Table of Contents

Attestation: ……………………………………………………………………………………………………………………………………………..I

Dedication: ………………………………………………………………………………………………………………………………………….….II

Author’s Declaration: …………………………………………………………………………………………………………………………….III

Acknowledgement: …………………………………………………………………………………………………………………………….…IV

Abstract: …………………………………………………………………………………………………………………………………………………V

Table of Contents: …………………………………………………………………………………………………………………………………VI

Figures List: ……………………………..…………………………………………………………………………………………………………….IX

List of Abbreviation: ……………………………………………………………………………………………………………………………...XI

Chapter 1 Introduction:

1.1 Introduction: ........................................................................................................................................... 4

1.2 Basic Theory: ........................................................................................................................................... 6
1.3 What is DNA? .......................................................................................................................................... 8
1.3.1 Structure of DNA: ............................................................................................................................ 9
1.3.2 Structure of Nucleotide: ................................................................................................................... 9
1.3.3 Chemical Structure of a Nucleotide: .............................................................................................. 10
1.3.3.1 Phosphodiester Bond: ............................................................................................................. 10
1.3.3.2 Hydrogen Bond: ...................................................................................................................... 10
1.4 DNA Computing: ................................................................................................................................... 12
1.4.1 Bonding Rules of Nucleotides: ...................................................................................................... 13
1.4.2 Potential of DNA Based Computing:............................................................................................. 14
1.5 Silicon Based Computing: ...................................................................................................................... 14
1.6 Organization of Study: .......................................................................................................................... 17
Chapter 2 Related Work:

2.1 Introduction: ......................................................................................................................................... 19

2.2 Operations on DNA: .............................................................................................................................. 20
2.2.1 Synthesis: ....................................................................................................................................... 20

Page | 1
 2.2.2 Denaturing, Annealing and Ligation:............................................................................................. 20
2.2.3 Hybridization Separation: .............................................................................................................. 21
2.2.4 Gel Electrophoresis: ....................................................................................................................... 21
2.2.5 Primer Extension and PCR: ........................................................................................................... 22
2.2.6 Restriction Enzymes: ..................................................................................................................... 23
2.3 The Main Steps in DNA Computing are: ............................................................................................... 23
2.4 Mechanism for DNA Replication: .......................................................................................................... 24
2.5 Evolution in DNA Computing: ............................................................................................................... 28
2.6 Carbon Nano Tubes:.............................................................................................................................. 30
2.7 Why DNA: .............................................................................................................................................. 31
2.8 Encryption and Decryption Data Files Algorithm:................................................................................. 33
2.8.1 Algorithm for Encryption: ............................................................................................................. 33
2.8.2 Decryption Algorithm: ................................................................................................................... 35
2.9 Assumption of Study: ............................................................................................................................ 39
2.10 Problem Statement: ............................................................................................................................ 40
Chapter 3 Design and Analysis:

3.1 Research Methodology: ........................................................................................................................ 42

3.2 Design and Analysis:.............................................................................................................................. 42
3.3 Hard Combinatorial Problem: ............................................................................................................... 45
3.4 Process of Both Computing Techniques: .............................................................................................. 47
3.4.1 PROCESS COMPARISON ........................................................................................................... 47
3.5 Storing Data in Both Techniques:.......................................................................................................... 49
3.5.1 STORAGE COMPARISON .......................................................................................................... 50
3.5.2 Density of DNA: ........................................................................................................................ 52
3.6 Cost Comparison of Both Techniques: .................................................................................................. 53
3.6.1 COST COMPARISON .................................................................................................................. 54
3.7 Energy Consumption of Both Techniques:............................................................................................ 55
3.7.1 ENERGY CONSUMPTION COMPARISON: ............................................................................. 56
Chapter 4 Results:

4.1 Data Collection and Analysis: ................................................................................................................ 58

4.2 Results: .................................................................................................................................................. 58

Page | 2
 4.2.1 Process: .......................................................................................................................................... 58
4.2.2 Storage: .......................................................................................................................................... 62
4.2.3 Cost: ............................................................................................................................................... 64
4.2.4 Power Consumption: ...................................................................................................................... 65
Chapter 5 Conclusion and Future Work:

5.1 Conclusion: ............................................................................................................................................ 66

5.2 Future Work: ......................................................................................................................................... 66
References: ................................................................................................................................................. 68

Page | 3
1.1 Introduction:

“Necessity is the mother of invention” communication is basic human need, so they

started communication by messenger on foot, then with help of pigeons, smoke signals, horses,

water, train, telephone, mobile and internet. As the need increases the innovations and inventions

take place accordingly.

In the beginning, communication was too slow then it increased gradually and became

faster with the passage of time. The means of communication took shift from one way of

communication to another. The technology enhanced with the change in requirements, the time

duration for sending and receiving data decreased and the amount of data to send and receive

increased. At present, the communication technology is established at Global level; Global world

has its own needs, because today every action creates moment far away. To full-fill Global world

needs researchers presented different solutions for different circumstances.

Computation enhanced gradually according to needs, so did the speed of processors, and

storage capacity, the needs of globalization are not yet been full-filled, because there are a lot of

limitations in the traditional method of communication, processing, storage means in

computation. Researchers, experts and technicians intend to increase the number of processors

by decreasing the size of transistors in chip, but it does not full-fill Global world needs. The

author in [1] argued that if we doubled the number of processors it will not double the processing

speed and a stage will come that it will start decreasing the speed of processing by increasing

number of processors. Traditional computation has the limitations of cost, time and huge energy

power. It is very difficult to overcome these limitations.

Page | 4
 The author in [2] proposed the idea of molecular level computation means

Deoxyribonucleic acid (DNA) based computation. The study involved comprehensive analysis

of up-to-date biology and DNA research as well as covering the knowledge of computation and

mathematics. Since computing is a method to store information and performing some operations

on that stored information. If anyone wants to have device for computing or to make Watson

crick complement or to play game of chess, need of an accurate input in the storage and apply

some sequence of procedures/instructions are required. Cell is using DNA for storing

information, for storing the blueprint of life, for storing the recipe of life as well as, nature [2]

used such a vast storing capability item for storage purpose. Enzymes like polymerase and

ligases were being used for operating on it, this all is enough for computing.

The author in [3] select problem known as Hamiltonian path problem and solves it. He

judged that one gram (1gm) of DNA; when it is dry, would occupy one cubic centimetre volume

and that can store data as one trillion compact disks (CD). He argued that DNA provide massive

parallelism in processing; in one second 1014 DNA flights number were simultaneously

concatenated in only 1/50 of teaspoon of solution.

DNA is double stranded sequence of four bases. DNA support for self-replication and it

contains coding for production of proteins. It means that DNA contain such encryption of coding

which may help and guide to produce protein for cell production which needed for body growth

[4].

A form of DNA computing is DNA chip, which is for self-treatment of disease; another

form is in order to control their processes by replacing living cells to develop genetic computer

programs [5].

Page | 5
 One form is “combine living organisms with traditional technique of computing for the

reason that brain of living organisms has the ability to understand such problems which may not

be handled by silicon based computer” [5].

Looking at the potential applications of DNA in various fields of computation researchers

explore it to store data in it because of DNA storage capacity, density, scalability, ideal for long

term stability and archival of data [6].

The author in [7] argues that DNA‟s are the core data storage molecules in all alive cells

of living things, years of evolution declared it, by testing these data molecules and specific

enzymes that could make replica of the data in DNA molecules or can transmit this data to other

DNA molecules; is an alternative of using electrical impulses for bits representation of data. The

chemical assets of these particles are being used in the DNA computer, patterns of combination

are examined or growth of the molecules. DNA perform this with creation of enzymes, that is so

called the ‟software‟, being used for the desired calculation [7].

1.2 Basic Theory:

DNA (Deoxyribonucleic acid) carries virus‟s genetic specificity and has exact self-

duplication property [8].

DNA‟s chemical formula is now well established. A long chain in which, the backbone is

made up of phosphate and sugar groups, they were joined in a regular 3/5 phosphate links. Each

sugar is committed to nitrogenous base, four (4) classes of them are usually found in DNA. Two

(2) of these are purines named guanine and adenine; the others are pyrimidine thymine and

cytosine.

Page | 6
 It have been noted that the chain broke down and sequence of the regular inter nucleotide

linkage made up. In contrast, as we can determine, as far as the sequence of nucleotides, is

totally irregular. Thus, DNA has regular, and some irregular features [8].

Long thin fiber sedimentation, light scattering obtained by physiochemical analysis,

dissemination and viscosity measurements. The estimated molecular weight is between 5 X 10

and 107 approximately, currently 3 <104 nucleotides of the center. Each measuring DNA, in

view of the large number of relatively strong bond that shows vague forms of DNA: Two distinct

forms of DNA exist. About 30 per cent of water is found in a relative humidity of about 75% in

which the crystalline form structure „A‟ [8].

The fibers in high humidity increase the length by 30% and assume it structure „B‟,

increase occur due to taking more water. This structure has a less ordered form, and mercury

appears to be crystalline, that is, all are parallel individual molecules packed together, but are not

regularly arranged in space. Some of the characteristic features which distinguish the two forms

are presented. B is a reversible change and infrastructure is likely to be connected in a simple

manner [8].

The structure of DNA lies on two (2) polynucleotide chains parallel to each other in

opposite directions and both are twisted. Both chains fastened by hydrogen bonds between the

bases, every single base are jointed to companion base on the next chain. Pairs of bases are

definite; adenine goes with thymine, and guanine goes along with cytosine [9].

In DNA structure, to notice and realize combination of symmetry and pseudo symmetry

and repetition is important. Phosphate sugar backbone of both chemical and structurally,

periodically repeats. This argument implies that the sugar-phosphate groups are related to the

Page | 7
balance by a screw axis in this case, and the spine is a simple helix. Again, two separate sugar

phosphate by double circulation backbones, are related to each other by the balance in this case

the fiber axis perpendicular to the two fold rotation axis [9].

The base‟s arrangement however does not repeats, and only shows the pseudo balance.

That is fixed to the area occupied by a pair of bases, and constant regions are related to each

other by the symmetry. But allowed till there is use of the base‟s pairs and restriction is

abandoned on occurrence of base‟s pairs at any point [9].

Using two hydrogen bonds many different ways are there to create pairing of four

common bases. The only satisfactory way which allows the occurrence of a series of four bases

on chain is the described one, and will fit in the x-ray data. There are other structural factors in

favor of particular suggested pairs. X-ray diffraction indicates that a substantial part of the DNA

has to be in the double helix form. This is really unfortunate way for determining ratio of the

DNA is in this configuration.

The titration arc and the investigative data propose that the bulk of the bases are paired. It

looks like that the particles are pleated in its biological form and possibly there is infrequent

regions where the configuration is improved [9].

1.3 What is DNA?

The principal and controller molecule of every cell is DNA [7]. Vital data is there in it

that converts to each consecutive generation. It synchronizes the creation of itself and also others

such is proteins. Changes in it may results serious consequences and destroying it beyond

healing the cell will die. Making changes in the DNA of cells in any multicellular living thing

produces deviations in the appearances of species. To change or evolve species, the natural

selection makes changes according to natural criteria but this took long period.

Page | 8
 DNA is nucleic acid and contains data, founds in nucleus in the cell of humans; RNA

(ribonucleic acid) is also nucleic acid. In DNA there are codons which guides the cell (along

with RNA) for production of new proteins that determines all biological traits goes generations

to generation. All these functions occur under a key which lies in DNA‟s molecular structure [7].

1.3.1 Structure of DNA:

Nucleotides are basic building blocks of DNA and nucleotides are four different parts.

DNA is large string of these four parts, which were called bases. DNA is also defined as that it is

a polymer. Number of small molecules strung together and forms DNA and these small

molecules are monomers of nucleotide [7].

1.3.2 Structure of Nucleotide:

It is the unit of nucleic acids as shown in figure 1. Each nucleotide consists of three (3)

components that are [7]:

1) Sugar deoxyribose and carbon atoms: one is sugar five (5) carbon atoms hydroxyl group (OH)

committed to three (3´) carbon.

2) Another group is phosphates which are linked to five (5') carbons.

3) The last one is nitrogenous base which is attached to one (1') carbon.

Page | 9
 Fig. 1. Nucleotides [7].

1.3.3 Chemical Structure of a Nucleotide:

Purines and pyrimidine‟s are two classes of nitrogen bases [7].

Adenine „A‟, Thymine „T‟, Guanine „G‟ and Cytosine „C‟ are bases.

Linking of Nucleotides:

Phosphodiester bond and Hydrogen bond are two ways for linking DNA monomers.

1.3.3.1 Phosphodiester Bond:

5 phosphate groups of 1 nucleotide are joined with the 3 OH group of the other strong

bond directionally; five to three (5 - 3) or three to five (3 - 5) both are possible [7].

1.3.3.2 Hydrogen Bond:

The bases of one nucleotide interacts with each other in a pair see figure 2. Basically

hydrogen is not strong chemical bond, it take place between hydrogen atoms and more

electronegative atoms such as fluorine and oxygen. Participating atoms can be located on either

Page | 10
case; it could be in adjacent nucleotide in a same molecule or also can be on adjacent nucleotides

on different DNA strands.

Hydrogen bonds act like covalent and ionic bonds, means that it does not involves in the

exchange or sharing of electrons. It can be easily formed and broken because they occur over

short distances. They also can stabilize molecule [7].

Fig. 2. DNA Strands Sugar, Phosphates and Nitrogen Bases [7].

In DNA there are strands which are formed of two substances that is sugar and phosphate

portions of the nucleotides. The middle parts of those stands are built of nitrogenous bases. The

Page | 11
nitrogenous bases on the 2 strands of DNA make pairs. The pairs are Purine with pyrimidine A

with T and G with C; they are apprehended together by weak bond of hydrogen.

“A – T” 2 hydrogen bonds

“C – G” 3 hydrogen bonds [7]

Foundation for current philosophies of molecular biology is the data transferring ability

of DNA molecules [10]. In DNA the data is enclosed in the form of nucleotide bases sequence.

Hydrogen bond is in a matching technique to form double stranded molecules from single

stranded oligonucleotides.

The basic biological mechanism of the development has been occurred in 1970 with

computational implementation. The following are the developments splicing, recombination and

mutation, they are genetic operations. In 1994 the researcher put the proof of idea that

recombinant properties DNA can really use gigantic parallelism for solving problems properly

programmed into single DNA strand [10].

1.4 DNA Computing:

To perform computational operations on DNA only single strand of it requires [11]. Two

types of strands are there in the computer of DNA, one is instructions and the other is input data

strand. Due to the leaning of single its strand, to form a double-helical structure in the occasion

of their nucleotide structures being corresponding to each other. Instruction strands splice

together with input data strands to produce the wanted output data strand [11].

As we know that the instruction strands are generic nucleotide codes for several

operations. The Turing machine concept about computation is found matched in DNA

Page | 12
computers. All the standards are successfully achieved. Now it focuses on the use of massive

parallelism and also the allocation of small slices of a computing task to processing components.

DNA structure allows the components of the problem to be represented in similar form to

the binary code structure that is (1‟s and 0‟s); the most basic form of computer languages is

named machine language. Huge numbers of unique DNA strands are encoded in a quaternary

number system which uses four nucleotides A, C, T, G and signifies the possible solutions to a

problem. The probable solution strands are allowed for responding to the „problem‟ strands [11].

1.4.1 Bonding Rules of Nucleotides:

The following are bonding rules for nucleotides which exist in the DNA; complement

strands to one another will bond, yielding group of entire DNA particles that contains the

solutions. A number of processing steps are performed on the resulting mixture to figure out a

correct solution from all the options.

The capability of performing trillions of reactions to occur concurrently runs the

equivalent of massively parallel processing in silicon based computers. In which a huge number

of searches through pools of data for the answers are executed at the same time [11].

In [12] the Idea of molecular scale arithmetic‟s has been given. How two luminous

sensors can do a job as a half-adder, taking ions as input and change happens in fluorescence as

output. It has been reported recently that the molecular scale computation technique based on a

whole set of catalytic nucleic acid based logic gates, satisfies the early recommendations that

nucleic acid catalysts created over modular design can provide a new form of DNA based

computation.

Page | 13
 It has recently been concluded in the construction of the first silicomimetic mechanism

having capability of playing a sort of tic-tac-toe game against a human opponent. A strategically

work is done for creating a group of three enhanced deoxiribozyme based logic gates. This takes

2 inputs and makes a decision to produce either one of the two possible outputs. This array was

first artificial decision making enzymatic linkage that mimics the half-adder and was the first

example of multi input artificial. Multi output decision making networks with molecules as both

inputs and outputs [12].

1.4.2 Potential of DNA Based Computing:

DNA computing retains number of advantages over silicon based complements. Some

leading advantages of DNA computer are given below [11].

1st, a process with DNA computing can be performed energy efficient in contrast to traditional

computing technique. Efficiency is roughly recorded as one million times.

2nd, it has potential to store an astonishing quantity of material for a certain space.

3rd, the element constituents are too much in quantity and easily can found.

4th, DNA computing tolerates for enormously parallel computation. Most researchers and experts

consider that it has capability of performing huge parallel processing [11].

1.5 Silicon Based Computing:

Since the early ages the computation and calculations have been performed by human [7].

Changes in computational techniques occurred, from manual way of computation to mechanical

technique and in modern ages to electronic technique.

Page | 14
 By definition computation is a technique of getting input, perform some sort of

manipulations store it and produce output. The size of it tremendously decreased and processing

power increased. Computer contains silicon built integrated circuits [7].

For digital computer the requirement is that it has ability of using logical functions for

arithmetic operations [12]. Adders are the logical circuits that perform addition in the central

processing unit (CPU). The half adder adds only two binary digits because it has only the ability

to add two bits, but it is a key for creating full adder, which can add three bits and can be

cascaded to yield serial adders for adding multi bit integers. Therefore, to construct half adder for

newly technique of computation is very hard. But molecular half adder logic gate have been

construction and tested [12].

Switch is one of the most vital components in the production of electronic appliances

such is CPU or memory [13]. More than millions switches jointed for construction of computers,

in new computer the transistors act as electronic switches. Transistors controls the flow of data

that either to pass or not under conditions mentioned for it. After the development of integrated

circuits (ICs) it has been probable to placed huge amount of transistors on a piece of silicon. IC‟s

are the silicon chips or can say traditional computing technique.

IC‟s are made of pure piece of silicon; generally its size is less than one centimeter (1cm)

square and almost half millimeter (1/2 mm) thick. It consists of hundreds of thousands of micro

miniature electronic circuit constituents; mostly they are transistors packed and connected in

layers under the surface. Control function, logic, and/or memory functions can be performed by

these components [13].

Page | 15
 Creating complex circuit by traditional silicon material is easier and economical than

other material so computer chip is made of silicon. To build qualitative insulator with silicon is

also very easy, add little oxygen to create silicon oxide. Accurate regulated voltage is required

for computational chip so that it manipulates data. Silicon was declared ideal reason behind is

that it can be prepared in form of either an effective and efficient insulator or semiconductors,

they are vital for regulatory of electrical current. It is also declared by experiment and researcher

study that silicon is economical resources with this aptitude. There are number of limitation in

silicon and researchers are eyeing for a better replacement [13].

In [14] components include in superscalar processor are discussed such are:

 Instruction cache: fetching and execution instructions are part of it.

 Instruction fetch unit: This unit fetches every clock cycle from instruction cache to fetch

buffer and the fetch unit retains support from memory management unit and also address

cache.

 Instruction decodes and renaming registers: The decoding of fetched instructions occurs

here and they are put in instruction buffer. The decode instructions contain reference of

architectural register, which is associated with renaming register.

 Issue unit: In instruction buffer the waiting instructions are examines by the issue unit

and assigns their number to functional unit.

 Reorder buffer: Order of the program is being stores for the issued instructions.

 Functional units: Transferring data between data cache and floating point or integer

register is performed in this register over one or more load/store units. Multimedia

operations are performed with single instruction by multimedia unit.

Page | 16
  Reservation stations: If the operands are not available then the instructions had to wait in

reservation station.

 Branch unit: Monitoring and updating of branch history table is performed in branch unit.

If any instruction is miss predicted it unrolls speculatively.

 Retire unit: The assurance of commit or retire are operated by retire unit. Although out of

program order functional unit may completed them.

 Physical registers: The physical register is architecture for floating point, integer and

rename floating point, integer.

 Data cache: This unit contains program data.

 Bus interface unit: To connect with environment the bus interface is used. It works like

secondary cache when it is not available on chip or it works like external memory [14].

1.6 Organization of Study:

The thesis is organized according to IEEE format; there are five chapters in this thesis

and each chapter contains number of topics.

The first chapter is Introductory includes Introduction of the study, Background

introducing the basics and origin of the study and also statement of the Problem is introduced in

chapter 1.

The 2nd chapter is named Literature Review; this chapter contains the data about work

done by number of researchers and different views of authors from very basic to advance level.

There are diagrams and tables to deliver and present the data efficiently. Explanatory notes are

Page | 17
there for properly conveying and marking the related study. Concrete, complete, correct data is

taken from different sources and presented gently. Chapter provides related data to the topic and

gives healthy base to chapter 3.

The 3rd chapter of the study contains the Design and Analysis part, contains diagrams and

calculations. The chapter contains comparative data and analysis performed with the help of the

proposed model in and mathematical calculations also involved in comparison process.

Chapter 4th is containing Results that are generated on base of designed methodology; the

results are presented with the help of graphs and figures.

5th chapter is named conclusion and future work; the chapter contain data about overall

study summary and also indicates to the future prospects of the study. Prediction of the author

based on the study is part of chapter. The chapter also contains the References of the data been

used in the study.

Page | 18
2.1 Introduction:

The computational skill of living thing has captivated researchers from last three or four

decades. They are focusing on implementation of living things in computational environment

[10]. There are some examples of it, artificial neural networks, genetic codes and cellular

automata. Their main focus is on the behavior simulating capability of natural computing devices

that founds in living things with the help of proper coding. So innovations are directional to

computation from biology point of view. Until the experiment of Adelman was not introduced

there was rare focus on implementation process of computational codes in biological things. The

fortunate thought was that it is only theoretical study and suggestions were all theoretical.

First the implementation proposal for DNA computing was motivated by logically

irreversible automata because of its capability of computation and the most important

thermodynamics. The objective was to decrease consumption of energy and proposal of

operations performed in DNA computing nearly thermal equilibrium [10].

The author realized after all the above mentioned information that enzymes have

transformative power and forecast role of enzyme‟s transformation in DNA based computation.

Chain reactions confirmed the computation of DNA, equilibrium in reactants and products, and

identical occurrence of reactions in both forward and backward direction helps in confirmation.

The author noticed that the means for achieving thermodynamic is this equilibrium

mechanism and also logical reversibility. These realizations and corresponding of reaction path

to the desired computation was trick in the proposed computation [10].

There came another proposal which consists of documentation for implementation of

traditional languages in this newly risen technique, it is basically called splicing system. Splicing

is double strands DNA molecules set; it is generated from the restriction enzyme activity and

Page | 19
represents over four nucleic acid alphabetic pairs. This study realizes that the enzymatic activity

retains computational power with the help of data storage capacity of DNA. A model for

computation in DNA is represented by splicing system, it provides path for implementing

traditional codes in in this new computing technique [10].

2.2 Operations on DNA:

The above mentioned number of models for DNA computation applies specific sequence

of operations on a set of strands; all of them commonly use these operations [15].

2.2.1 Synthesis:

Oligonucleotide has been synthesized to bring in order with the help of machine. Four

nucleotide bases in solution have been supplied to synthesizer, combined according to a user

entered sequence. Millions of copies are made by the instrument of requisite oligonucleotide and

placed them in a small vial contains solution [15].

2.2.2 Denaturing, Annealing and Ligation:

The solution is heated till the predefined temperature for strand composition; double

stranded DNA disbanded into single strands (denatured) [15].

Due to heating of solution the complementary bond of hydrogen between the strands

breaks up see in figure 3. This is weaker bond in contrast of covalent bond which occurs within

the strands; these strands are not damaged by the overall process.

The above process melts the solution and its reverse is Annealing, in this instead of

heating the opposite process mean the single strands solution is cooled down, which bonds the

complementary strands together.

Page | 20
 If any strands remains disconnected, not bonded and is single strand in the double

stranded DNA then it‟s been repaired by DNA ligase process [15].

Fig. 3. DNA Melting and Annealing [15].

2.2.3 Hybridization Separation:

The operation used in DNA computation for separation is so called hybridization. This

operation extracts any single strand from tube containing special short sequence; like, extract

those strands which match the following sequence ATGACT [15].

2.2.4 Gel Electrophoresis:

Computation with DNA technique requires sorting of strands by size and Gel

electrophoresis is the impressive technique to perform it. Basically Electrophoresis is defined as

that it is movement of charged particles in an electric field. So DNA moves towards positive pole

when it‟s been placed in electrical field because of negative charge it contains. The ratio of

particles migration in aqueous solution depends on the shape and electrical charge on it. Constant

charge on particles per unit length results in same speed of migration [15].

Page | 21
 The gel is a dense network of pores through which the particles travel necessarily see

figure 4.

Fig. 4. Gel Electrophoresis Photograph [15].

2.2.5 Primer Extension and PCR:

Polymerases of DNA perform several jobs; one of them is repairing and duplication of

DNA [15]. Given a short primer oligonucleotide, p, in the presence of nucleotide triphosphates,

the polymerase extends p (always in the 5 – 3 direction) if and only if p is bound to a longer

template oligonucleotide, t.

Another useful method of manipulating DNA is the Polymerase Chain Reaction (PCR).

PCR is a process that rapidly amplifies the amount of a particular molecule of DNA in a given

solution using primer extension by polymerase. The quantity of this molecule doubles with each

cycle of the reaction, giving an exponential growing to the number of strands. Most commonly

Page | 22
seen problem in DNA computation is how to read-out the final solution to a problem encoded as

a DNA strand, as the laboratory steps carried out it may result in quite dilute solution. PCR

solves this “needle in a haystack” problem, if a sought after molecule is present in the solution,

then it will be incredibly (exponentially) multiplied so that the volume of the solution will

“visibly” grow-this solves then the detection problem [15].

2.2.6 Restriction Enzymes:

Restriction endonucleases often referred to as restriction enzymes, distinguish a specific

sequence of DNA, called restriction site. Double-stranded DNA that contains restriction site

within its sequence is cut by the enzyme at that point [15].

2.3 The Main Steps in DNA Computing are:

Separate (T, s): given set T into the set + (T, s) of characters including character string s

have been separate by this operation, and set − (T, s) of character strings that do not contain

character string s also been separated [16].

Mix: sets T1 and T2 are been mixed into the union set T1 ∪ T2 using mix operation. This

operation corresponds to mixing test tubes T1 and T2.

Detect (T): operation returns „YES‟ if the test tube T is not empty, and „NO‟ if it is

empty. By the electrophoretic fluorescent method, the operation corresponds to an experimental

procedure that detects the existence of DNA molecules.

Amplify (T): amplify create multiple sets that are T1 and T2 with the same contents and

variables as the given set T. to amplify the amount of molecules polymerase chain reaction

(PCR) has been used [16].

Page | 23
2.4 Mechanism for DNA Replication:

The complementary nature suggests how it duplicates itself [8]. To imagine that how like

attracts like is difficult, and it has been suggested that self-duplication may involve the union of

each part with an opposite or complementary part. In the discussions it has generally been

suggested that protein and nucleic acid are complementary to each other and that self-replication

also involving the alternative syntheses of those two components.

It has also been proposed instead that the specificity of DNA self-replication is

accomplished without recourse to particular protein synthesis and that each of complementary

DNA chains serves as a template or mold for the formation onto itself of a new companion chain.

For this to take place the hydrogen bonds linking the complementary chains must breakdown and

the two chains unwind and separate from each other. It seems like that single chain may be itself

assume the helical form and serve as a mold onto which free nucleotides can link themselves by

forming hydrogen bonds [8].

Polymerization of the originators to form a new chain only occurs if the resulting chain

forms the proposed structure. This is possible because steric reasons would not allow monomers

"crystallized" onto the first chain to approach one another in such a way that they could be joined

together in a new chain. It is not obvious whether special enzymes are needed to carry out the

polymerization or whether the existing single helical chain may act effectively as an enzyme

could [8].

Recombinant DNA techniques have been developed for a wide class of operations on

DNA strands [17]. DNA operations were shown to be theoretically sufficient for universal

computation. Bio Molecular Computing (BMC) methods have been proposed to solve difficult

Page | 24
combinatorial search problems, such as the Hamiltonian path problem. Vast parallelism is

available in DNA to do the combinatorial search among large number of possible solutions.

Example is BMC method for breaking the Data Encryption Standard (DES). While these

methods for solving hard combinatorial search problems may succeed for fixed sized problems,

they are ultimately limited by their volume requirements, which may grow exponentially with

input size. However, BMC has many exciting features beyond pure combinatorial search. For

example, DNA is appealing media for data storage due to the very huge amounts of data storing

capability. They vastly exceed the storage capacities of any conventional electronic, magnetic or

optical media. A gram of DNA contains about one thousand and twenty one (1021) DNA bases,

or about one hundred and eight (108) Tera-bytes (TB). Hence, a few grams of DNA may have

the potential of storing all the data stored in the world [17].

Engineered DNA might be useful as a database medium for storing two broad classes of

data: (i) processed, biological sequences, and (ii) conventional data from binary, electronic

sources. Discussed method for fast and efficient associative searches within DNA databases is

using hybridization. Other BMC techniques might perform more refined database operations on

DNA data such as database join operations and various massively parallel operations on the

DNA data [17].

Between 1998 and 2001, for the first time some steps toward building cellular logic gates

were taken when logic gates were modeled and built a cellular inverter, an AND gate and two

other gates [11]. In the concentrations of two proteins the input bit and output bit are encoded for

sake of simplicity, named protein A and protein B. If the concentration of protein A is high, the

encoded input bit is a 1 in Figure 5; if the concentration is low; the encoded input bit is a 0 in

figure 6. Likewise, the focus of B tolerates to output bit.

Page | 25
 Fig. 5 Input Cellular Inverter [11].

Fig. 6 Output Cellular Inverter [11].

DNA loop has been created that contains 2 vital pieces one is the gene with codons for

producing protein „B‟ and a section of DNA to which protein „A‟ bonds.

For producing „B‟ protein the messenger RNA called a special molecule accumulates

itself beside the DNA, replicates the instructions of genes and carries them to the protein

Page | 26
producing industrial unit of cell. If „A‟ has high concentration then protein „A‟ molecules will

definitely bond to the loop of DNA and messenger RNA will be blocked from attaching to the

DNA. This does not allow the production cell to build protein „B‟. Protein B will be built in

abundance if the concentration of A is low. Hence, in the circuit, when A is low then B is high,

and vice versa [11].

The other gates are made in similar but with a little more complex ways. Engineers can

create range of exciting devices just by coupling the inverters together. In 2000 colleagues from

Boston University, made a memory device in E. coli shown in figure 7 out of two inverters for

which the output protein of one is the input protein of the other, and vice versa. In the same year

researchers from Rockefeller University in New York City made an oscillator in which 3

inverters are coupled in a loop subsequently that each output protein is the input protein of the

next inverter. In one test, a fluorescent protein became active whenever one of the proteins was

in its low state [11].

Fig. 7. E. Coli Bacteria [11].

Page | 27
2.5 Evolution in DNA Computing:

The first ever molecular logic gates were introduced in 1997 [7]. The genetic particles are

divided into fragments and the logic gates made up of DNA; detect these fragments as input

stream. To execute single output simply the splice of fragments are made. The study is aiming

that combining microchips with these logic gates will make breakthrough in this computational

technique [7].

DNA computer goes towards dry computation. The idea was implement and a glass plate

was been coated with thing layer of gold. Bonding of sulfur to gold occurs and linkages between

short DNA strands are made with sulfur atoms. Dunking procedure in DNA solution results in

self-assembly monolayer [7].

Main problem is divided into number of short and tiny problems, each part is coded in new

strand of molecules and bonding is allowed to offer possible solutions. After this the right answer

has been protected from such harmful attacks that occur chemically.

The new computing technique DNA based structure is focused for development, trying

for putting the diagnostics capability of different diseases. This computing technique can work in

fluids, like in blood samples and in inner side of body [7].

There is presented a MAYA-II DNA based computer. In its prototype complete tic-tac-

toe game was played under the logic gates operations that are molecular array of „YES‟ and

„AND‟ gates. A cell culture plate has been shown in figure 8. It has ability of performing moves

under the codon; the blue is human opponent of computer which is denoted by red squares [7].

Page | 28
 Fig. 8. MAYA - II [7].

More than 100 DNA circuits has been use in composition of MAYA-II. The previous

version was introduced in 2003. This was MAYA-I which could not perform all the required

moves of tic-tac-toe game. [7].

In view of today's high-tech consoles of games the experimental device looks nothing.

But it provides initial push to this new kind of computing. Study introduces MAYA-II computer

contains 9 cell structure wells. These cells were placed in a manner that looks like tic-tac-toe

grid. The well is given solution of coded DNA particles [7].

An autonomous device was introduced in 2001; that was DNA based and able to be

programmed and make calculations over it. It was experimented with the searching of even

number of 1‟s in list of binary digits. This device was enhanced in 2004.

An advance program was developed to enhance capabilities of bio molecular computers.

With the help of it the system became able to make logical decisions. A compiler for bridge

Page | 29
working between the code of DNA and computer programming language has been developed.

For example, „All men are mortal; Sajid is a man, so Sajid is mortal‟. A rule was fed and actual

data is confirmed by the computer correctly that sajid is mortal [7].

2.6 Carbon Nano Tubes:

In the view of above study it not looks like that there will be available any cheaper

material for building transistors other than silicon, it has been found after experiments and

researcher continuous effort that fast performer and small size component can create from

carbon. This will result in computers less consumer of energy [13].

Silicon is the most common semiconductor material used in transistors; it is economical

and easily handles to process, but it has its limitations. Reducing the size of transistor for

increasing the speed lead to other problems, the problems are; increased consumption of energy

and huge deviation in the properties of transistor. In this study they put the idea to use carbon

nanotube instead of silicon; it will result‟s in faster and small transistor. This nanotube is hollow

cylindrical shaped molecule. The diameter of it is about a nanometer (nm) and width is 1/50, 000

of human hair [13].

Another form was discussed that is semiconducting carbon nanotubes suggested that this

can be used in transistors but still there are unsolved problems in it. As overcome these problems

the transistors will be connected together and form circuit. These semiconductor nanotube

components can endure greater heat in contrast to traditional metal made components. It allows

working in more heated environment. [13].

Suggestions are there for using Graphite, shaped cylindrically nearly 1 nm diameter.

These light weight nanotubes are developed for different uses like fuel cells, television and

Page | 30
computer components. These applications can expand further. Scientists have found that other

materials besides carbon can form nanotubes [13].

2.7 Why DNA:

The author in [1] calculated that if F be the fraction of sequential calculations and (1-F) is

possibly parallelized fraction, then the maximum achievable speed by using P number of

processors is (1/F+ (1-F)/P).

In [1] the author calculated that if 90% processing or calculation is parallelized and 10%

remains sequential while 5 processors are used, it would roughly give you 3.6 maximum speed-

ups. If processors increased to 10 then it will speed up the processing up to 5.3 and if the

numbers of processors are increased to 20 then it will speed up calculations or processing to 6.9

which proves that doubling the hardware increase the speed only 30%. If 1000 processors are

used for calculations it will speed up 9.9 that mean a limit will come that the speed will start

decreasing [1].

To put the idea of DNA base computing in practice, the author in [3] selected a problem

for computation that is Hamiltonian path problem. The author assumed a map of cities connected

to each other with nonstop flights, for example “it is possible to travel directly from Boston to

Detroit but not vice versa. The goal is to determine whether a path exists that will initiate at the

starting city (Atlanta), finish at ending city (Detroit) and pass through each of the remaining

cities exactly one time”. Cities are given named in a manner that first name followed by last

name of the desired city. “Every C is replaced by G, every G by C, every A by T and every T by

A, for this particular problem only one Hamiltonian path exists, and it passes through Atlanta,

Page | 31
Boston, Chicago and Detroit in that order. In computation the path is represented by

GCAGTCGGACTGGGCTATGTCCGA, a DNA sequence of length 24” [3].

In [3] while solving the problem, the algorithm was applied as follows:

 Interconnect all the variables.

 Eliminate all paths except those which initiate from the beginner.

 Eliminate all paths that do not finish at the final variable.

 Exclude all paths which have length equal to the number of variables or exceed.

 Eliminate all paths that miss any variable.

 Check is whether any path remained left, is so then return “Yes” otherwise return “No”

[3].

Scientists have stored audio and text on fragments of DNA and then retrieved them, a

technique that may provide an approach to handle the vast data of the digital age. They encoded

Martin Luther King speech “I HAVE A DREAM”, a photograph and Shakespeare‟s 154 sonnets.

Later they were able to retrieve them with 99.99 % accuracy [6].

Companies, governments and universities face a challenge of storing the flood of

digitized information, videos, books, movies and songs over the internet, some experts explored

the answer in biology. They have found ways to encode data in cells and bacteria but these

biological elements eventually die. DNA is not alive, it could sit passively in a storage device for

thousands of years, and it is dense, stable and durable. Researchers at Harvard University

reported the encoding of an entire 54,000 word book in strand of DNA [6].

Page | 32
 In [5] an experiment was conducted, the author first downloaded 26 second speech clip of

King Luther in computer as well as and some other information to store. The data was in normal

computer code mean in the form of long string of „0‟s and „1‟s. They use a software program to

convert these „0‟s and „1‟s into letters A, C, G and T, the four bases that make DNA. The single

string was sliced into 150,000 fragments, each of 120 letters long. Each fragment contained 100

letters for encoding the data and the remaining 20 letters work for to index instruction for later

restoring the fragments in right order.

The information was sent to Aqilent Technologies where a laboratory machine used the

data and proper chemicals to manufacture physical string of actual DNA. Then data was

retrieved in Germany, DNA sequencing machine fired up lasers at the fragments and read their

inherited code. Then a computer program collected the fragments in correct order and converted

them back to „1‟s and „0‟s the data was same as it was before [5].

2.8 Encryption and Decryption Data Files Algorithm:

The author in [18] presented two different algorithms, one is for encryption purpose and

the other is for decryption.

2.8.1 Algorithm for Encryption:

Input: Random computer file

Output: Encrypted data in DNA sequences.

1.1 ||S|| = Total number of components in string S.

1.2 Computer files are read byte wise and converted the files respectively into equivalent ASCII

values which are stored in list S0 and S0 ∈ ∑*255 = {0, 1..., 255}.

Page | 33
1.3 Components from S0 are encrypted, converting to base-3 string of length 11, and stored in

string S1 where S1 ∈ Z*3 = {0, 1, 2} and ||S1|| = 0 (mod 11).

1.4 Codes for characters commas and colon are in S2 and S4 respectively. Extensions of input

files are stored in string S3.

1.5 N = ||S1||. String N is stored in S5 after encryption into codes.

1.6 n = ||S1|| + ||S2|| + ||S3|| + ||S4|| + ||S5||.

1.7 S6 S0 ∈ ∑*0 = {0} where (n + ||S6||) = 0 (mod 99).

1.8 S7 = S1.S2.S3.S4.S6.S5.

1.9 S7 is encrypted in string of DNA S8 where S8 ∈ ∑*DNA = {A, G, C, T} without repetition of

nucleotides.

1.10 Segments of 99 lengths are made from S8; segments are so called chunks and are stored in

list L.

1.11 Number of trits requisite to store index number of chunk is called μ, that is μ = ⌈ log3

(||S8||/99) ⌉.

1.12 In base-3 i is chunk index. ID is 2-trit string recognizing the original file. P is parity trit

computed as the addition (mod 3) of the odd located trits in ID and also in i; which is ID (1)

+ i(1)+ i(3) + i(5).

1.13 IX = ID. i. P and IX are encrypted into DNA same as in step 9 taking last character of

chunk from L as previous character for conversion. K (i) = L (i).IX (i), where K is list of

chunks.

1.14 From K the chunks could be synthesized as DNA oligonucleotides [18].

Page | 34
2.8.2 Decryption Algorithm:

Input: Sequences of DNA in which data/ file is encrypted [18].

Output: Original format of computer file.

1.1 List of DNA sequences = L having length n.

1.2 The string of nucleotides beginning at position 102 and finish at n - 1 in ith sequence from L,

is stored in S, where S= L (i) [102: n-1]. The string S of DNA is decrypted into ternary string

and stored in I. k = Decimal (I). Store DNA sequences L (i) [0:99] in database D with k as

key.

1.3 Sequences of DNA are read in increasing order from D of their corresponding keys and are

concatenated to form DNA string S1.

1.4 Decrypt S1 using nucleotide conversion ternary string S2.

1.5 11 trits block is read from S2 and converted into ASCII numbers. The corresponding

characters to these ASCII numbers are being stored in string S3. If trits has not been found

sequenced, with the help of error correction technique they are decrypted.

1.6 S3 has been break down based on occurring of last comma into string S4 and S5. File

extension has been extracted from S5 and stored in string S6.

1.7 Decrypted file has been generated using S4 and S6 [18].

Single strand of DNA performs all the required operations for computational purpose [19].

Now a day the focus is getting benefit of parallelism in it and as well as the work is going on to

overcome the limitation of solving small problems. DNA has flexible structure it helps in

representation of problem elements same to binary code structure. Trillion of unique strands of

DNA are able to represent all of the possible solutions to the problem.

In [19] the key properties for DNA computing are discussed as follows:

Page | 35
1 Synthesizing: In all presented models, a required polynomial length strand was used.

2 Mixing: He took the substances from two test tubes and put into third for combining and to

get union.

3 Annealing: By cooling the solution, a bond of two single stranded complementary DNA

sequences is taken.

4 Melting: While by heating the solution of double stranded DNA, single stranded

complementary components is taken.

5 Amplifying mean copying: The DNA polymerase enzymes one function is replication of

DNA. This reaction of replication takes place under the guidance of single strand DNA that

is template and oligonucleotide primer. This way makes copy of DNA strand.

6 Separating: It is taken by using a technique gel electrophoresis; it helps to separate different

strands of DNA by length.

7 Extracting: Affinity purification is used to extract strands which have given pattern as a

substring.

8 Cutting: A specific short sequence of DNA is distinguished by restriction endonucleases,

which are restriction site. The enzyme at that location cut the double stranded DNA which

contains the restriction site.

9 Ligating: Using DNA ligases paste DNA strands with compatible sticky ends. One enzyme

named DNA ligase bond together, in such a way that ending point strand is bond to a new

strand.

10 Substituting: Polymerase chain reaction site specific oligonucleotide mutagenesis Delete,

Insert and substitute DNA sequences.

Page | 36
11 Marking single strands by hybridization: To make them double stranded complementary

sequences are attached to the strands. By denaturing mean by separating complementary

strands the reverse action unmarking is obtained [19].

DNA guides the cell in making new proteins that govern all the biological behaviours and

gets copied from one generation to another [20]. It carries information between generations.

These molecules contain code for components that living organism needs for growth, production

and daily living. The genes play role in regulating other genes.

Genes are in the nucleus of cell, genes controls the protein production and ribosome are

structure of protein and are present outside of nucleus. Genes control it by sending message. The

information in genes is in the form of nitrogen bases and they are translated into 20 amino acids,

because protein based on amino acid [20].

The human body contain tens of thousands kind of proteins, which perform whole body

work [4]. There are two complementary chains in the structure of DNA, the bond is shown in

figure 9, each nucleotide in DNA has a sugar component joined to a phosphate group at one

point on the sugar, and at other point it attach to nitrogen which contains the bases.

Fig. 9. Bond [4].

Page | 37
 The chain in DNA has phosphate of one nucleotide connected to the sugar of the next

nucleotide to form strand of alternating sugar and phosphates with hanging nitrogenous base, the

structure of DNA base is shown in Figure 10.

Fig. 10. Structure of DNA Bases [4].

On one chain forms bond with T on the other strand but it is weak bond and every C on

strand bond to G on opposite chain, thus they are complementary, and the sequence in one can be

deduced from the other‟s sequence. Design information is transmitted as new DNA to new cells

during development and growth. The complementary of the two DNA strands allows their

information to be copied. These complimentary strands are coded as fundamental building

blocks.

AGG-CTC-AAG-TCC-TAG

TCC-GAG-TTC-AGG-ATC

DNA nucleotide bases exist in the form of codon and complimentary to each other that is

A-T-G-C are complimentary to each other, so these codons can be used for encoding and

decoding of data [4].

Page | 38
 „0‟s and „1‟s are used in silicon based computing for computational purpose because the

hardware only reacts on electrons action and for that the „0‟s and „1‟s was the possible way for

computation [6]. Electrons have only two states “on” or “off”. Every kind of Process, Storage

and Retrieving in traditional computing is operated on by using „0‟s and „1‟s form, different

types of conversions are done for it. The hardware is controlled by these „0‟s and „1‟s they are

ordered by it.

Now the trend is going to change, no need of electrons, no need of such expensive

hardware, no need of CD‟s, hard disks, processors which are in traditional computing, DNA

brought new thoughts, new kind of processing, storage, input output. The future of the computer

is DNA based; the „0‟s and „1‟s are now working on A C G T. The DNA processing is parallel

and the enzymes are controlled instead of hardware parts of traditional computer by the DNA,

they act like hardware and the software is DNA. Now there is no need of usage of potential

energy, the crises of energy do not effects the computing because it does not need of too much

energy [6].

2.9 Assumption of Study:

Everyone wants to have such a computer system which is cost efficient, reliable, and

integrated, perform parallel processing and has the ability to store huge amount of data.

Everything in daily life has interaction with computer which may generate huge amount of data

like Weather broadcasting system, camera snapshots, news, movies, videos, teleconferencing,

social media, hospitals, universities. All this increases the volume of data and need storage and

parallel processing. Limitations of conventional computations, drawbacks, technological usage in

daily life make sense to have solutions and to overcome those problems.

Page | 39
2.10 Problem Statement:

The traditional computation brought an enormous change in life style of humanity. It

impacts every aspect of life, and traditional computation rules the minds of humanity. The

thinking to change traditional computation to DNA based computing is not acceptable to users,

because the traditional computing does not allow to thinking otherwise. The impact of traditional

computing during last decades hides its limitations. People are not aware of DNA based

computing and also DNA based computing is not as richer yet.

The storage limitation of traditional computing, results in loss of huge amount of data, in

different big organizations, companies, industries and universities such as, like NADRA,

Ministry of Environment, Preston University, Shoukat Khanum Hospital and so on.

Limitation of parallel processing slows down the production, problem solving, operations

and services in organisations such as Atomic Energy Commission of Pakistan, Federal

Investigation Authority (FIA) and Pakistan Telecommunication Authority (PTA) face much

problem.

Cost is a major issue, replacement of devices for storage costs a lot to user, after months

and years old devices are replaced with new one such as floppy disc, magnetic tape, CDs and

low capacity Hard Drive, due to less capacity and reliability problems. Processors change with

name of bringing proficiency and fastness but actually it costs users because they have hope that

it is new and will solve their problems, processors like Intel 4004, 4040, 8080, 8008, Intel 8051,

80188, 80486DX, Pentium Pro, Pentium II/III/Xeon and so on.

Page | 40
 The limitation of unawareness about DNA based computing is focussed in this research.

Increasing number of processers in traditional computation is a limitation and is focussed in this

study. Another limitation which is focussed is cost of traditional computation which is compare

to DNA based computing.

Page | 41
3.1 Research Methodology:

To ensure that DNA based computing is suitable and efficient than the conventional

computing according to cost, Speed, parallelism and density of storage. Hybrid research

methodology means Quantitative and Qualitative approach will be used.

 Compare cost of DNA based computing with silicon based computing.

 Analyse parallelism in processing of both computational techniques.

 Mathematically prove density of storage capacity in both type of computation.

 Review the approach of both type of computation to hard problems.

 Educate students about the DNA based computing.

 Present the limitations of silicon based computing and their solutions given by DNA

base computing.

 Review on the state of art in DNA based computing.

The limitation of unawareness about DNA based computing will be overcome by

properly presenting the newly come-way of computation to the users of related field.

Mathematically comparison of cost, storage capacity, parallelism in processing and survey of

achievements in DNA based computing will all help to properly analyse both computation.

3.2 Design and Analysis:

The comparison diagram is designed and is divided in four portions, i) Combinatorial

Problem and Two techniques of computation, ii) the second portion contains the modules which

are focused in this research, iii) in third portion there is module of comparison iv) the last portion

Page | 42
has module for taking out put as a result see figure 11. In diagram also an analyzer (Researcher)

is placed, the researcher actually analyzes both computing techniques.

i. The first portion contains the module of picking combinatorial problem and to

analyze it by both techniques.

ii. The focused four pairs of modules are Process, Storage, Cost and Power

Consumption of both techniques.

iii. Here in this module the comparison of each module will be performed.

iv. This module shows the results.

Page | 43
 Fig. 11. Comparison Diagram.

Page | 44
3.3 Hard Combinatorial Problem:

Bio molecular computing (BMC) methods have been proposed to solve difficult

combinatorial search problems, such as the Hamiltonian path problem [3].

Hamiltonian path problem, the author assumed a map of cities connected to each other

with nonstop flights, for example “it is possible to travel directly from Boston to Detroit but not

vice versa. The goal is to determine whether a path exists that will initiate at the starting city

(Atlanta), finish at ending city (Detroit) and pass through each of the remaining cities exactly one

time”. He assigned sequence of DNA‟s to each city and it was declared that the first name will

follow the last name of the city which is destination. “Every C is replaced by G, every G by C,

every A by T and every T by A, for this particular problem only one Hamiltonian path exists,

and it passes through Atlanta, Boston, Chicago and Detroit in that order. In computation the path

is represented by GCAGTCGGACTGGGCTATGTCCGA, a DNA sequence of length 24” [3].

One other version was prepared late in 2004, during experiment they detect cancer out of

test tube and inject killing molecule which killed the effected molecules. [7].

The architecture shows that a hard combinatorial problem will pass through two different

techniques of computation. One called Silicon Based Computing and the other is DNA Based

Computing. Different attributes of them will compare and figure out the resultant data. Attributes

like Process, Storage, Cost and Power Consumption are focused in this research see figure 12.

Page | 45
 Fig. 12. Architecture of Analyzing Hard Problem.

The potential of DNA computer is that because of powerful computing, national

governments will use it for cracking secret codes. Airline system will map more efficient routes.

Such computing components that composed of living things or itself living thing are

shares 2 features, these features are self improving and injuries healing capability.

Page | 46
3.4 Process of Both Computing Techniques:

The model is for comparison of process, the process data is taken from the previous work and

compare for both computational techniques. Their performances were analyzed carefully and results

were made see figure 13.

Fig. 13. Comparison of Process

3.4.1 PROCESS COMPARISON

DNA Silicon
In a second 1014 DNA flights number The author in [1]calculated that if F be

were simultaneously concatenated in only 1/50 the fraction of sequential calculations and (1-F)

Page | 47
of teaspoon of solution.The ability of trillions is possibly parallelized fraction, then the

of reactions to occur simultaneously provides maximum achievable speed by using P number

the equivalent of massively parallel processing of processors is 1/F+ (1-F)/P).

in silicon-based computers, in which a huge The author in [1] calculated that if 90%

number of possible problem solutions or processing or calculation is parallelized and

searches through pools of information for the 10% remains sequential while 5 processor are

answer are performed at the same time [3]. used, it increase the speed 3.6 times and the

The new computational technique delay of 1.4 mean 28% occurs. If processors

tolerates exceptionally parallel processing. The increased to 10 then its speed up the processing

philosophers made a conclusion that this newly up to 5.3 and the delay of 47% notified, and if

technique has potential to perform at a level of the numbers of processors are increased to 20

ten power seventeen (1017) operations per then it will speed up calculations or processing

second (sec) or more, in other words a level to 6.9 which proves that doubling the hardware

that silicon-based computers will never ever be increase the speed only 30%. If 1000

able to match [22]. processors are used for calculations it will

Vast parallelism is available in DNA to speed up 9.9 and the delay will also increase to

do the combinatorial search among large 990.1 that is 0.99% delay. Conclusion is that a

number of possible solutions. Example is BMC limit will come that the speed will start

method for breaking the Data Encryption decreasing.

Standard (DES). While these methods for Traditional computers perform and

solving hard combinatorial search problems handle operations sequentially. CPU repeats

may succeed for fixed sized problems, they are the same operations “fetch and decode cycle”

ultimately limited by their volume again and again. The process performed very

Page | 48
requirements, which may grow exponentially fast and repeatedly again and again in a row.

with input size [17]. Increasing process speed of silicon based

The author in [19] discusses that 330 computing means escalation clock cycles 24.

trillion operations per second could be

performed by DNA.

3.5 Storing Data in Both Techniques:

The designed diagram is for comparing the storage capabilities of both techniques, first

see the storage of DNA based computing and note down the values and efficiency then storage of

Silicon based computing. After that the comparison and analysis of the calculated data will made

and results will be produced see figure 14.

Page | 49
 Fig. 14. Storage Comparison.

3.5.1 STORAGE COMPARISON:

DNA Silicon
The Storage medium is Nucleic Acid The Storage medium is semiconductor

and the potential of storage capacity in DNA is and required ten power twelve (1012) cubic

an attractive feature. Experiments have been nanometers to store 1 bit [2].The current areal

made and successfully retrieved fifty seven to density notified for magnetic based storage

ninety nine (57-99) base pairs of data that are stands at ten (10) GB per square inch (sq.

not property of bacterium. It is consider that a inch). If it does consider that the thickness of

milliliter (ml) of liquid can have 109 bacteria. hard disk platter is one (1) centimeter (cm) that

Page | 50
If assumption is made that an average of eighty gives one point five (1.5) GB/ml.

(80) base pairs of data for each bacterium,

using four (4) base pairs for each byte that can

approximately give nineteen (19) Gigabytes

(GB) of data for size of one (1)ml [11].

Above ten (10) trillion particles of

DNA can adjust in space of less than one (1)

cubic centimeter zero point zero six (0.06)

cubic inches having density of one (1) bit per

cubic nanometer. DNA computer can hold ten

(10) terabytes data with the mentioned amount

of DNA [7].

Data storing capability of DNA is its

impressive potential; it has been noticed

appealing media due to immense capability of

storage. Ten power twenty one (1021) DNA

bases are found in a gram of DNA, or about ten

power eight (108) Tera-bytes (TB). Hence, a

few grams of DNA may have the potential of

storing all the data stored in the world [17].

Page | 51
3.5.2 Density of DNA:

Amount of data that can be stored in unit gram of DNA is density of DNA, one gram

(1gm) of single stranded genetic code can encode 455 EB (Exabyte‟s) of data. In [18]

information density of 2.2 PB (Peta bytes) per gram of DNA has been achieved. Experimentally

succeeded data storing density for DNA based medium is 1.15 * 1020 = 115 EB (115000 PB).

Derivation of DNA information density:

Suppose in 1gm of DNA x bytes data can be encoded.

Total nucleotides that are required to store file is denoted by I

{I = 11 * x + 11 * 2 + 11 *log10(x)}

Here(x*11) are number of nucleotides for x bytes. (2 * 11) are nucleotides used for 2 extractors

declared in algorithm and (log1011 * (x)) are nucleotides used to store file size on DNA.

(C = ⌈I/99⌉) are number of chunks, every chunk will have ninety nine (99) nucleotide of file data.

Chunk length (l=⌈(log3(C))⌉+ 102) and(⌈log3(C)⌉)is index information of chunk.

Total number of nucleotides for storing file on DNA as

{[102 + log3 (I/99)] * (x *11 + 2 *11 + log10(x) *11)/99}

Hence storage capacity of DNA is maximum 443 (Exabyte‟s) considering (log10(x) = 20).

Number of nucleotide sin 1gm of DNA =(182 * 1019), so numbers of bytes that can be stored on

1gm of DNA are calculates as:

{182 * 1019 = (102 + log3(I/99)) * (x * 11 + 2 * 11 + log10(x) * 11)/99}

Page | 52
{x = 1.15 *1020 bytes (115 Exabyte‟s)}[18].

In [21] there is different results of density, file size, cost and features of DNA are discussed. The

author also showed results of his proposal see table 1.

Table 1. Views of Different Authors [21].

Church et.al [26] Goldman et.al [27] Proposed

Density 0.7 * 1015 B/g 2.2 * 1015 B/g 4.9 * 1020 B/g

File size 5.27MB 739KB File size: 17KB

Cost Not available $12, 600 $4, 023

Features Archival, no random- Archival, no random- Rewritable, random-

access access access

The author in [21] made comparison of different author‟s views and also given his own view.

The data storage density discussed by church et.al looks very low as compared to the others. I

have seen that the attention to the technique increases day by day, thus betterment and efficiency

occurs.

3.6 Cost Comparison of Both Techniques:

This diagram is used for comparing the cost, cost required for both techniques were taken

and analyzed. The production cost and availing the facility, material availability is also focused;

results were made and presented see figure 15.

Page | 53
 Fig. 15. Cost Comparison.

3.6.1 COST COMPARISON:

DNA Silicon
It was declared that DNA particles are Non- Recurring and High Cost

required for newly born computational Investment over every new IC chip design will

technique, its materials are available in every increase. As every new chip is more complex

living thing, so until and unless there is cellular from the preceding one, therefore reducing size

living thing it is available. This material is will not only add complexity to the design but

available everywhere and in enormous amount, it will also increase the overall cost per design

it can obtain easily. from one chip to another [22].

Page | 54
Cost is significant attribute for synthesis and Cost of silicon computing increases due

sequencing of DNA. Synthesis cost is 0.05$ to adaptation of newer versions of devices and

per base [18]. With the increase in the amount also software. The upcoming device or

of data to be stored in DNA there is negligible software is presented in such a way that it is

change in cost used to store unit MB associated going to solve each problem, but in real it

with DNA based storage. Cost of DNA won‟t. Need of market for better, efficient

computing grows with the increasing of file devices increases with usage and interference

size it using. The observation results that for of computing in daily life. The potential of

long term archival using DNA as storage silicon based computing for solving hard

medium becomes moderate and economical. problems and storing large amount of data is

not getting enhanced due to its limitations.

3.7 Energy Consumption of Both Techniques:

Figure 16 is designed to calculate power consumption of both techniques, attributes of

power are gathered and compared in comparison module, and then results were produced. The

power need for processing and storing data is analyzed carefully see figure 16.

Page | 55
 Fig. 16. Energy Comparison.

3.7.1 ENERGY CONSUMPTION COMPARISON:

DNA Silicon
A process with DNA computing in Supercomputers of current era are not

contrast of traditional computation is found very energy efficient, they executes power of

billion times efficient. Now there is no need 109 operations per joule [23].

usage of potential energy, the crises of energy An exclusive hurdle for advancement in

does not affect the computing because it does silicon-based chips includes consumption

not need too much energy [18]. power and heat dissipation. Power density has

In DNA the computational procedures been grown with the rate of rate S0.7 since

and operations takes place chemically, the couple of years for all generations. Size of

Page | 56
bonds made up naturally. So it does not transistors are becoming pretty small and that

consumes power energy, outside energy isn‟t small transistors consume small amount of

required [2]. power, but Integrated Circuit (IC) chip

It was observed during the research that DNA becoming denser and denser due to the number

computing can operates adjacent to thermal of transistors are being placed on it, hence it

equilibrium, really reducing energy dissipation. consumes enormous voltage to run all

It can perform about 2 x 1019 (irreversible) transistors and generates massive heat.

operations per joule [22]. In year 1971, the world's first single chip

microprocessor was introduced by Intel, that

first Intel microprocessor contained twenty

three hundreds (2,300) Transistors at ten (10)

µm, which was using 1/10th of watt.

Transistors were increased for speed up and

one of modern processor, a 3.2 GHz Pentium

IV extra edition consumes one hundred and

thirty five (135) watts.

In dual core 2.8GHz Pentium D the numbers of

transistors were increased to 167 Million and

notified two hundred and forty four (244) watts

power which it consumes [22].

Page | 57
4.1 Data Collection and Analysis:

Previous knowledge has been used for data collection. Teachers helped by discussion and

presentations to collect data, Gathered material from various published research papers,

conferences and books. Collect data from internet browsing, video clips and animations. Focused

area of study was about computation and biological structure, functions and properties of DNA.

Study has been made about different aspects of DNA in human life and Usage of DNA for

storage by nature. Analyses of Adelman experiment, comparing parallelism properties of DNA

with traditional silicon base computing also helped study about the topic. During the study social

media was found an attractive means for knowledge and information gathering daily news DON,

CNN, Discovery all is helpful.

4.2 Results:

The results are made on the bases of designed architecture in previous chapter. Comparison

diagrams for Process, Storage, Cost and Power Consumption are designed and results are made

after comparison of both computational techniques.

4.2.1 Process:

After comparison of different modules of both traditional and DNA based computing the

analysis are made. Results are remarkably attractive. Today‟s super computer performs

operations maximum up to thirty five point eight (35.8) teraflops, which really high speed of

operations occurrence, but the comparative operations of DNA is found hundred (100) teraflops

[22]. This is huge number, hundred (100) teraflops means hundred (100) trillions floating point

operations are operated in a sec see figure 17.

Page | 58
 Operations
120
T
e 100
r
a 80

60
f
l 40
o
p 20
s
0
Silicon DNA

Fig. 17 Operation Comparison Result.

DNA based computing performs 0.1 PetaFlops operations per sec and in comparison the

Silicon based computing only performs 0.0358 PetaFlops operations per sec see figure 18. This

is a huge difference in the performance of both techniques for computation.

Operations/sec
0.12
P
0.1
e
t 0.08
a
F 0.06
PetaFlops
l
o 0.04
p
0.02
s
0
Silicon DNA

Fig 18. Performance Comparison.

Page | 59
 Performing combinatorial problem with traditional computer setups sequential search

tree, generates each branch sequentially and save the shortest one. Performing for 20 cities by

hundred (100) million instructions/sec and also requires forty five (45) million GB memory,

mathematically it takes 2 years to generate the paths. While with DNA it looks possible due to

1015 is Nano mole of material. All operations are going to be performed in parallel.

Parallelism of DNA is really massive, one thousand and eighteen (1018) strands of DNA

gives ten thousands time greater operations then today‟s super computer.

The parallelism of silicon based computing were taken 90% and 10% process as

sequential; the calculated speed is shown in figure 19.

Fig 19. Speed Up

The delay noticed in 90% parallel and 10% sequential processing is shown in figure 20. It

proves that increasing number of processor on chip is not satisfactory any more.

Page | 60
 Percentage of Delay by Increasing Number of
Processors in Traditional Computing
120.0%

100.0% 99.0%
90.8%
D 80.0% 79.6% 83.1%
e 74.4%
65.5%
l 60.0%
a 47.0% Series1
40.0%
y
28.0%
20.0%

0.0%
5 10 20 30 40 50 100 1000
Number of Processors

Fig 20. Delay.

Suppose if parallel processing increase to 95% and only 5% processing remains

sequential then possible speed up with 5 to 2000 processors is shown in figure 21.

95% Parallelism Speed Up

25
S
p 20 19.26 19.62 19.81
e 16.8
15 14.49
e
d 10
Series1
6.89
U 5 4.16
p
0
5 10 50 100 500 1000 2000
Number of Processors

Fig 21. Speed Up with 95% Parallelism.

Page | 61
Only 5% processing remains sequential and 95% is parallel then the delay was noticed that a

limit will reach where it does not increase the speed instead it will start decreasing the speed see

figure 22.

Delay with 5% Sequential

120.00%

100.00% 96.15%98.04%99.01%
D 83.20%
80.00%
e 71.02%
l 60.00%
a Series1
40.00%
y 31.10%
20.00% 16.60%
0.00%
5 10 50 100 500 1000 2000
Number of Processors

Fig 22. Delay with 5% Sequential Processing.

The delay is seen increasing enormously, due to this increase in delay; the number of

processors are not going to increase any more. In comparison the parallelism in DNA computing

increases exponentially.

4.2.2 Storage:

In DNA of bacteria the replications occurs very fast that is equal to thousand (1000)

bits/sec data storage rate, it looks handsome number but as compared to traditional computing it

is very slow. This rate is significantly higher of silicon based computing, but after the 1 st

replication it increases exponentially in DNA that is 2n. Number of strands increases with the

Page | 62
replication and the bit/ sec data rate also increases, operations executes in parallel. After thirty

(30) iterations it reaches one thousands (1000 Gbits/sec) see figure 23. In this situation the

today‟s fast hard drive is beyond to DNA storage rate [11].

Number of Replications
R 1200000000000
e 1000000000000
p
800000000000
l
i 600000000000
c 400000000000
a
200000000000
t
i 0
bits/sec bits/sec
o
Series1 1000 1000000000000
n
After 30 Iterations

Fig 23. Exponential Increase in Replications.

Today‟s video tapes acquire 1012 cubic nanometers space to store one bit data, but DNA

requires only one cubic Nano meter for one bit data, that mean DNA stores data 1012 times more

data in equal space. Data of trillion CD‟s adjust only in one cubic centimeter of DNA [18].

In [21] the author made comparison of different researcher‟s views about storing data.

The density of data storage in DNA is given in figure 24.

Page | 63
 DNA Storage Density
6E+20

5E+20
Density Capibility
4E+20

3E+20
DNA Storage Density
2E+20 DNA Storage Density

1E+20

0
Church et.al Goldman Z. Huimin
et.al
Different Views

Fig 24. DNA Density

DNA has density of eighteen (18 Mbits/ inch), and traditional computing storage stores

1/100,000 times less information in same space.

4.2.3 Cost:

The cost of DNA and traditional computing is analyzed and the result is that increasing

number of transistors on chip attracts users for hoping betterment but actually it costs them,

nothing else.

The production cost of traditional computing goes significantly above the DNA cost

because as far as there is availability of cellular organism there is DNA. No need to plant an

industry and produce it, enormous availability makes it an attractive source. To achieve DNA

Page | 64
level storage and parallel processes, exponential increase in cost will be required for silicon

computing industry.

4.2.4 Power Consumption:

DNA computing is one billion times energy efficient as compared to silicon computation

[24]. The building block of DNA is chemical bonding and the bonding happens in absence of

power source. In DNA 2*1019 operations performed per joule, while in silicon based computing

109 operations occur per joule shown in figure 25.

DNA Performs 99.99% more Operations/joule

As Compare to Silcon Based Computing
120.00%

100.00%

80.00%

60.00% Comparison of Power

Consumption
40.00%

20.00%

0.00%
1

Fig 25 Number of Operations/Joule

Heat and power consumption is seen a hurdle for the future of silicon based computing.

S.7 is the grown rate of power consumption in it, 2300 transistors upgrades to 167 million

transistors and power consumption increased from 1/10th of watt to 244 watt in dual core 2.8

GHz processor [11].

Page | 65
5.1 Conclusion:

It is concluded from the study that DNA is authentic medium for computation; it’s getting

famous round the world for the features it holds. The researcher works hard on it, the number of

limitations in traditional computing is hopefully going to solve in DNA computing. Especially the storage

and processing are going to be enhanced to the level of today’s global world need.

DNA based computing is suitable to use in all the cases compared in the study that is processing,

storage, cost and energy consumption all are supporting this technique for future computation. There is

still a lot to do for its maturity and research is newer stop process. There are certain limitations but will

soon overcome and these limitations will be its advantages. This kind of computation is really beneficial

for humanity in view of every aspect, either health care or social life. It helps human in each and every

situation.

DNA based computing is the part of biological, chemical and computer study. It combines

number of fields and researchers have attraction in it. It involves researchers of most dangerous

diseases and they have solutions in it. Disease like cancer will hopefully diagnose and treated via this

technique of computing.

The study is introducing the newly technique of computing in Pakistan to open doors of research

for the student of Computer science, Information technology, Nano science and Biochemical. The study

helps the organizations in selecting and acquiring of new technology according to their needs.

5.2 Future Work:

Working with silicon based computing we use different algorithms to hide data in

pictures, in diagrams, that we called cryptography. Using same technique of data encryption, in

Page | 66
near future an algorithm could be written that will read the combination of DNA genes and will

also predict the sequence and data would be stored in encrypted form.

As we know that DNA has encrypted code for production of proteins according to the

genes requirement for growth of leaving thing, the change in the inheritance genes occurs due to

the environmental effects like the temperature, the standard of food, the hobbies, the mind sitting

etc., the codons in DNA directly take input from these variables and produced cells and genes

accordingly with an amount of changes from the parental genes.

Page | 67
References:

1. G. M. Amdahl. “Validity of the Single-Processor Approach to Achieving Large Scale

Computing Capabilities”. AFIPS Conference Proceedings, P: 483–485, 1967.

2. D. Nixon “GSEC Assignment Version 1.3 DNA and DNA Computing in Security

Practices” Is the Future in Our Genes? GSEC Assignment Version 1.3 P: 1-14, 2002.

3. L. M. Adleman, “Molecular computation of solutions to combinatorial problems”.

Published by American Association for the advancement of Science, vol 266(11) P:

1021-1024, 1994.

4. P. Rakheja (“Integrating DNA Computing in International Data Encryption Algorithm”)

International Journal of Computer Applications (0975 – 8887) Volume 26, No.3, P: 1-6,

July 2011.

5. G. T. Naik (“Storing Digital Data in DNA”) Scientists Store King Speech, Shakespeare

Sonnets in DNA WSJ 11/30/2014.

6. S. N Kumar, “A Proper Approach on DNA Based Computer”. American Journal of

Nanomaterials, 3(1), P: 1-14, 2015.

7. P. S. Sugathan, “DNA COMPUTING”, DEPARTMENT OF COMPUTER SCIENCE

COCHIN UNIVERSITY OF SCIENCE AND TECHNOLOGY COCHIN – 682022,

2010.

8. J. D. Watson and F. H. C. Crick, “THE STRUCTURE OF DNA”, Cold Spring Harbor

Symposia Quantitative Biology, Cavendish laboratory, Cambridge, England, P: 123-131,

1953.

Page | 68
 9. F. H. C. Crick, “THE STRUCTURE OF DNA”, The Medical Research Council Unit for

the Study of the molecular Structure of Biological Systems, The Cavendish Laboratory,

Cambridge, England, P: 532-538, 1953.

10. R. Deaton, M. Garzon, J. Rose, D. R. Franceschetti, S. E. Stevens, “DNA Computing: A

Review”, IOS Press, Fundamenta Informaticae 30, P: 23-41, 1997.

11. A. Maurya, A. Nair and S. Sanyal “An Overview of the Evolutionary Trends in

Molecular Computing using DNA”, Veermata Jijabai Technological Institute India, 2011.

12. M. N. Stojanovic and D. Stefanovic, “Deoxyribozyme-Based Half-Adder”, Contribution

from the DiVision of Clinical Pharmacology and Experimental Therapeutics, Department

of Medicine, Columbia University, Box 84, 630 W 168th Street, New York, New York

10032, and Department of Computer Science, University of New Mexico, Albuquerque,

New Mexico 87131, PP: 6673-6676, 2003.

13. S. Kumar, G. Pant, V. Sharma and P. Bisht, “Nanotechnology in Computers”,

International Journal of Information & Computation Technology, ISSN 0974-2239

Volume 4, Number 15, P: 1597-1603, 2014.

14. J. Silc, T. Ungerer and B. Robic, “A survey of new research directions in

microprocessors”, a Computer Systems Department, Jozef Stefan Institute, Jamova 39,

1001 Ljubljana, Slovenia, Department of Computer Design and Fault Tolerance,

University of Karlsruhe, Germany, Faculty of Computer and Information Science,

University of Ljubljana, Slovenia, P: 175–190, 2000.

15. M. Amos, G. Paun, G. Rozenberg and A. Salomaa, “Topics in the theory of DNA

computing”, Theoretical Computer Science 287, P: 3 – 38, 2002.

Page | 69
 16. J. Watada, “DNA Computing and its Application”, Computational Intelligence: A

compendium, Volume 115 of the series study in computational intelligence. P: 1065-

1089, 2008.

17. A. Gehani, T. LaBean, and J. Reif, “DNA-Based Cryptography”, Department of

Computer Science, Duke University, Box 90129, Durham, NC 27708-0129, 2000.

18. V. Dhameliya, D. Limbachiya, M. Khakhar and M. K. Gupta, “On Optimal Family of

Codes for DNA Storage”, arxiv: 1501.07133v1 [cs.IT], 28 Jan, 2015.

19. S. Sharma, CSE-DNA-Computing-report (“seminar report on working of web search

engine”) published in www.Studymafia.org, P: 1-25, Feb 2009 to May 2009.

20. L. C. Micheal, H. Damman, R. A. Lue, C. K. Yoon and R. Morel “Discover Biology” 3rd

Edition, w.w. Norton, USA, 2004.

21. S. M. H. T. Yazdi, Y. Yuan, M. Jian, Z. Huimin and O. Milenkovic, “A Rewritable,

Random-Access DNA-Based Storage System”, arxiv: 1505.02199v1 [cs.IT], 8 May,

2015.

22. S. Hassan, Himaira and M. Asghar, “LIMITATION OF SILICON BASED

COMPUTATION AND FUTURE PROSPECTS”, Second International Conference on

Communication Software and Networks, 978-0-7695-3961-4/10, IEEE DOI

10.1109/ICCSN, 2010.

23. K. Hammed, “DNA Computation Based Approach for Enhanced Computing Power”,

International Journal of Emerging Sciences ISSN: 2222-4254 1(1), April 2011.

24. A. Srivastava and J. Ramadas, “Analogy and Gesture for Mental Visualization of DNA

Structure”, Citation: Srivastava, Analogy and Gesture for Mental Visualization of DNA

Page | 70
 Structure. In Treagust, D.F. & Tsui, C.-Y. (Eds.), Multiple Representations in Biological

Education, Dordrecht, The Netherlands: Springer. P: 311-329, 2013.

25. J. C. Adams, “On the Application of DNA Based Computation”, Candidate for HBA

2000, BESc (CE), Richard Ivey School of Business / Faculty of Engineering

Science University of Western Ontario London, Ontario Canada, 2000.

26. G. M. Church, Y. Gao, and S. Kosuri, “Next-generation digital information storage in

DNA,” Science, vol. 337, no. 6102, pp. 1628–1628, 2012.

27. N. Goldman, P. Bertone, S. Chen, C. Dessimoz, E. M. LeProust, B. Sipos, and E. Birney,

“Towards practical, high-capacity, low-maintenance information storage in synthesized

DNA,” Nature, 2013.

Page | 71