<s>
DNA	B-Application
computing	I-Application
is	O
an	O
emerging	O
branch	O
of	O
unconventional	B-Device
computing	I-Device
which	O
uses	O
DNA	O
,	O
biochemistry	O
,	O
and	O
molecular	O
biology	O
hardware	O
,	O
instead	O
of	O
the	O
traditional	O
electronic	O
computing	O
.	O
</s>
<s>
Research	O
and	O
development	O
in	O
this	O
area	O
concerns	O
theory	O
,	O
experiments	O
,	O
and	O
applications	O
of	O
DNA	B-Application
computing	I-Application
.	O
</s>
<s>
Since	O
the	O
initial	O
Adleman	O
experiments	O
,	O
advances	O
have	O
occurred	O
and	O
various	O
Turing	B-Architecture
machines	I-Architecture
have	O
been	O
proven	O
to	O
be	O
constructible	O
.	O
</s>
<s>
This	O
expanded	O
the	O
horizon	O
of	O
DNA	B-Application
computing	I-Application
into	O
the	O
realm	O
of	O
memory	O
technology	O
although	O
the	O
in	O
vitro	O
demonstrations	O
were	O
made	O
almost	O
after	O
a	O
decade	O
.	O
</s>
<s>
The	O
field	O
of	O
DNA	B-Application
computing	I-Application
can	O
be	O
categorized	O
as	O
a	O
sub-field	O
of	O
the	O
broader	O
DNA	O
nanoscience	O
field	O
started	O
by	O
about	O
a	O
decade	O
before	O
Len	O
Adleman	O
's	O
demonstration	O
.	O
</s>
<s>
In	O
1994	O
Leonard	O
Adleman	O
presented	O
the	O
first	O
prototype	O
of	O
a	O
DNA	B-Application
computer	I-Application
.	O
</s>
<s>
In	O
Adleman	O
's	O
experiment	O
,	O
the	O
Hamiltonian	O
Path	O
Problem	O
was	O
implemented	O
notationally	O
as	O
"	O
travelling	B-Algorithm
salesman	I-Algorithm
problem	I-Algorithm
"	O
.	O
</s>
<s>
In	O
2002	O
,	O
J	O
.	O
Macdonald	O
,	O
D	O
.	O
Stefanović	O
and	O
M	O
.	O
Stojanović	O
created	O
a	O
DNA	B-Application
computer	I-Application
able	O
to	O
play	O
tic-tac-toe	O
against	O
a	O
human	O
player	O
.	O
</s>
<s>
The	O
corresponding	O
bin	O
becomes	O
fluorescent	O
,	O
indicating	O
which	O
box	O
is	O
being	O
played	O
by	O
the	O
DNA	B-Application
computer	I-Application
.	O
</s>
<s>
One	O
of	O
the	O
challenges	O
of	O
DNA	B-Application
computing	I-Application
is	O
its	O
speed	O
.	O
</s>
<s>
Such	O
localized	O
DNA	B-Application
computing	I-Application
techniques	O
have	O
shown	O
to	O
potentially	O
reduce	O
the	O
computation	O
time	O
by	O
.	O
</s>
<s>
Subsequent	O
research	O
on	O
DNA	B-Application
computing	I-Application
has	O
produced	O
,	O
bringing	O
the	O
technology	O
one	O
step	O
closer	O
to	O
the	O
silicon-based	O
computing	O
used	O
in	O
(	O
for	O
example	O
)	O
PCs	B-Device
.	O
</s>
<s>
While	O
both	O
the	O
designs	O
face	O
some	O
issues	O
(	O
such	O
as	O
reaction	O
leaks	O
)	O
,	O
this	O
appears	O
to	O
represent	O
a	O
significant	O
breakthrough	O
in	O
the	O
field	O
of	O
DNA	B-Application
computing	I-Application
.	O
</s>
<s>
Using	O
strand	O
displacement	O
reactions	O
(	O
SRDs	O
)	O
,	O
reversible	O
proposals	O
are	O
presented	O
in	O
for	O
implementing	O
reversible	O
gates	O
and	O
circuits	O
on	O
DNA	B-Application
computers	I-Application
by	O
combining	O
DNA	B-Application
computing	I-Application
and	O
reversible	O
computing	O
techniques	O
.	O
</s>
<s>
This	O
paper	O
also	O
proposes	O
a	O
universal	O
reversible	O
gate	O
library	O
(	O
URGL	O
)	O
for	O
synthesizing	O
n-bit	O
reversible	O
circuits	O
on	O
DNA	B-Application
computers	I-Application
with	O
an	O
average	O
length	O
and	O
cost	O
of	O
the	O
constructed	O
circuits	O
better	O
than	O
the	O
previous	O
methods	O
.	O
</s>
<s>
The	O
most	O
fundamental	O
operation	O
in	O
DNA	B-Application
computing	I-Application
and	O
molecular	O
programming	O
is	O
the	O
strand	O
displacement	O
mechanism	O
.	O
</s>
<s>
Beside	O
simple	O
strand	O
displacement	O
schemes	O
,	O
DNA	B-Application
computers	I-Application
have	O
also	O
been	O
constructed	O
using	O
the	O
concept	O
of	O
toehold	O
exchange	O
.	O
</s>
<s>
This	O
class	O
of	O
DNA	B-Application
computers	I-Application
does	O
not	O
require	O
enzymes	O
or	O
any	O
chemical	O
capability	O
of	O
the	O
DNA	O
.	O
</s>
<s>
The	O
full	O
stack	O
for	O
DNA	B-Application
computing	I-Application
looks	O
very	O
similar	O
to	O
a	O
traditional	O
computer	O
architecture	O
.	O
</s>
<s>
This	O
opened	O
the	O
way	O
to	O
design	O
and	O
synthesis	O
of	O
biochemical	O
controllers	O
since	O
the	O
expressive	O
power	O
of	O
CRNs	O
is	O
equivalent	O
to	O
a	O
Turing	B-Architecture
machine	I-Architecture
.	O
</s>
<s>
Enzyme-based	O
DNA	B-Application
computers	I-Application
are	O
usually	O
of	O
the	O
form	O
of	O
a	O
simple	O
Turing	B-Architecture
machine	I-Architecture
;	O
there	O
is	O
analogous	O
hardware	O
,	O
in	O
the	O
form	O
of	O
an	O
enzyme	O
,	O
and	O
software	O
,	O
in	O
the	O
form	O
of	O
DNA	O
.	O
</s>
<s>
Benenson	O
,	O
Shapiro	O
and	O
colleagues	O
have	O
demonstrated	O
a	O
DNA	B-Application
computer	I-Application
using	O
the	O
FokI	O
enzyme	O
and	O
expanded	O
on	O
their	O
work	O
by	O
going	O
on	O
to	O
show	O
automata	O
that	O
diagnose	O
and	O
react	O
to	O
prostate	O
cancer	O
:	O
under	O
expression	O
of	O
the	O
genes	O
PPAP2B	O
and	O
GSTP1	O
and	O
an	O
over	O
expression	O
of	O
PIM1	O
and	O
HPN	O
.	O
</s>
<s>
DNA	O
nanotechnology	O
has	O
been	O
applied	O
to	O
the	O
related	O
field	O
of	O
DNA	B-Application
computing	I-Application
.	O
</s>
<s>
A	O
DX	O
array	O
has	O
been	O
demonstrated	O
whose	O
assembly	O
encodes	O
an	O
XOR	O
operation	O
;	O
this	O
allows	O
the	O
DNA	O
array	O
to	O
implement	O
a	O
cellular	O
automaton	O
which	O
generates	O
a	O
fractal	B-Application
called	O
the	O
Sierpinski	O
gasket	O
.	O
</s>
<s>
DNA	B-Application
computing	I-Application
is	O
a	O
form	O
of	O
parallel	B-Operating_System
computing	I-Operating_System
in	O
that	O
it	O
takes	O
advantage	O
of	O
the	O
many	O
different	O
molecules	O
of	O
DNA	O
to	O
try	O
many	O
different	O
possibilities	O
at	O
once	O
.	O
</s>
<s>
For	O
certain	O
specialized	O
problems	O
,	O
DNA	B-Application
computers	I-Application
are	O
faster	O
and	O
smaller	O
than	O
any	O
other	O
computer	O
built	O
so	O
far	O
.	O
</s>
<s>
Furthermore	O
,	O
particular	O
mathematical	O
computations	O
have	O
been	O
demonstrated	O
to	O
work	O
on	O
a	O
DNA	B-Application
computer	I-Application
.	O
</s>
<s>
DNA	B-Application
computing	I-Application
does	O
not	O
provide	O
any	O
new	O
capabilities	O
from	O
the	O
standpoint	O
of	O
computability	O
theory	O
,	O
the	O
study	O
of	O
which	O
problems	O
are	O
computationally	O
solvable	O
using	O
different	O
models	O
of	O
computation	O
.	O
</s>
<s>
if	O
the	O
space	O
required	O
for	O
the	O
solution	O
of	O
a	O
problem	O
grows	O
exponentially	O
with	O
the	O
size	O
of	O
the	O
problem	O
(	O
EXPSPACE	O
problems	O
)	O
on	O
von	B-Architecture
Neumann	I-Architecture
machines	I-Architecture
,	O
it	O
still	O
grows	O
exponentially	O
with	O
the	O
size	O
of	O
the	O
problem	O
on	O
DNA	O
machines	O
.	O
</s>
<s>
A	O
compiler	O
has	O
been	O
written	O
in	O
Perl	B-Language
.	O
</s>
<s>
The	O
slow	O
processing	O
speed	O
of	O
a	O
DNA	B-Application
computer	I-Application
(	O
the	O
response	O
time	O
is	O
measured	O
in	O
minutes	O
,	O
hours	O
or	O
days	O
,	O
rather	O
than	O
milliseconds	O
)	O
is	O
compensated	O
by	O
its	O
potential	O
to	O
make	O
a	O
high	O
amount	O
of	O
multiple	O
parallel	B-Operating_System
computations	I-Operating_System
.	O
</s>
<s>
However	O
,	O
it	O
is	O
much	O
harder	O
to	O
analyze	O
the	O
answers	O
given	O
by	O
a	O
DNA	B-Application
computer	I-Application
than	O
by	O
a	O
digital	O
one	O
.	O
</s>
