Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more ➡
Download
Standard view
Full view
of .
Add note
Save to My Library
Sync to mobile
Look up keyword
Like this
1Activity
×
0 of .
Results for:
No results containing your search query
P. 1
Computing with DNA

Computing with DNA

Ratings: (0)|Views: 145|Likes:
Published by slartibartfastibast
http://www.alterlab.org/teaching/BME303/Adleman.pdf
http://www.alterlab.org/teaching/BME303/Adleman.pdf

More info:

Published by: slartibartfastibast on Jan 10, 2013
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See More
See less

08/16/2013

pdf

text

original

 
C
omputer. The word conjuresup images of keyboards andmonitors. Terms like “ROM,”“RAM,” “gigabyte” and “megahertz”come to mind. We have grown accus-tomed to the idea that computationtakes place using electronic compo-nents on a silicon substrate.But must it be this way? The comput-er that you are using to read thesewords bears little resemblance to a PC.Perhaps our view of computation is toolimited. What if computers were ubiq-uitous and could be found in manyforms? Could a liquid computer exist inwhich interacting molecules performcomputations? The answer is yes. Thisis the story of the DNA computer.
Rediscovering Biology
M
y involvement in this story beganin 1993, when I walked into amolecular biology lab for the first time.Although I am a mathematician andcomputer scientist, I had done a bit of AIDS research, which I believed and stillbelieve to be of importance [see “Bal-anced Immunity,” by John Rennie;
Sci-entific American,
May 1993]. Unfor-tunately, I had been remarkably unsuc-cessful in communicating my ideas tothe AIDS research community. So, in aneffort to become a more persuasive ad-vocate, I decided to acquire a deeperunderstanding of the biology of HIV.Hence, the molecular biology lab. There,under the guidance of Nickolas Chelya-pov (now chief scientist in my own lab-oratory), I began to learn the methodsof modern biology.I was fascinated. With my own hands,I was creating DNA that did not existin nature. And I was introducing it intobacteria, where it acted as a blueprint forproducing proteins that would changethe very nature of the organism.During this period of intense learn-ing, I began reading the classic text
TheMolecular Biology of the Gene,
co-au-thored by James D. Watson of Watson-Crick fame. My concept of biology wasbeing dramatically transformed. Biolo-gy was no longer the science of thingsthat smelled funny in refrigerators (myview from undergraduate days in the1960s at the University of California atBerkeley). The field was undergoing arevolution and was rapidly acquiringthe depth and power previously associ-ated exclusively with the physical sci-ences. Biology was now the study of in-formation stored in DNA
strings of four letters:
A, T, G
and
C
for the basesadenine, thymine, guanine and cyto-sine
and of the transformations thatinformation undergoes in the cell. Therewas mathematics here!Late one evening, while lying in bedreading Watson’s text, I came to a de-scription of DNA polymerase. This isthe king of enzymes
the maker of life.Under appropriate conditions, given astrand of DNA, DNA polymerase pro-duces a second “Watson-Crick” com-plementary strand, in which every
C
isreplaced by a
G,
every
G
by a
C,
every
A
by a
and every
by an
A
. For ex-ample, given a molecule with the se-quence
CATGTC,
DNA polymerasewill produce a new molecule with thesequence
GTACAG.
The polymeraseenables DNA to reproduce, which inturn allows cells to reproduce and ulti-mately allows you to reproduce. For astrict reductionist, the replication of DNA by DNA polymerase is what lifeis all about.DNA polymerase is an amazing littlenanomachine, a single molecule that
Computing with DNA
54
Scientific American
August 1998
Computing with DNA
The manipulation of DNA to solvemathematical problems is redefining what is meant by “computation”
by Leonard M. Adleman
Copyright 1998 Scientific American, Inc.
 
“hops” onto a strand of DNA and slidesalong it, “reading” each base it passesand “writing” its complement onto anew, growing DNA strand. While lyingthere admiring this amazing enzyme, Iwas struck by its similarity to somethingdescribed in 1936 by Alan M. Turing,the famous British mathematician. Tur-ing
and, independently, Kurt Gödel,Alonzo Church and S. C. Kleene
hadbegun a rigorous study of the notion of “computability.” This purely theoreti-cal work preceded the advent of actualcomputers by about a decade and led tosome of the major mathematical resultsof the 20th century [see “UnsolvedProblems in Arithmetic,” by HowardDeLong;
Scientific American,
March1971; and “Randomness in Arithme-tic,” by Gregory J. Chaitin;
ScientificAmerican,
 July 1988].For his study, Turing had invented a“toy” computer, now referred to as aTuring machine. This device was notintended to be real but rather to be con-ceptual, suitable for mathematical in-vestigation. For this purpose, it had tobe extremely simple
and Turing suc-ceeded brilliantly. One version of hismachine consisted of a pair of tapesand a mechanism called a finite control,which moved along the “input” tapereading data while simultaneously mov-ing along the “output” tape reading andwriting other data. The finite controlwas programmable with very simple in-structions, and one could easily write aprogram that would read a string of 
A,T, C
and
G
on the input tape and writethe Watson-Crick complementary stringon the output tape. The similarities withDNA polymerase could hardly havebeen more obvious.But there was one important piece of information that made this similaritytruly striking: Turing’s toy computerhad turned out to be universal
simpleas it was, it could be programmed tocompute anything that was computableat all. (This notion is essentially the con-tent of the well-known “Church’s the-sis.”) In other words, one could programa Turing machine to produce Watson-Crick complementary strings, factornumbers, play chess and so on. This re-alization caused me to sit up in bed andremark to my wife, Lori, “Jeez, thesethings could compute.” I did not sleepthe rest of the night, trying to figure outa way to get DNA to solve problems.My initial thinking was to make aDNA computer in the image of a Tur-ing machine, with the finite control re-placed by an enzyme. Remarkably, es-sentially the same idea had been sug-gested almost a decade earlier by CharlesH. Bennet and Rolf Landauer of IBM[see “The Fundamental Physical Limitsof Computation”;
Scientific Ameri-can,
 July 1985]. Unfortunately, while anenzyme (DNA polymerase) was knownthat would make Watson-Crick comple-ments, it seemed unlikely that any exist-ed for other important roles, such asfactoring numbers.This brings up an important factabout biotechnologists: we are a com-munity of thieves. We steal from thecell. We are a long way from being able
Computing with DNA
    T    O    M    O    N    A    R    A    S    H    I    M    A
DNA MOLECULES
with their sequences of adenine, thymine, guanine and cytosine (repre-sented by the letters
 A, T, G
and
C
)
can beused to store information and to perform com-putations. The molecule shown here in color,
GCAGTCGGACTGGGCTATGTCCGA,
en-codes the solution to the sample HamiltonianPath Problem on the next page.
Scientific American
August 1998 55
Copyright 1998 Scientific American, Inc.
 
to create de novo miraculous molecularmachines such as DNA polymerase.Fortunately, three or four billion yearsof evolution have resulted in cells thatare full of wondrous little machines. Itis these machines, stolen from the cell,that make modern biotechnology possi-ble. But a molecular machine that wouldplay chess has apparently never evolved.So if I were to build a DNA computerthat could do something interesting, Iwould have to do it with the tools thatwere at hand. These tools were essen-tially the following:1.
Watson-Crick pairing.
As statedearlier, every strand of DNA has its Wat-son-Crick complement. As it happens,if a molecule of DNA in solution meetsits Watson-Crick complement, then thetwo strands will anneal
that is, twistaround each other to form the famousdouble helix. The strands are not cova-lently bound but are held together byweak forces such as hydrogen bonds. If a molecule of DNA in solution meets aDNA molecule to which it is not com-plementary (and has no long stretchesof complementarity), then the two mol-ecules will not anneal.2.
Polymerases.
Polymerases copy in-formation from one molecule into an-other. For example, DNA polymerasewill make a Watson-Crick complemen-tary DNA strand from a DNA template.In fact, DNA polymerase needs a “start
Computing with DNA
56
Scientific American
August 1998
C
onsider a map of cities connected by certain nonstopflights (
top right 
). For instance, in the example shown here,it is possible to travel directly from Boston to Detroit but not viceversa. The goal is to determine whether a path exists that willcommence at the start city (Atlanta), finish at the end city (De-troit) and pass through each of the remaining cities exactly once.In DNA computation, each city is assigned a DNA sequence(
 ACTTGCAG
for Atlanta) that can be thought of as a first name(
 ACTT 
) followed by a last name (
GCAG
). DNA flight numbers canthen be defined by concatenating the last name of the city of origin with the first name of the city of destination (
bottom right 
). The complementary DNA city names are the Watson-Crick complements of the DNA city names in which every
is replacedby a
G,
every
G
by a
C,
every
 A
by a
T,
and every
by an
 A.
(To sim-plify the discussion here, details of the 3
versus 5
ends of theDNA molecules have been omitted.) For this particular problem,only one Hamiltonian path exists,and it passes through Atlanta,Boston, Chicago and Detroit in thatorder. In the computation, thispath is represented by
GCAGTCG-GACTGGGCTATGTCCGA,
a DNA se-quence of length 24. Shown at theleft is the map with seven citiesand 14 nonstop flights used in theactual experiment.
L.M.A.
    S    L    I    M     F    I    L    M    S
STARTEND
ATLANTA TO BOSTONBOSTON TO CHICAGOCHICAGO TO DETROITCOMPLEMENT OF BOSTONCOMPLEMENT OF CHICAGO
GC
A
 A
 A
A
A  
  A
AA
G  
T  
   T
  T
 T
   T
 T
    T
C  
C  
C
 C
C  
  C
  C
   C
C  C  
  C
G  G  
G
G  
 G
 G
G  
  G
G    
G  
Hamiltonian Path Problem
CITYATLANTABOSTONCHICAGODETROITFLIGHTATLANTA-BOSTONATLANTA-DETROITBOSTON-CHICAGOBOSTON-DETROITBOSTON-ATLANTACHICAGO-DETROITDNA NAMEACTTGCAGTCGGACTGGGCTATGTCCGAGCAADNA FLIGHT NUMBERGCAGTCGGGCAGCCGAACTGGGCTACTGCCGAACTGACTTATGTCCGACOMPLEMENTTGAACGTCAGCCTGACCCGATACAGGCTCGTT
BostonAtlantaDetroitChicago
WATSON-CRICKANNEALING,in which
C
s pairwith
G
s and
 A
s joinwith
s, will result inDNA flight-numberstrands (shown here: At-lanta to Boston, Bostonto Chicago, and Chicagoto Detroit) being held end-to-end by strands encodingthe complementary DNA citynames (shown here: Bostonand Chicago).
Copyright 1998 Scientific American, Inc.

You're Reading a Free Preview

Download
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->