
o
v
el
s
im
ilar
ity
s
co
r
in
g
ap
p
r
o
ac
h
with
C
NN
ca
lled
as
E
MSPLA
-
C
N
N.
T
h
e
m
et
h
o
d
o
lo
g
y
f
o
r
th
e
E
MSPLA
-
C
NN
is
p
r
esen
ted
b
el
o
w.
3.
M
E
T
H
O
D
I
n
th
is
m
eth
o
d
o
lo
g
y
s
ec
tio
n
,
we
will
f
ir
s
t
d
is
cu
s
s
th
e
ar
ch
i
tectu
r
e.
Nex
t,
we
will
co
v
er
th
e
d
ataset
u
s
ed
.
Fo
llo
win
g
th
at,
we
will
ex
p
lain
h
o
w
th
e
p
r
o
tein
-
lig
an
d
s
ar
e
r
ep
r
esen
ted
.
W
e
will
th
en
p
r
esen
t
th
e
p
r
o
p
o
s
ed
E
MSPLA
.
Fin
ally
,
we
will
d
is
cu
s
s
h
o
w
th
e
C
NN
is
u
s
ed
f
o
r
f
ea
tu
r
e
e
x
tr
ac
tio
n
f
r
o
m
th
e
p
r
o
tein
-
lig
an
d
s
tr
u
ctu
r
al
s
im
ilar
ity
.
T
h
en
f
in
ally
,
th
is
wo
r
k
d
is
cu
s
s
es th
e
class
if
ier
u
s
ed
in
th
is
wo
r
k
.
3
.
1
.
Arc
hite
ct
ure
T
h
e
p
r
o
p
o
s
ed
E
MSPLA
-
C
NN
ar
ch
it
ec
tu
r
e
f
ir
s
t
ev
alu
ates
th
e
g
iv
en
p
r
o
tein
-
lig
an
d
b
in
d
in
g
s
tr
u
ctu
r
e
as
g
iv
en
in
Fig
u
r
e
1
.
T
h
e
ev
a
lu
atio
n
is
d
o
n
e
b
y
ev
alu
atin
g
th
e
p
r
o
tein
-
lig
an
d
s
eq
u
en
ce
s
.
T
h
en
th
e
p
r
o
tein
-
lig
an
d
s
eq
u
en
ce
s
ar
e
s
ep
ar
ated
u
s
in
g
B
E
R
T
m
o
d
el
f
o
r
f
u
r
th
er
ev
alu
atio
n
u
s
in
g
th
e
p
r
o
p
o
s
ed
E
MSPLA
s
im
ilar
ity
ev
alu
atio
n
.
Af
ter
t
h
e
ev
alu
atio
n
o
f
th
e
s
im
ilar
ity
,
th
e
s
eq
u
en
ce
g
o
es
to
th
e
C
NN
wh
er
e
it
h
as
two
co
m
p
o
n
en
ts
,
a
n
eu
r
al
-
n
etwo
r
k
lay
er
,
an
d
atten
tio
n
n
etwo
r
k
.
T
h
e
C
NN
is
u
s
ed
as
f
ea
t
u
r
e
e
x
tr
ac
to
r
wh
e
r
e
th
e
lo
ca
l
f
ea
tu
r
es
am
o
n
g
th
e
n
ei
g
h
b
o
r
r
esid
u
es
ar
e
e
x
tr
ac
ted
.
Fu
r
th
er
,
f
o
r
e
x
tr
ac
tin
g
m
o
r
e
f
ea
tu
r
es
f
r
o
m
th
e
p
r
o
tein
s
eq
u
en
ce
s
,
t
h
e
C
NN
was
s
tack
ed
in
b
lo
ck
s
.
L
o
n
g
s
eq
u
en
ce
s
o
f
p
r
o
tein
a
n
d
th
ei
r
b
in
d
in
g
r
esid
u
es
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
MS
P
LA
fo
r
a
cc
u
r
a
te
fea
tu
r
e
mo
lecu
la
r
ex
tr
a
ctio
n
fr
o
m
p
r
o
tein
-
lig
a
n
d
i
n
tera
ctio
n
s
(
S
r
in
i
d
h
i Ku
lka
r
n
i V
)
583
wer
e
ca
p
tu
r
ed
u
s
in
g
th
e
atten
t
io
n
n
etwo
r
k
.
Fin
ally
,
o
u
tp
u
ts
f
r
o
m
b
o
th
th
e
n
e
two
r
k
wer
e
m
er
g
ed
an
d
p
ass
ed
to
th
e
f
in
al
co
n
n
ec
ted
lay
er
,
wh
er
e
th
e
b
in
ar
y
p
r
e
d
ictio
n
to
o
k
p
lace
u
s
in
g
th
e
class
if
ier
p
r
o
p
o
s
ed
in
th
is
wo
r
k
.
T
h
e
is
s
u
e
o
f
o
v
er
f
itti
n
g
was
p
r
ev
en
ted
b
y
u
tili
zin
g
th
e
d
r
o
p
o
u
t
an
d
weig
h
d
ec
ay
in
g
ap
p
r
o
ac
h
(
n
o
t
u
s
ed
in
co
n
v
o
l
u
ti
o
n
lay
er
)
.
T
h
e
co
m
p
l
ete
f
lo
w
o
f
th
e
ar
ch
itectu
r
e
is
p
r
esen
ted
in
Fig
u
r
e
1
.
Fig
u
r
e
1
.
Ar
c
h
itectu
r
e
3.
2
.
Da
t
a
s
et
p
re
pro
ce
s
s
ing
Fo
r
ev
alu
atin
g
th
is
wo
r
k
th
e
PDB
b
n
in
d
v
.
2
0
2
0
[
2
5
]
d
ata
s
et
was
u
tili
ze
d
.
T
h
e
PD
B
b
i
n
d
v
.
2
0
2
0
d
ataset
in
clu
d
es
b
in
d
in
g
-
s
ite
in
f
o
r
m
atio
n
f
o
r
1
9
,
4
4
3
co
m
p
lex
es
o
f
p
r
o
tein
-
lig
an
d
th
at
h
a
v
e
b
ee
n
id
e
n
tifie
d
th
r
o
u
g
h
ex
p
er
im
en
tatio
n
.
T
h
e
in
f
o
r
m
atio
n
is
av
ailab
le
in
b
o
t
h
MO
L
2
an
d
PDB
f
ile
f
o
r
m
at
s
.
T
h
e
s
elec
tio
n
o
f
co
m
p
lex
p
r
o
tein
-
lig
an
d
f
r
o
m
th
e
PDB
b
in
d
d
ataset
wa
s
d
o
n
e
b
y
co
n
s
id
er
i
n
g
d
if
f
e
r
en
t
d
ec
is
io
n
s
wh
ich
ar
e
d
is
cu
s
s
ed
as
f
o
llo
ws.
Sin
ce
th
e
to
tal
len
g
th
o
f
th
e
s
eq
u
en
ce
o
f
p
r
o
tein
-
lig
a
n
d
d
if
f
er
s
f
r
o
m
o
n
e
p
r
o
tein
-
lig
a
n
d
to
an
o
th
er
a
n
d
s
in
ce
in
clu
s
io
n
o
f
s
eq
u
e
n
ce
s
th
at
ar
e
to
o
le
n
g
th
y
ca
n
d
eg
r
a
d
e
E
MSPLA
-
C
NN
p
er
f
o
r
m
an
ce
,
he
n
ce
,
in
th
is
wo
r
k
,
f
ir
s
t
im
p
o
s
ed
a
r
estrictio
n
th
at
th
e
p
r
o
tei
n
-
lig
an
d
s
eq
u
en
ce
c
o
u
ld
n
o
t
e
x
ce
ed
1
5
0
0
am
in
o
-
ac
id
r
esid
u
es.
T
h
is
led
t
o
th
e
ex
clu
s
io
n
o
f
co
m
p
lex
p
r
o
tei
n
-
lig
an
d
s
eq
u
en
ce
s
f
r
o
m
th
e
d
atasets
wh
en
th
e
len
g
th
o
f
p
r
o
tein
-
lig
a
n
d
s
eq
u
en
ce
was
m
o
r
e
t
h
an
1
5
0
0
.
T
h
e
s
ec
o
n
d
s
tep
was
th
e
elim
in
atio
n
o
f
c
o
m
p
lex
es
co
n
tain
in
g
lig
an
d
s
wh
ic
h
p
r
o
v
ed
n
o
t
co
m
p
atib
le
with
th
e
R
DKit
an
d
Op
en
b
ab
el
li
b
r
ar
i
es.
Af
ter
co
m
p
letio
n
o
f
PDB
b
in
d
p
r
ep
r
o
ce
s
s
in
g
a
to
tal
o
f
3
4
,
0
0
5
p
r
o
tein
-
lig
a
n
d
s
eq
u
en
ce
s
wer
e
ac
h
ie
v
ed
.
U
tili
zin
g
in
d
iv
id
u
al
PDB
I
Ds,
th
is
wo
r
k
ac
q
u
ir
e
d
d
ata
to
tr
ain
f
o
r
2
8
,
7
2
8
co
m
p
l
ex
p
r
o
tein
-
lig
an
d
s
eq
u
e
n
ce
s
.
T
o
s
to
p
in
f
o
r
m
atio
n
f
r
o
m
leak
in
g
o
u
t
as
p
r
esen
te
d
in
[
2
6
]
,
we
s
ep
ar
ated
th
e
t
est
an
d
tr
ain
in
g
d
atasets
an
d
d
id
n
’
t
u
s
e
id
en
tical
p
r
o
tein
-
lig
a
n
d
s
eq
u
e
n
ce
s
.
B
y
c
o
m
p
ar
in
g
th
e
s
tr
u
ctu
r
al
s
im
il
ar
ities
o
f
b
o
th
test
in
g
an
d
tr
ai
n
in
g
p
r
o
tein
-
lig
an
d
u
tili
zin
g
th
e
s
u
g
g
ested
E
MSPLA
,
th
is
wo
r
k
was
a
b
le
to
cr
ea
te
u
n
s
ee
n
-
p
r
o
tein
test
d
atas
ets,
wh
ich
we
th
en
u
s
ed
to
ass
ess
b
in
d
in
g
s
ite
p
r
e
d
ictio
n
s
f
o
r
u
n
k
n
o
wn
p
r
o
tein
-
l
ig
an
d
s
tr
u
ctu
r
es.
3.
3
.
E
nh
a
nced
m
o
lecula
r
s
im
ila
rit
y
pro
t
ein
-
lig
a
nd
a
lig
ner
Acc
o
r
d
in
g
to
r
esear
ch
[
1
]
,
a
p
r
o
tein
-
lig
an
d
s
eq
u
en
ce
is
co
m
p
o
s
ed
o
f
a
s
eq
u
e
n
ce
o
f
a
m
in
o
-
ac
id
s
ar
r
an
g
e
d
in
a
p
ar
ticu
lar
o
r
d
e
r
with
lig
an
d
s
co
n
ca
ten
atin
g
d
if
f
er
en
t
p
r
o
tein
s
.
I
t
h
as
b
ee
n
d
e
ter
m
in
ed
th
at
th
er
e
ar
e
r
o
u
g
h
ly
2
0
d
is
tin
ctiv
e
a
m
in
o
-
ac
id
s
(
s
tan
d
a
r
d
am
i
n
o
-
a
cid
s
r
esid
u
es),
an
d
th
e
c
h
ar
a
cter
is
tic
s
o
f
p
r
o
tein
-
lig
an
d
s
eq
u
en
ce
s
v
a
r
y
b
ased
o
n
th
e
p
ar
ticu
lar
s
eq
u
en
ce
o
f
th
ese
am
in
o
-
ac
id
s
.
I
t
h
as
b
ee
n
o
b
s
er
v
ed
th
at
ce
r
tain
p
r
o
tein
s
m
ig
h
t
h
av
e
s
y
n
th
etic
r
esid
u
es
(
n
o
n
-
s
tan
d
ar
d
r
esid
u
es).
I
n
o
r
d
er
t
o
g
en
e
r
ate
th
e
2
1
r
esid
u
es
id
en
tifie
d
in
th
e
p
r
o
tein
-
lig
a
n
d
s
eq
u
en
ce
s
,
th
is
wo
r
k
u
tili
ze
d
E
MSPLA
,
a
p
r
e
-
tr
ain
ed
ap
p
r
o
ac
h
th
at
is
b
ased
o
n
th
e
Sm
ith
-
W
ater
m
an
(
SW
)
ap
p
r
o
ac
h
s
p
ec
if
ically
d
e
s
ig
n
ed
f
o
r
an
aly
zin
g
lar
g
e
-
s
ca
le
p
r
o
tein
-
lig
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
7
,
No
.
1
,
J
an
u
ar
y
20
2
5
:
58
0
-
5
8
9
584
s
eq
u
en
ce
s
.
C
o
n
s
id
er
th
at
0
an
d
1
is
u
s
ed
f
o
r
r
e
p
r
esen
tin
g
two
p
r
o
tein
-
lig
an
d
s
eq
u
e
n
ce
s
wh
ich
ar
e
i
d
en
tifie
d
u
s
in
g
B
E
R
T
m
o
d
el.
T
h
e
m
ai
n
aim
o
f
th
e
SW
ap
p
r
o
ac
h
is
to
co
m
p
u
te
th
e
s
im
ilar
ity
am
o
n
g
t
h
e
two
-
p
r
o
tei
n
lig
an
d
s
eq
u
en
ce
s
.
T
h
e
s
im
ilar
ity
is
s
to
r
ed
in
a
m
atr
ix
f
o
r
m
wh
ich
is
ca
lled
a
s
an
d
d
en
o
te
d
as
.
Usi
n
g
th
e
an
d
b
ac
k
-
tr
ac
k
in
g
a
p
p
r
o
a
ch
,
b
est
alig
n
m
en
t
is
ac
h
iev
e
d
.
Fu
r
th
er
,
c
o
n
s
id
er
wh
ich
d
en
o
tes
s
ize
o
f
0
p
r
o
tein
-
lig
a
n
d
s
eq
u
e
n
ce
an
d
d
en
o
tes
s
ize
o
f
1
p
r
o
tein
-
lig
a
n
d
s
eq
u
en
ce
.
I
t
is
ass
u
m
ed
th
at
f
o
r
0
,
th
er
e
ex
is
t
+
1
p
r
ef
ix
es,
Similar
ly
,
f
o
r
1
,
th
er
e
ex
is
ts
+
1
p
r
ef
ix
es.
Als
o
,
th
e
0
an
d
1
ca
n
co
n
tain
n
u
ll
s
eq
u
en
ce
s
.
C
o
n
s
id
er
[
1
…
]
wh
ich
r
ep
r
esen
ts
th
e
p
r
ef
ix
f
o
r
an
d
[
]
wh
ich
r
ep
r
esen
ts
p
r
ef
ix
f
o
r
r
esp
ec
tiv
ely
,
wh
er
e
r
ep
r
esen
ts
th
e
s
eq
u
en
ce
s
.
Fu
r
th
er
,
th
e
am
o
n
g
th
e
p
r
e
f
ix
es
0
[
1
…
]
an
d
1
[
1
…
]
is
d
en
o
ted
as
,
.
T
h
e
s
ize
is
d
en
o
ted
as
(
+
1
,
+
1
)
.
T
h
e
in
itial
co
lu
m
n
an
d
r
o
w
o
f
is
in
itially
s
et
to
0
.
T
h
e
f
o
r
th
e
s
eq
u
en
ce
is
r
ep
r
ese
n
ted
as
,
wh
ich
is
ev
alu
ated
u
tili
zin
g
(
1
)
.
,
=
{
0
,
,
−
1
,
−
1
−
(
,
)
(
1
)
I
n
(
1
)
,
(
,
)
p
u
r
p
o
s
e
is
to
p
r
o
v
id
e
r
esu
lts
,
i.e
.
,
m
is
m
atch
o
r
th
er
e
is
a
m
atch
b
etwe
en
two
s
eq
u
en
ce
s
.
I
f
th
er
e
is
a
m
atc
h
,
th
e
n
,
t
h
is
is
r
ep
r
esen
ted
as
0
[
]
=
1
[
]
an
d
(
,
)
=
.
I
f
t
h
er
e
is
n
o
m
atch
,
th
en
it
is
r
ep
r
esen
ted
as
0
[
]
≠
1
[
]
an
d
(
,
)
=
.
Als
o
,
in
(
1
)
,
th
e
an
d
d
en
o
te
th
e
m
atr
ix
with
r
esp
ec
t
to
th
e
p
r
o
tein
-
lig
an
d
af
f
in
ity
g
ap
ap
p
r
o
ac
h
.
T
h
e
m
atr
ix
an
d
ar
e
u
tili
ze
d
f
o
r
ev
alu
atin
g
th
e
g
ap
s
,
h
en
ce
,
ar
e
ev
al
u
ated
u
s
i
n
g
(
2
)
an
d
(
3
)
r
esp
ec
tiv
ely
.
,
=
{
,
−
1
−
,
−
1
−
(
2
)
,
=
{
−
1
,
−
−
1
,
−
(
3
)
I
n
(
2
)
a
n
d
(
3
)
,
an
d
d
en
o
te
f
ir
s
-
g
ap
an
d
s
u
cc
ess
iv
e
-
g
ap
p
e
n
alty
.
I
n
th
is
wo
r
k
,
th
e
b
ac
k
-
r
a
ck
in
g
ap
p
r
o
ac
h
h
as
b
ee
n
u
s
ed
with
SW
f
o
r
f
in
d
in
g
b
est
alig
n
er
am
o
n
g
0
an
d
1
.
T
h
e
p
r
o
ce
s
s
o
f
b
ac
k
-
tr
ac
k
in
g
s
tar
ts
u
s
in
g
th
e
h
av
in
g
th
e
m
a
x
im
u
m
s
co
r
e
an
d
g
o
es
to
th
e
f
in
al
c
ell
wh
er
e
a
n
u
ll v
alu
e
ce
l
l
is
r
ea
ch
ed
.
T
h
e
p
u
r
p
o
s
e
o
f
SW
ap
p
r
o
ac
h
is
to
p
r
o
v
id
e
b
est
s
im
ilar
ity
s
co
r
e
an
d
b
est
alig
n
er
s
.
T
h
e
c
o
m
p
u
tatio
n
o
f
tak
es m
o
r
e
tim
e
as it h
as to
m
atch
an
d
p
er
f
o
r
m
b
ac
k
-
tr
ac
k
in
g
p
r
o
ce
s
s
s
im
u
ltan
eo
u
s
ly
.
Fu
r
th
er
,
in
th
e
EMS
PLA
,
f
o
r
s
o
lv
in
g
th
e
ab
o
v
e
is
s
u
es,
an
in
d
ex
in
g
ap
p
r
o
ac
h
is
u
s
ed
b
y
u
tili
zin
g
q
u
er
y
an
d
r
ef
er
e
n
ce
s
eq
u
e
n
ce
d
en
o
te
d
as
an
d
.
T
h
e
s
eq
u
en
ce
is
f
o
r
th
e
s
eq
u
en
ce
is
co
n
s
tr
u
cted
in
t
h
e
f
o
r
m
at
o
f
d
ir
ec
ted
-
ac
y
clic
-
g
r
ap
h
(
DAG)
wh
ich
r
ep
r
esen
ts
th
e
0
s
eq
u
en
ce
.
Fo
r
s
eq
u
en
c
e,
a
p
r
ef
ix
-
t
r
ee
s
tr
u
ctu
r
e
is
co
n
s
tr
u
cted
wh
ic
h
r
ep
r
esen
ts
th
e
1
s
eq
u
en
ce
.
T
h
e
p
r
e
f
ix
-
tr
ee
is
co
n
s
tr
u
cte
d
i
n
s
u
ch
a
m
a
n
n
e
r
s
u
ch
th
at
s
im
ilar
ity
am
o
n
g
t
h
e
ed
g
es
an
d
n
o
d
es
ca
n
b
e
f
o
u
n
d
i
n
a
f
aster
way
.
I
n
th
e
f
i
n
al
p
r
o
ce
s
s
o
f
th
is
ap
p
r
o
ac
h
,
a
d
is
tin
ctiv
e
s
tr
in
g
is
ac
h
iev
ed
wh
ich
is
a
s
u
b
-
s
tr
in
g
o
f
s
eq
u
en
ce
.
Fu
r
th
er
,
ev
er
y
n
o
d
e
in
th
e
g
iv
en
tr
ee
is
d
en
o
ted
u
s
in
g
a
n
d
in
ter
v
al
s
u
f
f
ix
-
ar
r
ay
.
B
y
u
s
in
g
th
e
tr
av
er
s
in
g
ap
p
r
o
ac
h
,
th
e
n
o
d
es
wh
ich
cr
ea
te
th
e
s
tr
in
g
ar
e
s
o
r
ted
i
n
lex
ico
g
r
a
p
h
ically
way
.
Fu
r
th
er
,
u
s
in
g
th
e
p
r
ef
ix
-
tr
ee
,
th
e
n
o
d
es
wh
ich
ar
e
s
im
ilar
o
r
i
d
en
tical
to
s
u
f
f
ix
-
a
r
r
ay
a
r
e
u
s
ed
to
cr
ea
te
a
Pre
f
ix
DAG
(
P
-
DAG)
.
E
v
er
y
n
o
d
e
in
P
-
DAG
is
u
s
ed
to
d
en
o
te
s
in
g
le
o
r
m
o
r
e
s
u
b
-
s
tr
in
g
o
f
s
eq
u
en
ce
.
C
o
n
s
id
er
(
)
an
d
(
)
ar
e
t
wo
f
u
n
ctio
n
s
wh
ich
ar
e
u
s
ed
f
o
r
co
n
s
tr
u
ctin
g
p
r
e
f
ix
-
tr
ee
an
d
P
-
DAG,
wh
er
e
r
ep
r
esen
ts
or
s
eq
u
en
ce
.
I
n
t
h
e
E
MSPLA
tech
n
iq
u
e,
th
e
(
)
an
d
(
)
ar
e
c
o
m
p
u
ted
in
itially
.
C
o
n
s
id
er
wh
ich
d
en
o
tes
r
o
o
t
-
n
o
d
e
f
o
r
(
)
an
d
d
en
o
tes
r
o
o
t
-
n
o
d
e
f
o
r
(
)
.
T
h
e
b
est
am
o
n
g
th
e
an
d
s
eq
u
en
ce
s
ar
e
ev
alu
ated
u
s
in
g
d
y
n
am
ic
-
p
r
o
g
r
a
m
m
in
g
ap
p
r
o
ac
h
.
Fu
r
th
er
,
we
c
o
n
s
id
er
th
r
ee
,
i.e
.
,
,
,
,
,
an
d
,
wh
ich
a
r
e
in
itially
s
et
to
n
u
ll
b
y
co
n
s
id
er
in
g
t
h
e
r
o
o
t
-
n
o
d
es
(
)
an
d
(
)
.
Fo
r
ev
er
y
p
ar
e
n
t
in
tr
ee
in
(
)
,
ev
alu
atio
n
is
d
o
n
e
f
o
r
|
,
|
,
an
d
|
.
T
h
e
ev
alu
atio
n
o
f
|
is
d
o
n
e
u
s
in
g
(
4
)
.
|
=
{
,
−
}
−
(
4
)
W
h
er
e,
r
ep
r
esen
ts
p
r
o
tein
-
lig
an
d
o
p
en
-
g
ap
p
e
n
alty
an
d
r
ep
r
esen
ts
th
e
p
r
o
tein
-
lig
an
d
g
ap
-
ex
ten
s
io
n
f
u
n
ctio
n
.
Fu
r
th
e
r
,
|
is
ev
alu
at
ed
b
y
u
tili
zin
g
(
5
)
.
I
n
(
5
)
,
r
ep
r
esen
ts
p
ar
en
t
f
o
r
n
o
d
e
in
(
)
.
Fin
ally
,
th
e
ev
alu
atio
n
o
f
|
is
d
o
n
e
u
s
in
g
(
6
)
.
I
n
(
6
)
,
(
,
;
,
)
d
en
o
tes
th
e
am
o
n
g
th
e
ed
g
es
(
,
)
an
d
(
,
)
.
T
h
e
ev
al
u
atio
n
o
f
t
h
e
all
th
e
,
i.e
.
,
,
,
an
d
is
d
o
n
e
u
s
in
g
(
7
)
.
|
=
{
,
−
}
−
(
5
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
MS
P
LA
fo
r
a
cc
u
r
a
te
fea
tu
r
e
mo
lecu
la
r
ex
tr
a
ctio
n
fr
o
m
p
r
o
tein
-
lig
a
n
d
i
n
tera
ctio
n
s
(
S
r
in
i
d
h
i Ku
lka
r
n
i V
)
585
|
=
{
+
(
,
;
,
)
,
,
,
0
}
(
6
)
(
,
,
)
=
{
(
|
′
,
|
′
,
|
′
)
ℎ
|
′
>
0
(
−
∞
,
−
∞
,
−
∞
)
ℎ
(
7
)
I
n
(
7
)
,
′
=
∈
(
)
|
an
d
(
)
r
ep
r
esen
t
th
e
p
ar
en
t
-
n
o
d
es
f
o
r
.
Fu
r
th
er
,
d
en
o
tes
b
est
f
o
r
s
u
b
-
s
tr
in
g
,
th
at
is
s
im
ilar
ity
am
o
n
g
s
u
b
-
s
tr
in
g
an
d
s
u
b
-
s
tr
in
g
.
W
h
en
>
0
,
th
en
t
h
e
s
u
b
-
s
tr
in
g
h
as
a
g
o
o
d
m
atch
with
s
u
b
-
s
tr
in
g
.
As
th
e
SW
tech
n
iq
u
e
p
r
o
v
id
ed
b
etter
alig
n
er
b
u
t
to
o
k
m
o
r
e
tim
e
f
o
r
c
o
m
p
u
tatio
n
,
h
e
n
ce
,
th
e
E
MSPLA
r
ed
u
ce
s
tim
e
b
y
r
ed
u
cin
g
th
e
co
m
p
u
tatio
n
an
d
b
y
u
s
in
g
th
e
tr
av
er
s
al
ap
p
r
o
ac
h
o
n
(
)
an
d
(
)
.
T
h
e
d
y
n
am
ic
-
p
r
o
g
r
am
m
in
g
h
elp
s
to
i
d
en
tify
b
etter
m
atch
es
a
m
o
n
g
th
e
p
r
o
tein
-
li
g
an
d
s
eq
u
en
ce
s
.
Fu
r
th
er
,
a
m
atc
h
in
g
-
p
air
(
,
)
is
co
n
s
tr
u
cted
wh
en
a
n
o
p
ti
m
al
is
ac
h
iev
ed
,
th
at
is
wh
en
is
h
ig
h
er
an
d
s
u
f
f
ix
-
ar
r
ay
is
o
f
s
ize
with
in
th
e
th
r
esh
o
ld
v
alu
e.
Fu
r
th
er
,
in
th
e
m
atch
in
g
p
r
o
ce
s
s
,
th
e
o
p
tim
al
m
atch
is
ac
h
iev
ed
f
r
o
m
(
,
)
an
d
b
y
a
n
aly
s
is
o
f
s
u
f
f
ix
-
ar
r
a
y
s
o
f
an
d
s
eq
u
en
ce
s
.
Fu
r
th
e
r
,
t
h
e
p
r
o
tei
n
-
lig
an
d
s
a
r
e
u
s
ed
f
o
r
ex
tr
ac
ti
n
g
f
ea
tu
r
es
u
s
in
g
th
e
C
NN
w
h
ich
is
d
is
cu
s
s
ed
in
th
e
n
ex
t sectio
n
.
3.
4
.
F
e
a
t
ure
e
x
t
r
a
ct
io
n us
ing
CNN
Usi
n
g
th
e
p
r
o
tein
-
lig
an
d
s
tr
u
ctu
r
e
s
im
ilar
ity
ac
h
iev
ed
f
r
o
m
E
MSPLA
,
th
e
o
u
tp
u
t
is
p
ass
ed
o
n
th
e
C
NN
lay
er
f
o
r
f
ea
tu
r
e
e
x
tr
ac
tio
n
.
T
h
e
E
MSPLA
s
co
r
e
s
th
e
p
r
o
tein
-
lig
a
n
d
in
b
etwe
en
0
an
d
1
in
d
icatin
g
th
at
h
ig
h
er
s
co
r
e
h
as
s
tr
u
ctu
r
e
s
im
ilar
to
th
e
n
ex
t
p
r
o
tein
-
lig
an
d
s
eq
u
en
ce
.
W
h
en
e
v
er
,
t
h
e
s
im
i
lar
ity
s
co
r
e
am
o
n
g
two
p
r
o
tein
-
lig
a
n
d
h
as
m
o
r
e
th
an
two
s
co
r
es,
th
en
th
e
m
a
x
i
m
u
m
p
air
-
wis
e
ch
ain
E
MSPL
A
-
s
co
r
e
was
g
iv
en
as
s
im
ilar
ity
s
co
r
e.
Als
o
,
th
e
d
ataset
wh
ich
was
co
n
s
id
er
ed
f
o
r
test
in
g
co
m
p
r
is
ed
p
r
o
tein
-
lig
an
d
s
h
a
v
in
g
≤
40%
o
f
s
im
ilar
ity
wh
en
c
o
m
p
ar
e
d
with
th
e
p
r
o
tein
-
lig
an
d
tr
ai
n
in
g
s
et.
3.
4
.
1
.
E
m
bedd
ing
s
T
h
e
E
MSPLA
-
C
NN
ap
p
r
o
ac
h
f
u
r
th
er
u
s
es B
E
R
T
ap
p
r
o
ac
h
f
o
r
ex
tr
ac
tio
n
o
f
am
in
o
-
ac
i
d
em
b
ed
d
in
g
s
r
ep
r
esen
ted
as
∈
×
f
r
o
m
t
h
e
p
r
o
tein
-
lig
an
d
s
eq
u
en
ce
as
it
h
elp
s
to
ex
tr
ac
t
b
etter
f
ea
tu
r
es
wh
en
s
ep
ar
atin
g
th
e
am
i
n
o
ac
id
s
a
n
d
p
r
o
tein
s
f
r
o
m
th
e
p
r
o
tein
-
li
g
an
d
s
eq
u
en
ce
.
T
h
e
r
ep
r
esen
t
h
ig
h
est
s
eq
u
en
ce
len
g
th
o
f
p
r
o
tei
n
an
d
r
ep
r
ese
n
ts
h
id
d
en
-
s
ize.
T
h
e
s
eq
u
en
ce
s
wh
ich
a
r
e
less
in
co
m
p
ar
is
o
n
with
ar
e
s
et
to
ze
r
o
.
Usi
n
g
th
e
am
in
o
-
ac
i
d
em
b
ed
d
in
g
s
,
th
e
p
r
o
tein
em
b
e
d
d
in
g
s
ar
e
ac
h
iev
e
d
b
y
av
e
r
a
g
in
g
am
in
o
-
ac
id
em
b
ed
d
in
g
s
an
d
r
e
p
r
esen
ted
as
∈
.
I
n
an
y
cir
cu
m
s
tan
ce
w
h
er
e
th
e
p
r
o
tein
-
lig
an
d
h
as
m
u
lti
-
ch
ain
s
eq
u
en
ce
s
tr
u
ctu
r
e,
th
e
n
is
d
iv
id
ed
in
to
s
in
g
le
s
eq
u
en
ce
s
tr
u
ctu
r
e
an
d
e
v
er
y
am
in
o
-
ac
id
e
m
b
ed
d
in
g
ex
tr
ac
tio
n
tak
es
p
lace
.
Fu
r
th
e
r
m
o
r
e
,
th
e
p
r
o
tein
s
h
av
in
g
m
u
lti
-
c
h
ain
s
eq
u
en
ce
s
tr
u
ctu
r
e
,
th
e
p
r
o
tei
n
em
b
e
d
d
in
g
s
a
r
e
ac
h
iev
ed
in
th
e
s
im
ilar
way
,
i.
e.
,
b
y
a
v
er
ag
in
g
am
in
o
-
ac
id
e
m
b
ed
d
in
g
s
an
d
is
r
ep
r
esen
ted
u
s
in
g
∈
×
,
wh
er
e
r
ep
r
esen
ts
to
tal
p
r
o
te
in
-
ch
ain
s
.
Fin
ally
,
th
e
p
o
s
itio
n
em
b
e
d
d
in
g
s
r
ep
r
esen
ted
as
∈
×
ar
e
m
er
g
ed
with
to
ev
alu
ate
th
e
p
o
s
itio
n
o
f
ev
er
y
r
esid
u
e.
T
h
e
ch
ain
em
b
ed
d
in
g
s
∈
×
ar
e
f
u
r
th
er
m
er
g
ed
with
f
o
r
m
o
d
ellin
g
m
u
tli
-
ch
ain
s
eq
u
en
ce
s
tr
u
ctu
r
e
o
f
p
r
o
tein
.
T
h
e
r
ep
r
esen
ts
h
ig
h
est
to
tal
ch
ain
s
.
T
h
e
r
esid
u
al
em
b
ed
d
i
n
g
∈
×
ar
e
d
en
o
ted
u
s
in
g
(
8
)
.
I
n
(
8
)
,
′
∈
×
r
ep
r
esen
ts
th
e
ch
ain
-
em
b
ed
d
in
g
w
ith
r
esp
ec
t to
ch
ain
o
f
ev
e
r
y
r
esid
u
e.
=
(
+
+
′
)
(
8
)
3.
4
.
2
.
F
ea
t
ure
e
x
t
ra
ct
i
o
n
I
n
th
is
wo
r
k
,
1
-
d
im
en
s
io
n
C
NN
ap
p
r
o
ac
h
was
u
s
ed
f
o
r
e
x
tr
ac
tin
g
f
ea
tu
r
es
f
r
o
m
th
e
n
eig
h
b
o
r
i
n
g
r
esid
u
es
f
r
o
m
th
e
p
r
o
tein
-
lig
a
n
d
s
eq
u
e
n
ce
s
.
I
n
th
e
tr
ain
in
g
s
et,
95%
o
f
th
e
p
r
o
tein
-
lig
a
n
d
b
in
d
i
n
g
s
h
ad
≤
23
am
in
o
-
ac
id
r
esid
u
es.
Usi
n
g
th
is
,
th
e
C
NN
ap
p
r
o
ac
h
was
co
n
s
tr
u
cted
in
s
u
c
h
a
way
th
at
t
h
e
f
ea
tu
r
es
co
u
ld
b
e
ex
tr
ac
ted
f
r
o
m
th
e
p
r
o
tein
-
lig
an
d
b
in
d
in
g
s
eq
u
e
n
ce
.
I
n
t
h
is
wo
r
k
,
th
e
C
NN
h
ad
th
r
ee
b
lo
ck
s
wh
ich
wer
e
s
tack
ed
o
n
ea
c
h
lay
er
o
f
C
NN,
s
u
ch
th
at
f
ea
t
u
r
es
co
u
ld
b
e
ex
tr
ac
ted
in
a
h
ier
a
r
ch
al
wa
y
in
th
e
last
s
tack
ed
C
NN
lay
er
.
T
h
e
s
tack
ed
lay
e
r
h
ad
d
i
f
f
er
en
t
k
er
n
el
wid
th
s
as
p
r
esen
ted
in
Fig
u
r
e
2
,
i.e
.
,
7
,
5
a
n
d
3
with
d
ilatio
n
r
ate
o
f
3
,
2
an
d
1
r
es
p
ec
tiv
ely
.
I
n
C
NN,
th
e
c
o
n
v
o
lu
tio
n
al
k
er
n
el
×
×
tr
an
s
f
o
r
m
e
d
t
h
e
×
in
p
u
t
f
ea
tu
r
es
in
to
×
1
×
f
ea
tu
r
es
u
s
in
g
(
9
)
.
I
n
(
9
)
,
r
ep
r
esen
ts
th
e
d
ilatio
n
-
r
ate,
r
ep
r
esen
t
s
to
ta
l
ch
an
n
els an
d
r
ep
r
esen
ts
k
e
r
n
e
l
-
wid
th
.
[
]
=
∑
[
+
∙
]
[
]
=
1
(
9
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
7
,
No
.
1
,
J
an
u
ar
y
20
2
5
:
58
0
-
5
8
9
586
Fig
u
r
e
2
.
Ar
c
h
itectu
r
e
o
f
p
r
o
p
o
s
ed
C
NN
m
o
d
el
in
E
MSPLA
-
C
NN
3.
4
.
3
.
P
o
s
it
io
n
-
ba
s
ed
a
t
t
ent
io
n m
ec
ha
nis
m
T
h
e
E
MSPLA
-
C
NN
u
til
izes
t
h
e
co
n
v
o
lu
tio
n
lay
er
s
f
o
r
ca
p
tu
r
in
g
d
e
p
en
d
e
n
cy
am
o
n
g
n
ei
g
h
b
o
r
in
g
p
r
o
tein
-
lig
a
n
d
b
in
d
in
g
r
esid
u
es.
Nev
er
th
eless
,
b
y
u
s
in
g
s
tack
ed
lay
er
s
o
f
C
NN
o
r
b
y
u
s
in
g
m
o
r
e
k
e
r
n
el
-
wid
th
s
,
m
ax
im
u
m
p
r
o
tein
-
lig
an
d
b
i
n
d
in
g
d
ep
e
n
d
en
cies
ca
n
b
e
ex
tr
ac
te
d
.
Als
o
,
th
is
in
c
r
e
ases
th
e
co
s
t
wh
ile
co
m
p
u
tatio
n
.
Hen
ce
,
to
s
o
lv
e
th
is
is
s
u
e,
an
d
t
o
ca
p
t
u
r
e
m
ax
im
u
m
p
r
o
tein
-
lig
a
n
d
b
in
d
i
n
g
d
ep
en
d
e
n
cies,
th
is
wo
r
k
p
r
esen
ts
a
n
o
v
el
p
o
s
itio
n
-
b
ased
em
b
e
d
d
in
g
atten
tio
n
ap
p
r
o
ac
h
.
T
h
e
p
o
s
itio
n
-
b
ased
em
b
ed
d
in
g
atten
tio
n
h
elp
s
in
ex
t
r
ac
tin
g
m
ax
im
u
m
f
ea
tu
r
es
f
r
o
m
th
e
p
r
o
tein
-
lig
an
d
b
in
d
in
g
s
eq
u
en
ce
s
.
Hen
ce
,
th
e
wh
ich
r
ep
r
esen
ts
k
ey
an
d
∈
×
wh
ich
r
e
p
r
esen
ts
its
r
esp
ec
tiv
e
v
alu
e
is
o
b
tain
ed
f
r
o
m
th
e
C
NN
o
u
tp
u
t la
y
er
.
Fu
r
th
er
,
th
e
∈
×
wh
ich
d
e
f
in
es th
e
q
u
er
y
is
ev
alu
ated
u
s
in
g
(
1
0
)
.
=
(
+
+
)
(
10
)
W
h
er
e,
∈
×
r
ep
r
esen
ts
th
e
em
b
ed
d
in
g
s
wh
ich
co
n
s
is
t
o
f
p
r
o
tei
n
-
lig
an
d
em
b
ed
d
in
g
s
wit
r
esp
ec
t
to
ea
ch
ch
ain
-
em
b
e
d
d
in
g
r
esid
u
e.
T
h
e
r
elatio
n
s
h
ip
am
o
n
g
ev
er
y
p
r
o
tein
-
lig
a
n
d
p
o
s
itio
n
an
d
n
ex
t
p
r
o
tein
-
lig
an
d
p
o
s
itio
n
is
ev
alu
ated
u
s
in
g
(
1
1
)
.
T
h
e
ℎ
ℎ
in
(
1
1
)
is
ev
alu
a
ted
u
s
in
g
(
1
2
)
.
T
h
e
atten
tio
n
co
m
p
o
n
en
t
in
(
1
2
)
is
ev
alu
ated
u
s
in
g
(
1
3
)
.
ℎ
(
,
,
)
=
(
ℎ
1
,
ℎ
2
,
…
,
ℎ
)
(
11
)
ℎ
=
(
ℎ
,
ℎ
,
ℎ
)
(
12
)
(
,
,
)
=
(
√
)
(
13
)
I
n
(
1
2
)
,
ℎ
b
elo
n
g
s
to
×
,
ℎ
b
elo
n
g
s
to
×
,
ℎ
b
elo
n
g
s
to
×
an
d
ℎ
b
elo
n
g
s
to
×
wh
ich
ar
e
co
n
s
id
er
as
p
ar
am
eter
m
atr
ix
an
d
r
ep
r
esen
ts
to
tal
h
ea
d
s
p
r
esen
t
in
ℎ
.
T
h
e
f
in
al
o
u
tp
u
t
ac
h
iev
ed
b
y
th
e
(
1
3
)
is
f
u
r
t
h
er
u
s
ed
as
an
i
n
p
u
t to
th
e
f
ee
d
-
f
o
r
war
d
-
n
etwo
r
k
(
FF
N)
wh
ich
is
f
u
lly
co
n
n
ec
ted
.
T
h
e
FF
N
co
m
p
r
is
es
o
f
two
b
i
-
lin
ea
r
tr
a
n
s
f
o
r
m
ati
o
n
h
av
in
g
Gau
s
s
ian
-
E
r
r
o
r
L
in
ea
r
-
Un
it
ac
tiv
atio
n
f
u
n
ctio
n
i
n
b
etwe
en
o
f
FF
N
lay
er
s
.
(
)
=
(
1
+
1
)
2
+
2
(
14
)
Fu
r
th
er
,
th
e
n
o
r
m
aliza
tio
n
a
n
d
r
esid
u
al
r
elatio
n
s
h
ip
ar
e
a
p
p
li
ed
to
ac
h
iev
e
b
etter
ac
ce
ler
ati
o
n
f
o
r
c
o
m
p
u
tatio
n
an
d
f
o
r
s
tab
le
g
r
a
d
ien
ts
.
T
h
e
f
in
al
o
u
t
p
u
t
is
ac
h
iev
ed
as
∈
×
wh
ich
r
ep
r
esen
ts
th
e
in
f
o
r
m
a
t
io
n
b
etwe
en
th
e
d
e
p
en
d
e
n
cies o
f
p
r
o
tein
-
lig
an
d
s
eq
u
en
ce
s
.
3.
4
.
4
.
Cla
s
s
if
ier
T
h
e
E
MSPLA
-
C
NN
u
s
e
s
th
e
f
in
al
o
u
tp
u
t
∈
×
f
o
r
p
r
e
d
ictio
n
o
f
th
e
p
r
o
tein
-
lig
an
d
b
in
d
in
g
s
b
y
m
e
r
g
in
g
th
e
p
r
o
t
ein
-
lig
an
d
e
m
b
ed
d
in
g
s
an
d
.
T
h
is
is
d
o
n
e
u
s
in
g
(
1
5
)
.
T
h
e
class
if
icatio
n
b
lo
ck
co
n
s
is
ts
o
f
co
n
n
ec
ted
la
y
er
s
with
4
-
h
id
ee
n
lay
er
s
.
Fu
r
th
er
,
th
e
o
u
tp
u
t
is
u
s
ed
f
o
r
f
i
n
al
p
r
ed
ictio
n
an
d
class
if
icatio
n
o
f
ev
er
y
p
o
s
itio
n
o
f
p
r
o
tein
-
lig
an
d
s
eq
u
en
ce
s
as
b
in
d
er
o
r
n
o
n
-
b
in
d
e
r
.
T
h
e
o
u
tp
u
t
is
ev
alu
ated
u
s
in
g
R
MSE
wh
ich
is
ev
alu
ated
u
s
in
g
(
1
6
)
.
I
n
(
1
6
)
,
̂
ar
e
p
r
ed
icted
v
alu
es,
ar
e
o
b
s
er
v
ed
v
alu
es,
an
d
i
s
to
tal
o
b
s
er
v
atio
n
s
.
Usi
n
g
t
h
is
p
r
o
p
o
s
ed
E
MSPLA
-
C
NN
ap
p
r
o
ac
h
,
t
h
e
r
esu
lts
wer
e
ev
alu
ated
,
wh
ich
ar
e
d
is
cu
s
s
ed
in
th
e
n
ex
t sectio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
MS
P
LA
fo
r
a
cc
u
r
a
te
fea
tu
r
e
mo
lecu
la
r
ex
tr
a
ctio
n
fr
o
m
p
r
o
tein
-
lig
a
n
d
i
n
tera
ctio
n
s
(
S
r
in
i
d
h
i Ku
lka
r
n
i V
)
587
=
⊕
(
15
)
(
)
=
(
1
+
1
)
2
+
2
(
16
)
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
d
ev
el
o
p
m
en
t
o
f
th
e
p
r
o
te
in
-
lig
an
d
i
n
ter
ac
tio
n
p
r
ed
ictio
n
s
y
s
tem
was
ac
co
m
p
lis
h
ed
u
s
in
g
b
o
th
Py
th
o
n
an
d
C
#
p
r
o
g
r
am
m
in
g
lan
g
u
ag
es,
lev
er
a
g
in
g
th
eir
r
e
s
p
ec
tiv
e
s
tr
en
g
th
s
to
cr
ea
te
a
r
o
b
u
s
t
an
d
e
f
f
icien
t
s
o
lu
tio
n
.
Py
th
o
n
,
k
n
o
wn
f
o
r
its
ex
ten
s
iv
e
lib
r
ar
ies
an
d
f
r
a
m
ewo
r
k
s
in
d
ata
p
r
o
ce
s
s
in
g
an
d
ML
,
p
lay
ed
a
cr
u
cial
r
o
le
in
th
is
wo
r
k
.
L
i
b
r
ar
ies
s
u
ch
as
Nu
m
Py
,
Pan
d
as,
an
d
SciPy
f
ac
ilit
ated
d
ata
m
an
ip
u
latio
n
an
d
an
aly
s
is
,
wh
ile
DL
f
r
am
ew
o
r
k
s
lik
e
T
en
s
o
r
Flo
w
an
d
Py
T
o
r
ch
wer
e
u
tili
ze
d
f
o
r
m
o
d
el
tr
ain
in
g
an
d
ev
alu
atio
n
.
Ad
d
itio
n
ally
,
c
h
e
m
in
f
o
r
m
atics
to
o
ls
lik
e
R
DKit
wer
e
em
p
lo
y
e
d
to
h
an
d
le
m
o
lecu
lar
d
ata.
On
th
e
o
th
er
h
an
d
,
C
#
was
in
s
tr
u
m
en
tal
in
in
teg
r
atin
g
th
ese
m
o
d
el
s
in
to
lar
g
er
s
o
f
twar
e
s
y
s
tem
s
,
en
s
u
r
i
n
g
ef
f
icien
t
d
ep
lo
y
m
e
n
t
an
d
s
ca
lab
ilit
y
th
r
o
u
g
h
th
e
.
NE
T
C
o
r
e
f
r
am
ewo
r
k
.
C
#
lib
r
ar
ies
s
u
ch
as
ML
.
NE
T
an
d
Acc
o
r
d
.
NE
T
wer
e
u
s
ed
f
o
r
ML
task
s
an
d
s
cien
tific
co
m
p
u
tatio
n
s
with
in
th
e
ap
p
licatio
n
en
v
ir
o
n
m
en
t.
T
h
is
d
u
al
-
lan
g
u
ag
e
ap
p
r
o
ac
h
,
co
m
b
in
in
g
Py
th
o
n
’
s
f
lex
i
b
ilit
y
an
d
co
m
p
r
eh
e
n
s
iv
e
ML
ec
o
s
y
s
tem
with
C
#
’
s
p
er
f
o
r
m
an
ce
an
d
in
teg
r
ati
o
n
ca
p
ab
ilit
ies,
r
esu
lted
in
a
p
o
wer
f
u
l
s
y
s
tem
f
o
r
p
r
ed
ict
in
g
p
r
o
tein
-
lig
a
n
d
in
ter
ac
tio
n
s
.
T
h
e
h
ar
d
war
e
r
e
q
u
ir
em
en
ts
in
clu
d
e
an
I
n
tel
C
o
r
e
i7
p
r
o
ce
s
s
o
r
,
1
6
GB
o
f
R
AM
,
an
d
at
leas
t
5
0
0
GB
o
f
f
r
ee
s
to
r
ag
e,
with
a
d
e
d
icate
d
GPU
r
ec
o
m
m
en
d
ed
f
o
r
DL
task
s
.
T
h
e
s
o
f
twar
e
r
eq
u
ir
em
en
ts
en
co
m
p
ass
W
in
d
o
ws
1
1
,
alo
n
g
s
id
e
Py
th
o
n
v
er
s
io
n
3
.
8
o
r
h
i
g
h
er
an
d
.
NE
T
C
o
r
e
3
.
1
o
r
h
i
g
h
er
f
o
r
C
#
.
T
h
is
co
m
p
r
eh
e
n
s
iv
e
s
etu
p
e
n
s
u
r
ed
th
at
th
e
s
y
s
tem
is
b
o
th
ef
f
e
ctiv
e
an
d
s
ca
lab
le,
ca
p
ab
le
o
f
h
an
d
lin
g
c
o
m
p
lex
co
m
p
u
tatio
n
al
task
s
in
th
e
r
ea
l
m
o
f
p
r
o
tein
-
lig
a
n
d
in
ter
ac
tio
n
s
.
T
h
e
ev
alu
atio
n
o
f
th
e
E
MSPLA
-
C
NN
wa
s
d
o
n
e
u
s
in
g
PDB
b
in
d
d
ataset
[
2
5
]
.
T
h
e
PDB
b
in
d
v
.
2
0
2
0
d
ataset
in
clu
d
es
b
in
d
in
g
-
s
ite
in
f
o
r
m
atio
n
f
o
r
1
9
,
4
4
3
co
m
p
lex
es
o
f
p
r
o
tein
-
lig
an
d
th
at
h
a
v
e
b
ee
n
id
e
n
tifie
d
th
r
o
u
g
h
ex
p
e
r
im
en
tatio
n
.
T
h
e
in
f
o
r
m
atio
n
is
av
ailab
le
in
b
o
th
MO
L
2
an
d
PDB
f
ile
f
o
r
m
a
ts
.
T
h
e
ev
alu
atio
n
was
d
o
n
e
u
s
in
g
R
MSE
s
co
r
e
.
Als
o
,
co
m
p
ar
is
o
n
was
d
o
n
e
with
cu
r
r
e
n
t
s
tate
-
of
-
a
r
t
-
wo
r
k
s
.
T
h
e
r
esu
lts
ar
e
s
h
o
wn
in
Fig
u
r
e
3
an
d
i
n
T
ab
le
1
.
T
h
e
Fig
u
r
e
3
an
d
T
ab
le
1
p
r
esen
ts
a
co
m
p
a
r
is
o
n
o
f
v
a
r
io
u
s
m
o
d
els
b
ased
o
n
th
eir
R
MSE
in
p
r
ed
ictin
g
p
r
o
tein
-
li
g
an
d
i
n
ter
ac
tio
n
s
,
wh
er
e
a
lo
wer
R
MSE
in
d
icate
s
b
etter
p
r
ed
ictiv
e
ac
cu
r
ac
y
.
Am
o
n
g
t
h
e
m
o
d
els
lis
ted
,
th
e
p
r
o
p
o
s
ed
E
MSPLA
-
C
NN
s
tan
d
s
o
u
t
with
th
e
lo
w
est
R
MSE
o
f
0
.
6
7
,
s
u
g
g
esti
n
g
it
h
as
th
e
h
ig
h
est
p
r
ed
ictiv
e
ac
cu
r
ac
y
.
T
h
is
m
o
d
el
u
tili
ze
s
C
N
Ns
to
ca
p
tu
r
e
co
m
p
lex
s
p
atial
f
ea
tu
r
es
o
f
p
r
o
tein
-
lig
an
d
in
t
er
ac
tio
n
s
,
h
ig
h
lig
h
tin
g
th
e
ef
f
ec
tiv
en
ess
o
f
DL
tech
n
iq
u
es.
SQM2
.
2
0
,
a
n
o
th
e
r
r
ec
en
t
m
o
d
el
f
r
o
m
2
0
2
4
,
also
d
em
o
n
s
tr
ates
h
ig
h
ac
cu
r
ac
y
with
an
R
MSE
o
f
0
.
6
9
,
in
d
icatin
g
s
ig
n
if
ican
t
im
p
r
o
v
em
e
n
ts
o
v
e
r
ea
r
lier
m
o
d
els.
I
n
co
n
tr
ast,
A
u
to
Do
ck
Vin
a,
d
esp
ite
b
ein
g
a
2
0
2
4
m
o
d
el,
s
h
o
ws
th
e
h
ig
h
est
R
MSE
o
f
2
.
8
5
,
in
d
ic
atin
g
lo
wer
p
r
ed
i
ctiv
e
p
er
f
o
r
m
an
ce
wh
ich
is
d
u
e
to
its
m
eth
o
d
o
lo
g
y
.
Oth
er
m
o
d
els
lik
e
MSGG
(
2
0
2
0
)
a
n
d
SEGSA
-
DT
A
(
2
0
2
3
)
s
h
o
w
m
o
d
er
ate
p
e
r
f
o
r
m
an
ce
with
R
MSE
s
o
f
1
.
2
7
an
d
1
.
3
1
9
,
r
esp
ec
tiv
ely
,
wh
ile
I
GN
an
d
R
F
-
Sco
r
e,
b
o
th
f
r
o
m
2
0
2
4
,
h
a
v
e
s
lig
h
tly
h
ig
h
er
R
MSE
s
o
f
1
.
8
2
an
d
1
.
8
9
.
Dee
p
DT
A
(
2
0
2
4
)
,
a
DL
-
b
ase
d
m
o
d
el,
p
e
r
f
o
r
m
s
r
ea
s
o
n
ab
ly
well
with
an
R
M
SE
o
f
1
.
3
4
b
u
t
d
o
es
n
o
t
m
atch
th
e
ac
cu
r
ac
y
o
f
SQM2
.
2
0
an
d
E
MSPLA
-
C
NN.
T
h
ese
r
es
u
lts
in
d
icate
th
at
n
ewe
r
m
o
d
els,
esp
ec
ially
th
o
s
e
in
co
r
p
o
r
atin
g
a
d
v
an
ce
d
ML
an
d
DL
tec
h
n
iq
u
es,
te
n
d
to
h
av
e
b
etter
p
r
ed
ictiv
e
p
er
f
o
r
m
an
ce
.
T
h
e
im
p
r
o
v
em
e
n
ts
s
ee
n
in
m
o
d
els
lik
e
SQ
M2
.
2
0
an
d
E
MSPLA
-
C
NN
u
n
d
er
s
co
r
e
th
e
im
p
o
r
tan
ce
o
f
co
n
tin
u
ed
in
n
o
v
atio
n
in
co
m
p
u
tatio
n
al
m
eth
o
d
o
lo
g
ies
to
en
h
an
ce
th
e
ac
cu
r
ac
y
an
d
r
elia
b
ilit
y
o
f
p
r
o
tein
-
lig
an
d
in
ter
ac
tio
n
p
r
e
d
ictio
n
s
.
Fig
u
r
e
3
.
R
MSE
s
co
r
es a
ch
iev
ed
b
y
d
if
f
er
e
n
t m
o
d
els u
s
in
g
PDB
b
in
d
d
ataset
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
7
,
No
.
1
,
J
an
u
ar
y
20
2
5
:
58
0
-
5
8
9
588
T
ab
le
1
.
R
MSE
s
co
r
es c
o
m
p
ar
is
o
n
with
cu
r
r
en
t state
-
of
-
ar
t
wo
r
k
s
an
d
E
MSPLA
ap
p
r
o
ac
h
M
o
d
e
l
s,
r
e
f
e
r
e
n
c
e
,
y
e
a
r
R
M
S
E
M
S
G
G
,
[
1
7
]
,
2
0
2
0
1
.
2
7
S
EG
S
A
_
D
TA
,
[
2
2
]
,
2
0
2
3
1
.
3
1
9
S
Q
M
2
.
2
0
,
[
2
4
]
,
2
0
2
4
0
.
6
9
A
u
t
o
D
o
c
k
V
i
n
a
,
[
2
6
]
,
2
0
2
4
2
.
8
5
I
G
N
,
[
2
6
]
,
2
0
2
4
1
.
8
2
RF
-
S
c
o
r
e
,
[
2
6
]
,
2
0
2
4
1
.
8
9
D
e
e
p
D
TA
,
[
2
6
]
,
2
0
2
4
1
.
3
4
EM
S
P
LA
-
C
N
N
,
[
P
r
o
p
o
s
e
d
]
,
2
0
2
4
0
.
6
7
5.
CO
NCLU
SI
O
N
I
n
th
is
wo
r
k
,
th
is
wo
r
k
d
ev
elo
p
ed
an
d
e
v
alu
ated
th
e
E
MSPLA
-
C
NN,
a
n
o
v
el
p
r
o
tein
-
lig
a
n
d
f
ea
tu
r
e
ex
tr
ac
tio
n
s
y
s
tem
th
at
c
o
m
b
i
n
es
C
NNs
an
d
a
p
o
s
itio
n
-
b
as
ed
atten
tio
n
m
ec
h
an
is
m
.
T
h
e
E
MSPLA
wa
s
u
s
ed
f
o
r
id
e
n
tify
in
g
th
e
s
im
ilar
ity
b
etwe
en
th
e
p
r
o
tein
s
an
d
C
NN
was
u
s
ed
to
ex
tr
ac
t
lo
ca
l
f
ea
tu
r
es
f
r
o
m
n
eig
h
b
o
r
in
g
r
esid
u
es
f
r
o
m
t
h
e
p
r
o
t
ein
s
eq
u
en
ce
s
.
B
y
lev
er
a
g
in
g
th
e
s
tr
en
g
th
s
o
f
Py
t
h
o
n
an
d
C
#
,
t
h
e
s
y
s
tem
b
en
ef
ited
f
r
o
m
th
e
ex
ten
s
iv
e
ML
lib
r
ar
ies
in
Py
th
o
n
an
d
th
e
in
teg
r
atio
n
an
d
p
e
r
f
o
r
m
an
ce
ca
p
ab
ilit
ies
o
f
C
#
.
T
h
e
E
MSPLA
ap
p
r
o
ac
h
u
til
ized
th
e
PDB
b
in
d
v
.
2
0
2
0
d
ataset,
an
d
th
r
o
u
g
h
r
ig
o
r
o
u
s
p
r
ep
r
o
ce
s
s
in
g
an
d
in
n
o
v
ativ
e
m
o
d
elin
g
tech
n
iq
u
es,
th
e
E
MSPL
A
ac
h
iev
ed
s
ig
n
if
ican
t
im
p
r
o
v
e
m
en
ts
in
p
r
ed
ictiv
e
ac
cu
r
ac
y
,
as
ev
id
en
ce
d
b
y
an
R
MSE
o
f
0
.
6
7
.
T
h
e
E
MSPLA
-
C
NN
m
o
d
el
’
s
ab
ilit
y
to
ca
p
tu
r
e
b
o
th
lo
ca
l
an
d
lo
n
g
-
d
is
tan
ce
d
ep
en
d
e
n
cies
in
p
r
o
tein
s
eq
u
e
n
c
es,
f
ac
ilit
ated
b
y
th
e
C
NN
an
d
atten
tio
n
m
o
d
u
les,
u
n
d
er
s
c
o
r
es
its
r
o
b
u
s
tn
es
s
an
d
e
f
f
ec
tiv
en
ess
.
Ad
d
itio
n
ally
,
th
e
E
MSPLA
co
n
tr
i
b
u
tes
to
p
r
ec
is
e
s
tr
u
ctu
r
al
s
im
ilar
ity
a
s
s
es
s
m
en
ts
,
f
u
r
th
e
r
en
h
an
cin
g
th
e
m
o
d
el
’
s
p
r
ed
ict
iv
e
p
er
f
o
r
m
a
n
ce
.
T
h
e
r
esu
lts
i
n
d
icate
th
at
in
te
g
r
atin
g
ad
v
a
n
ce
d
DL
tech
n
iq
u
es
with
tr
ad
itio
n
al
s
eq
u
en
ce
alig
n
m
en
t
m
eth
o
d
s
ca
n
s
u
b
s
tan
tially
im
p
r
o
v
e
th
e
ac
c
u
r
ac
y
o
f
p
r
o
tein
-
lig
an
d
in
ter
ac
tio
n
p
r
ed
ictio
n
s
.
T
h
is
w
o
r
k
n
o
t o
n
ly
s
ets
a
n
ew
b
en
ch
m
ar
k
in
p
r
ed
ictiv
e
p
er
f
o
r
m
an
c
e
b
u
t
also
p
r
o
v
id
es
a
s
ca
lab
le
an
d
ef
f
icien
t
f
r
am
e
wo
r
k
f
o
r
f
u
tu
r
e
d
ev
elo
p
m
en
ts
in
co
m
p
u
tatio
n
al
b
i
o
lo
g
y
.
C
o
n
tin
u
ed
in
n
o
v
atio
n
in
co
m
p
u
tatio
n
al
m
eth
o
d
o
lo
g
ies,
as
d
em
o
n
s
tr
ated
h
e
r
e,
is
cr
u
cial
f
o
r
ad
v
an
cin
g
th
e
u
n
d
er
s
tan
d
in
g
an
d
p
r
ed
ictio
n
o
f
p
r
o
tein
-
lig
an
d
i
n
ter
ac
tio
n
s
.
Fo
r
th
e
f
u
tu
r
e
wo
r
k
,
th
e
wo
r
k
will
b
e
ex
te
n
d
ed
f
o
r
p
r
ed
ictin
g
an
d
ev
alu
atin
g
th
e
p
r
o
tein
-
lig
an
d
b
in
d
in
g
af
f
in
ity
.
RE
F
E
R
E
NC
E
S
[
1
]
K
.
A
.
C
a
r
p
e
n
t
e
r
a
n
d
R
.
B
.
A
l
t
m
a
n
,
“
D
a
t
a
b
a
ses
o
f
l
i
g
a
n
d
-
b
i
n
d
i
n
g
p
o
c
k
e
t
s
a
n
d
p
r
o
t
e
i
n
-
l
i
g
a
n
d
i
n
t
e
r
a
c
t
i
o
n
s
,
”
C
o
m
p
u
t
a
t
i
o
n
a
l
a
n
d
S
t
r
u
c
t
u
r
a
l
B
i
o
t
e
c
h
n
o
l
o
g
y
J
o
u
r
n
a
l
,
v
o
l
.
2
3
,
p
p
.
1
3
2
0
–
1
3
3
8
,
M
a
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
s
b
j
.
2
0
2
4
.
0
3
.
0
1
5
.
[
2
]
R
.
M
o
r
r
i
s,
K
.
A
.
B
l
a
c
k
,
a
n
d
E
.
J.
S
t
o
l
l
a
r
,
“
U
n
c
o
v
e
r
i
n
g
p
r
o
t
e
i
n
f
u
n
c
t
i
o
n
:
f
r
o
m
c
l
a
ssi
f
i
c
a
t
i
o
n
t
o
c
o
m
p
l
e
x
e
s,
”
Ess
a
y
s
i
n
Bi
o
c
h
e
m
i
st
r
y
,
v
o
l
.
6
6
,
n
o
.
3
,
p
p
.
2
5
5
–
2
8
5
,
A
u
g
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
4
2
/
e
b
c
2
0
2
0
0
1
0
8
.
[
3
]
R
.
I
.
Te
l
e
a
n
u
,
A
.
-
G
.
N
i
c
u
l
e
sc
u
,
E.
R
o
z
a
,
O
.
V
l
a
d
â
c
e
n
c
o
,
A
.
M
.
G
r
u
m
e
z
e
sc
u
,
a
n
d
D
.
M
.
Te
l
e
a
n
u
,
“
N
e
u
r
o
t
r
a
n
smi
t
t
e
r
s
—
k
e
y
f
a
c
t
o
r
s
i
n
n
e
u
r
o
l
o
g
i
c
a
l
a
n
d
n
e
u
r
o
d
e
g
e
n
e
r
a
t
i
v
e
d
i
s
o
r
d
e
r
s
o
f
t
h
e
c
e
n
t
r
a
l
n
e
r
v
o
u
s
s
y
s
t
e
m
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
M
o
l
e
c
u
l
a
r
S
c
i
e
n
c
e
s
,
v
o
l
.
2
3
,
n
o
.
1
1
,
p
.
5
9
5
4
,
M
a
y
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
i
j
ms
2
3
1
1
5
9
5
4
.
[
4
]
H
.
L
u
e
t
a
l
.
,
“
R
e
c
e
n
t
a
d
v
a
n
c
e
s
i
n
t
h
e
d
e
v
e
l
o
p
m
e
n
t
o
f
p
r
o
t
e
i
n
–
p
r
o
t
e
i
n
i
n
t
e
r
a
c
t
i
o
n
s
m
o
d
u
l
a
t
o
r
s
:
m
e
c
h
a
n
i
s
ms
a
n
d
c
l
i
n
i
c
a
l
t
r
i
a
l
s,
”
S
i
g
n
a
l
T
r
a
n
s
d
u
c
t
i
o
n
a
n
d
T
a
rg
e
t
e
d
T
h
e
ra
p
y
,
v
o
l
.
5
,
n
o
.
1
,
S
e
p
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
3
9
2
-
0
2
0
-
0
0
3
1
5
-
3.
[
5
]
X
.
D
u
e
t
a
l
.
,
“
I
n
si
g
h
t
s
i
n
t
o
p
r
o
t
e
i
n
–
l
i
g
a
n
d
i
n
t
e
r
a
c
t
i
o
n
s
:
me
c
h
a
n
i
sms
,
m
o
d
e
l
s,
a
n
d
met
h
o
d
s
,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
M
o
l
e
c
u
l
a
r
S
c
i
e
n
c
e
s
,
v
o
l
.
1
7
,
n
o
.
2
,
p
.
1
4
4
,
Ja
n
.
2
0
1
6
,
d
o
i
:
1
0
.
3
3
9
0
/
i
j
ms
1
7
0
2
0
1
4
4
.
[
6
]
P
.
O
j
e
d
a
-
M
a
y
e
t
a
l
.
,
“
D
y
n
a
mi
c
c
o
n
n
e
c
t
i
o
n
b
e
t
w
e
e
n
e
n
z
y
ma
t
i
c
c
a
t
a
l
y
si
s a
n
d
c
o
l
l
e
c
t
i
v
e
p
r
o
t
e
i
n
m
o
t
i
o
n
s
,
”
B
i
o
c
h
e
m
i
st
ry
,
v
o
l
.
6
0
,
n
o
.
2
8
,
p
p
.
2
2
4
6
–
2
2
5
8
,
Ju
l
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
2
1
/
a
c
s.
b
i
o
c
h
e
m.
1
c
0
0
2
2
1
.
[
7
]
K
.
D
z
o
b
o
,
“
Th
e
r
o
l
e
o
f
n
a
t
u
r
a
l
p
r
o
d
u
c
t
s
a
s
so
u
r
c
e
s
o
f
t
h
e
r
a
p
e
u
t
i
c
a
g
e
n
t
s
f
o
r
i
n
n
o
v
a
t
i
v
e
d
r
u
g
d
i
sc
o
v
e
r
y
,
”
C
o
m
p
re
h
e
n
si
v
e
Ph
a
rm
a
c
o
l
o
g
y
,
p
p
.
4
0
8
–
4
2
2
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
B
9
7
8
-
0
-
12
-
8
2
0
4
7
2
-
6
.
0
0
0
4
1
-
4.
[
8
]
R
.
F
r
i
e
d
ma
n
,
“
C
o
m
p
u
t
a
t
i
o
n
a
l
s
t
u
d
i
e
s
o
f
p
r
o
t
e
i
n
–
d
r
u
g
b
i
n
d
i
n
g
a
f
f
i
n
i
t
y
c
h
a
n
g
e
s
u
p
o
n
m
u
t
a
t
i
o
n
s
i
n
t
h
e
d
r
u
g
t
a
r
g
e
t
,
”
WIRE
s
C
o
m
p
u
t
a
t
i
o
n
a
l
Mo
l
e
c
u
l
a
r
S
c
i
e
n
c
e
,
v
o
l
.
1
2
,
n
o
.
1
,
A
u
g
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
2
/
w
c
ms.1
5
6
3
.
[
9
]
N
.
V
e
r
ma
e
t
a
l
.
,
“
S
S
n
e
t
:
a
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
f
o
r
p
r
o
t
e
i
n
-
l
i
g
a
n
d
i
n
t
e
r
a
c
t
i
o
n
p
r
e
d
i
c
t
i
o
n
,”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
M
o
l
e
c
u
l
a
r
S
c
i
e
n
c
e
s
,
v
o
l
.
2
2
,
n
o
.
3
,
p
.
1
3
9
2
,
Ja
n
.
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
i
j
ms
2
2
0
3
1
3
9
2
.
[
1
0
]
H
.
W
a
n
g
,
“
P
r
e
d
i
c
t
i
o
n
o
f
p
r
o
t
e
i
n
–
l
i
g
a
n
d
b
i
n
d
i
n
g
a
f
f
i
n
i
t
y
v
i
a
d
e
e
p
l
e
a
r
n
i
n
g
mo
d
e
l
s
,
”
Bri
e
f
i
n
g
s
i
n
b
i
o
i
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
2
5
,
n
o
.
2
,
Ja
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
9
3
/
b
i
b
/
b
b
a
e
0
8
1
.
[
1
1
]
Y
.
W
a
n
g
,
Q
.
J
i
a
o
,
J
.
W
a
n
g
,
X
.
C
a
i
,
W
.
Z
h
a
o
,
a
n
d
X
.
C
u
i
,
“
P
r
e
d
i
c
t
i
o
n
o
f
p
r
o
t
e
i
n
-
l
i
g
a
n
d
b
i
n
d
i
n
g
a
f
f
i
n
i
t
y
w
i
t
h
d
e
e
p
l
e
a
r
n
i
n
g
,
”
C
o
m
p
u
t
a
t
i
o
n
a
l
a
n
d
S
t
ru
c
t
u
r
a
l
Bi
o
t
e
c
h
n
o
l
o
g
y
J
o
u
r
n
a
l
,
v
o
l
.
2
1
,
p
p
.
5
7
9
6
–
5
8
0
6
,
Ja
n
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
s
b
j
.
2
0
2
3
.
1
1
.
0
0
9
.
[
1
2
]
E
.
E
l
b
a
sa
n
i
,
S
.
N
.
N
j
i
m
b
o
u
o
m,
T.
-
J.
O
h
,
E
.
-
H
.
K
i
m
,
H
.
L
e
e
,
a
n
d
J.
-
D
.
K
i
m
,
“
G
C
R
N
N
:
g
r
a
p
h
c
o
n
v
o
l
u
t
i
o
n
a
l
r
e
c
u
r
r
e
n
t
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
c
o
m
p
o
u
n
d
–
p
r
o
t
e
i
n
i
n
t
e
r
a
c
t
i
o
n
p
r
e
d
i
c
t
i
o
n
,
”
BM
C
b
i
o
i
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
2
2
,
n
o
.
S
5
,
N
o
v
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
8
6
/
s
1
2
8
5
9
-
0
2
2
-
0
4
5
6
0
-
x.
[
1
3
]
A
.
D
h
a
k
a
l
,
C
.
M
c
K
a
y
,
J.
J.
Ta
n
n
e
r
,
a
n
d
J.
C
h
e
n
g
,
“
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
i
n
t
h
e
p
r
e
d
i
c
t
i
o
n
o
f
p
r
o
t
e
i
n
–
l
i
g
a
n
d
i
n
t
e
r
a
c
t
i
o
n
s
:
r
e
c
e
n
t
a
d
v
a
n
c
e
s a
n
d
f
u
t
u
r
e
d
i
r
e
c
t
i
o
n
s
,
”
Bri
e
f
i
n
g
s
i
n
Bi
o
i
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
2
3
,
n
o
.
1
,
N
o
v
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
9
3
/
b
i
b
/
b
b
a
b
4
7
6
.
[
1
4
]
W
.
Je
o
n
a
n
d
D
.
K
i
m,
“
F
P
2
V
EC
:
a
n
e
w
m
o
l
e
c
u
l
a
r
f
e
a
t
u
r
i
z
e
r
f
o
r
l
e
a
r
n
i
n
g
mo
l
e
c
u
l
a
r
p
r
o
p
e
r
t
i
e
s
,
”
Bi
o
i
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
3
5
,
n
o
.
2
3
,
p
p
.
4
9
7
9
–
4
9
8
5
,
M
a
y
2
0
1
9
,
d
o
i
:
1
0
.
1
0
9
3
/
b
i
o
i
n
f
o
r
m
a
t
i
c
s/
b
t
z
3
0
7
.
[
1
5
]
K
.
C
h
o
u
d
h
a
r
y
e
t
a
l
.
,
“
R
e
c
e
n
t
a
d
v
a
n
c
e
s
a
n
d
a
p
p
l
i
c
a
t
i
o
n
s
o
f
d
e
e
p
l
e
a
r
n
i
n
g
m
e
t
h
o
d
s
i
n
ma
t
e
r
i
a
l
s
sc
i
e
n
c
e
,
”
n
p
j
C
o
m
p
u
t
a
t
i
o
n
a
l
Ma
t
e
r
i
a
l
s,
v
o
l
.
8
,
n
o
.
1
,
A
p
r
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
2
4
-
0
2
2
-
0
0
7
3
4
-
6.
[
1
6
]
Y
.
S
i
e
t
a
l
.
,
“
D
e
e
p
r
e
p
r
e
s
e
n
t
a
t
i
o
n
l
e
a
r
n
i
n
g
o
f
p
a
t
i
e
n
t
d
a
t
a
f
r
o
m
e
l
e
c
t
r
o
n
i
c
h
e
a
l
t
h
r
e
c
o
r
d
s
(
EH
R
)
:
a
s
y
st
e
ma
t
i
c
r
e
v
i
e
w
,
”
J
o
u
r
n
a
l
o
f
Bi
o
m
e
d
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
1
1
5
,
p
.
1
0
3
6
7
1
,
M
a
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
b
i
.
2
0
2
0
.
1
0
3
6
7
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
MS
P
LA
fo
r
a
cc
u
r
a
te
fea
tu
r
e
mo
lecu
la
r
ex
tr
a
ctio
n
fr
o
m
p
r
o
tein
-
lig
a
n
d
i
n
tera
ctio
n
s
(
S
r
in
i
d
h
i Ku
lka
r
n
i V
)
589
[
1
7
]
S
.
W
a
n
g
,
Z.
Li
,
S
.
Zh
a
n
g
,
M
.
Ji
a
n
g
,
X
.
W
a
n
g
,
a
n
d
Z.
W
e
i
,
“
M
o
l
e
c
u
l
a
r
p
r
o
p
e
r
t
y
p
r
e
d
i
c
t
i
o
n
b
a
s
e
d
o
n
a
mu
l
t
i
c
h
a
n
n
e
l
s
u
b
st
r
u
c
t
u
r
e
g
r
a
p
h
,
”
I
E
EE
A
c
c
e
ss
,
v
o
l
.
8
,
p
p
.
1
8
6
0
1
–
1
8
6
1
4
,
J
a
n
.
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
a
c
c
e
ss.
2
0
2
0
.
2
9
6
8
5
3
5
.
[
1
8
]
D
.
D
e
n
g
,
X
.
C
h
e
n
,
R
.
Z
h
a
n
g
,
Z.
Le
i
,
X
.
W
a
n
g
,
a
n
d
F
.
Z
h
o
u
,
“
X
G
r
a
p
h
B
o
o
st
:
e
x
t
r
a
c
t
i
n
g
g
r
a
p
h
n
e
u
r
a
l
n
e
t
w
o
r
k
-
b
a
se
d
f
e
a
t
u
r
e
s
f
o
r
a
b
e
t
t
e
r
p
r
e
d
i
c
t
i
o
n
o
f
mo
l
e
c
u
l
a
r
p
r
o
p
e
r
t
i
e
s
,
”
J
o
u
r
n
a
l
o
f
C
h
e
m
i
c
a
l
I
n
f
o
rm
a
t
i
o
n
a
n
d
Mo
d
e
l
i
n
g
,
v
o
l
.
6
1
,
n
o
.
6
,
p
p
.
2
6
9
7
–
2
7
0
5
,
M
a
y
2
0
2
1
,
d
o
i
:
1
0
.
1
0
2
1
/
a
c
s.j
c
i
m
.
0
c
0
1
4
8
9
.
[
1
9
]
J.
L
i
a
n
d
X
.
Ji
a
n
g
,
“
M
o
l
-
B
ER
T
:
a
n
e
f
f
e
c
t
i
v
e
m
o
l
e
c
u
l
a
r
r
e
p
r
e
se
n
t
a
t
i
o
n
w
i
t
h
B
E
R
T
f
o
r
m
o
l
e
c
u
l
a
r
p
r
o
p
e
r
t
y
p
r
e
d
i
c
t
i
o
n
,”
Wi
re
l
e
ss
C
o
m
m
u
n
i
c
a
t
i
o
n
s
a
n
d
M
o
b
i
l
e
C
o
m
p
u
t
i
n
g
,
v
o
l
.
2
0
2
1
,
p
p
.
1
–
7
,
S
e
p
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
1
/
7
1
8
1
8
1
5
.
[
2
0
]
Z.
Y
a
n
g
,
W
.
Zh
o
n
g
,
L.
Z
h
a
o
,
a
n
d
C
.
Y
.
-
C
.
Ch
e
n
,
“
M
G
r
a
p
h
D
TA
:
d
e
e
p
m
u
l
t
i
sc
a
l
e
g
r
a
p
h
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
e
x
p
l
a
i
n
a
b
l
e
d
r
u
g
–
t
a
r
g
e
t
b
i
n
d
i
n
g
a
f
f
i
n
i
t
y
p
r
e
d
i
c
t
i
o
n
,
”
C
h
e
m
i
c
a
l
S
c
i
e
n
c
e
,
v
o
l
.
1
3
,
n
o
.
3
,
p
p
.
8
1
6
–
8
3
3
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
3
9
/
d
1
s
c
0
5
1
8
0
f
.
[
2
1
]
N
.
W
e
n
,
G
.
L
i
u
,
J.
Z
h
a
n
g
,
R
.
Z
h
a
n
g
,
Y
.
F
u
,
a
n
d
X
.
H
a
n
,
“
A
f
i
n
g
e
r
p
r
i
n
t
s
b
a
s
e
d
m
o
l
e
c
u
l
a
r
p
r
o
p
e
r
t
y
p
r
e
d
i
c
t
i
o
n
met
h
o
d
u
si
n
g
t
h
e
B
ER
T
m
o
d
e
l
,
”
J
o
u
r
n
a
l
o
f
C
h
e
m
i
n
f
o
r
m
a
t
i
c
s
,
v
o
l
.
1
4
,
n
o
.
1
,
O
c
t
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
8
6
/
s
1
3
3
2
1
-
0
2
2
-
0
0
6
5
0
-
3.
[
2
2
]
Y
.
G
u
e
t
a
l
.
,
“
P
r
o
t
e
i
n
–
l
i
g
a
n
d
b
i
n
d
i
n
g
a
f
f
i
n
i
t
y
p
r
e
d
i
c
t
i
o
n
w
i
t
h
e
d
g
e
a
w
a
r
e
n
e
ss
a
n
d
s
u
p
e
r
v
i
s
e
d
a
t
t
e
n
t
i
o
n
,
”
i
S
c
i
e
n
c
e
,
v
o
l
.
2
6
,
n
o
.
1
,
p
p
.
1
0
5
8
9
2
–
1
0
5
8
9
2
,
J
a
n
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
s
c
i
.
2
0
2
2
.
1
0
5
8
9
2
.
[
2
3
]
L.
X
u
,
S
.
P
a
n
,
L
.
X
i
a
,
a
n
d
Z.
L
i
,
“
M
o
l
e
c
u
l
a
r
p
r
o
p
e
r
t
y
p
r
e
d
i
c
t
i
o
n
b
y
c
o
mb
i
n
i
n
g
LST
M
a
n
d
G
A
T,
”
Bi
o
m
o
l
e
c
u
l
e
s
,
v
o
l
.
1
3
,
n
o
.
3
,
p
p
.
5
0
3
–
5
0
3
,
M
a
r
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
b
i
o
m
1
3
0
3
0
5
0
3
.
[
2
4
]
A
.
P
e
c
i
n
a
,
J
.
F
a
n
f
r
l
í
k
,
M
.
Le
p
š
í
k
,
a
n
d
J.
Ř
e
z
á
č
,
“
S
Q
M
2
.
2
0
:
s
e
m
i
e
m
p
i
r
i
c
a
l
q
u
a
n
t
u
m
-
m
e
c
h
a
n
i
c
a
l
sc
o
r
i
n
g
f
u
n
c
t
i
o
n
y
i
e
l
d
s
D
F
T
-
q
u
a
l
i
t
y
p
r
o
t
e
i
n
–
l
i
g
a
n
d
b
i
n
d
i
n
g
a
f
f
i
n
i
t
y
p
r
e
d
i
c
t
i
o
n
s
i
n
m
i
n
u
t
e
s,
”
N
a
t
u
re
c
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
1
5
,
n
o
.
1
,
F
e
b
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
2
4
-
4
5
4
3
1
-
8.
[
2
5
]
B
e
g
i
n
n
e
r
’
s
G
u
i
d
e
t
o
t
h
e
P
D
B
b
i
n
d
D
a
t
a
b
a
s
e
(
v
.
2
0
2
0
)
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
:
/
/
w
w
w
.
p
d
b
b
i
n
d
.
o
r
g
.
c
n
/
d
o
w
n
l
o
a
d
/
p
d
b
b
i
n
d
_
2
0
2
0
_
i
n
t
r
o
.
p
d
f
[
2
6
]
J.
Li
e
t
a
l
.
,
“
Le
a
k
p
r
o
o
f
P
D
B
B
i
n
d
:
a
r
e
o
r
g
a
n
i
z
e
d
d
a
t
a
se
t
o
f
p
r
o
t
e
i
n
-
l
i
g
a
n
d
c
o
mp
l
e
x
e
s
f
o
r
m
o
r
e
g
e
n
e
r
a
l
i
z
a
b
l
e
b
i
n
d
i
n
g
a
f
f
i
n
i
t
y
p
r
e
d
i
c
t
i
o
n
,
”
A
r
X
i
v
,
p
.
a
r
X
i
v
:
2
3
0
8
.
0
9
6
3
9
v
2
,
M
a
y
2
0
2
4
,
A
c
c
e
sse
d
:
M
a
y
2
0
,
2
0
2
4
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
p
u
b
me
d
.
n
c
b
i
.
n
l
m.
n
i
h
.
g
o
v
/
3
7
6
4
5
0
3
7
/
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
S
r
in
id
h
i
K
u
lk
a
r
n
i
V
o
b
tai
n
e
d
h
is
b
a
c
h
e
lo
r
d
e
g
re
e
in
I
n
fo
rm
a
ti
o
n
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
fr
o
m
Visv
e
sv
a
ra
y
a
Tec
h
n
o
l
o
g
ica
l
Un
i
v
e
rsity
Be
lg
a
u
m
,
M
a
ste
rs
in
Co
m
p
u
te
r
S
c
ien
c
e
a
n
d
En
g
in
e
e
rin
g
fro
m
Visv
e
sv
a
ra
y
a
Tec
h
n
o
l
o
g
ica
l
Un
i
v
e
rsity
Be
lg
a
u
m
,
M
a
ste
r
o
f
Arts
in
S
a
n
sk
rit
fr
o
m
Ka
rn
a
tak
a
sta
te
o
p
e
n
Un
iv
e
rsit
y
,
Dip
l
o
m
a
in
Dv
a
it
a
Ve
d
a
n
ta
fro
m
Ti
ru
p
a
t
h
i
S
a
n
s
k
rit
Un
i
v
e
rsity
.
He
h
a
s
1
0
y
e
a
rs
o
f
tea
c
h
i
n
g
E
x
p
e
ri
e
n
c
e
a
t
u
n
d
e
rg
ra
d
u
a
te
lev
e
l
o
f
En
g
in
e
e
rin
g
.
He
h
a
s
g
u
i
d
e
d
m
o
re
th
a
n
1
5
P
r
o
jec
ts
a
n
d
8
m
i
n
i
p
ro
jec
ts
a
t
u
n
d
e
rg
ra
d
u
a
te
lev
e
l.
He
h
a
s
n
u
m
e
ro
u
s
re
se
a
rc
h
p
a
p
e
rs
in
h
is
c
r
e
d
it
i
n
n
a
ti
o
n
a
l
a
n
d
in
tern
a
ti
o
n
a
l
c
o
n
fe
re
n
c
e
.
He
h
a
s
a
lso
p
u
b
li
sh
e
d
a
p
a
p
e
r
i
n
re
fe
rre
d
jo
u
rn
a
l
.
He
h
a
s
a
ls
o
p
a
rti
c
ip
a
ted
i
n
se
v
e
ra
l
F
a
c
u
lt
y
De
v
e
lo
p
m
e
n
t
P
ro
g
ra
m
s
a
n
d
w
o
r
k
sh
o
p
s.
His
a
re
a
s
o
f
i
n
tere
st
i
n
c
l
u
d
e
s
o
c
ial
n
e
two
rk
s,
c
l
o
u
d
c
o
m
p
u
ti
n
g
,
Io
T
,
b
i
g
d
a
ta,
d
e
sig
n
a
n
d
a
n
a
l
y
sis
o
f
a
l
g
o
rit
h
m
s,
a
n
d
d
a
ta
stru
c
tu
re
s
.
He
h
a
s
t
h
e
h
o
n
o
rs
o
f
L
ife
M
e
m
b
e
r
o
f
IS
T
E,
CS
I
a
n
d
is
a
n
a
ss
istan
t
p
ro
f
e
ss
o
r
a
t
Jy
o
th
y
In
stit
u
te
o
f
Tec
h
n
o
l
o
g
y
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
su
d
h
a
su
re
n
0
5
@
g
m
a
il
.
c
o
m
.
Dr
.
G
a
n
e
sh
Dha
n
d
a
p
a
n
i
h
a
s
c
o
m
p
lete
d
h
is
M
CA
in
Bh
a
r
a
th
iar
Un
iv
e
rsity
Co
imb
a
to
re
,
M
.
P
h
il
a
t
M
.
S
U
n
iv
e
rsit
y
,
Ti
ru
n
e
lv
e
li
,
M
.
Tec
h
.
a
t
S
a
ty
a
b
h
a
m
a
Un
i
v
e
rsity
Ch
e
n
n
a
i
a
n
d
P
h
.
D.
in
Im
a
g
e
p
ro
c
e
ss
in
g
a
t
Bh
a
ra
th
iar
Un
i
v
e
rsity
.
He
h
a
s
2
6
+
y
e
a
rs
o
f
e
x
p
e
rien
c
e
in
tea
c
h
in
g
a
n
d
h
is
p
ro
fe
ss
io
n
a
l
e
x
c
e
ll
e
n
c
e
is
th
ro
u
g
h
re
su
lt
-
o
rien
ted
a
p
p
ro
a
c
h
,
h
a
rd
w
o
rk
,
se
lf
-
m
o
ti
v
a
ti
o
n
a
n
d
p
e
rse
v
e
ra
n
c
e
in
tea
c
h
i
n
g
c
a
rrier.
He
h
a
s
p
u
b
li
sh
e
d
p
a
p
e
rs
in
m
a
n
y
S
c
o
p
u
s
jo
u
rn
a
ls,
c
o
n
fe
re
n
c
e
s
a
n
d
UG
C
c
a
re
jo
u
r
n
a
ls.
His
a
r
e
a
s
o
f
re
se
a
rc
h
a
n
d
tea
c
h
in
g
a
re
ima
g
e
p
ro
c
e
ss
i
n
g
,
c
o
m
p
u
t
e
r
n
e
two
r
k
s,
c
ry
p
t
o
g
ra
p
h
y
a
n
d
c
lo
u
d
se
c
u
rit
y
,
so
ftwa
re
e
n
g
in
e
e
rin
g
,
a
n
d
m
a
c
h
i
n
e
lea
rn
in
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
g
a
n
e
sh
d
@g
m
a
il
.
c
o
m
.
Dr
.
K
u
r
e
e
c
k
a
l
V.
Ra
m
e
sh
is
th
e
p
ro
fe
ss
o
r
a
n
d
h
e
a
d
o
f
t
h
e
De
p
a
rtme
n
t
o
f
Bio
tec
h
n
o
lo
g
y
a
t
JA
IN
(De
e
m
e
d
-
to
-
b
e
Un
i
v
e
rsity
)
in
Be
n
g
a
lu
ru
.
He
sp
e
c
ialize
s
in
Ag
ricu
lt
u
ra
l
M
icro
b
io
l
o
g
y
a
n
d
Bio
in
f
o
rm
a
ti
c
s,
h
o
l
d
i
n
g
a
P
h
.
D
.
in
Ag
ricu
lt
u
ra
l
M
icro
b
io
l
o
g
y
fro
m
th
e
Un
iv
e
rsity
o
f
Ag
ricu
l
tu
ra
l
S
c
ien
c
e
s,
Ba
n
g
a
lo
re
(2
0
0
4
).
Wi
th
a
c
a
re
e
r
sp
a
n
n
in
g
o
v
e
r
two
d
e
c
a
d
e
s,
Dr.
Ra
m
e
sh
h
a
s
h
e
l
d
v
a
ri
o
u
s
a
c
a
d
e
m
ic
a
n
d
c
o
n
s
u
lt
i
n
g
p
o
siti
o
n
s,
in
c
l
u
d
i
n
g
ro
les
a
s
P
ro
fe
ss
o
r
a
n
d
C
o
o
r
d
in
a
t
o
r,
a
n
d
Lec
tu
re
r
a
t
p
r
o
m
in
e
n
t
i
n
sti
tu
ti
o
n
s
in
Be
n
g
a
lu
ru
.
His
re
se
a
rc
h
fo
c
u
se
s
o
n
c
o
m
p
u
tati
o
n
a
l
b
io
lo
g
y
,
p
a
rti
c
u
larl
y
in
3
D
m
o
d
e
li
n
g
o
f
p
ro
tei
n
s
fro
m
m
icro
b
ial
s
o
u
rc
e
s
su
c
h
a
s
TB
b
a
c
teria
a
n
d
S
ARS
-
C
o
V
-
2
,
d
o
c
k
in
g
n
o
v
e
l
p
h
y
t
o
c
h
e
m
ica
l
li
g
a
n
d
s
o
n
m
o
d
e
led
p
ro
tein
s,
a
n
d
in
sili
c
o
st
u
d
ies
o
f
th
e
str
u
c
tu
ra
l
a
n
d
fu
n
c
ti
o
n
a
l
d
iv
e
rsity
o
f
b
e
h
a
v
i
o
ra
l
g
e
n
e
s
in
so
c
ial
o
rg
a
n
ism
s
li
k
e
A
p
is
m
e
ll
ifera
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
k
v
.
ra
m
e
sh
@ja
in
u
n
i
v
e
rsity
.
a
c
.
i
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.