Current Research In Dna Computing Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

This paper talks about the division of the current research in DNA computing. There are many advantages of the new methods implemented over the ones which have been used in the past. The paper includes them along with the applications from which these methods are derived from. In addition to this, it explains the characteristics of a Biomolecular computer. But the DNA computer has still got many challenges to face. The new methods of computation have to find its place in the pragmatic world even though it has proved to be much better than the conventional methods. The DNA applications along with their feasibility have been discussed in this paper. This paper has been written keeping in mind that the reader should not be necessarily acquainted with in-depth knowledge of biochemistry. A basic knowledge of DNA computing though might be required.

Adleman in 1994 [1] pioneered the technique of DNA computation by using DNA molecules for computation purpose. Its work suggested that the development of this technology would lead to construction of new architectural paradigms which shall be advantageous over conventional computers. Adleman's experiments suggested that DNA has massive parallelism as a result of which it can solve potentially complex problems in limited time. Adleman used a brute force algorithm to solve the Hamiltonian pathway. His approach of using a redundant algorithm was never meant for practical applications but it did indicate the feasible computational ability of DNA.

Adleman's application of biomolecular engineering to solve the Hamiltonian algorithm urged the others to follow him. The others started to work to generate more complex and effective computational methods by applying similar pathway solutions to difficult computational problems. It has been proved that the working of some of the DNA computer is similar to that of a Turing machines. Many theoretical concepts and mathematical models have been developed to show the same. Apart from this efforts have been taken to improvise its working by mitigating the errors and identifying bio related engineering schemes and applications for solving new challenging problems. They also concentrated on solving the major problem of available storage space. But to find its application in real-life and its pragmatic viability, there is still a long way to go.

While the practical application of the DNA computers is highly questionable, many experiments suggest that this biomolecule can form the basis of emerging computational paradigm. The advantages of developing a DNA computer can fall into three categories.

  1. Massive parallel processing of DNA would allow generation of polynomial time solutions which otherwise is highly impractical using conventional computers.
  2. Fundamental abilities of DNA include high information storage and close interaction with emerging biological techniques.
  3. Integration of physical sciences and computer sciences for exploring the limitations of computability and manipulating biomolecular chemistry.

DNA computation can be viewed as a combinatorial field which encompasses many interrelated disciplines.


The most important benefit presented by the projected schemes of DNA computing is its capability to conduct high degree of parallelism. This might create the possibility to compute a large number of intractable problems. Another advantage could be its ability to solve easy but lengthy problems and the polynomial problems as well. If DNA is used for solving huge problems, then its computation will be called Classic.

Classical models are superior to the old methods because of its following benefits:

  1. It provides high degree of parallelism
  2. It has the ability to perform massive searching.
  3. It has a huge memory.
  4. Solves difficult problems.

Classical DNA models have the following drawbacks

  1. The whole process takes a lot of time( measured in hours or days) with continuous mechanical and human intervention.
  2. This model generates a very high number of solution sets even for simple problems( 1+2=3).
  3. Error rates are high; this model has no proof-reading ability and biological techniques which are included in the process like in vivo, in vitro technique are highly inconvenient and cumbersome.

The above mentioned advantages and disadvantages indicate that a DNA computer would probably be good for solving decomposable problems. Such problems include a sequence of non related tasks; the ability of DNA to store large quantities of data and perform massive parallelism would make it optimum to solve such problems. However, for non-decomposable problems which require large amount of sequential processing DNA computer might be ineffective. [2]

Thus, in DNA classic computing, there are a large number of distinct computational schemes, each having a set of capabilities, benefits and certain feasibility and applications. The surface-based algorithm which is used to compute the minimum set cover is an instance of the difference between the DNA and Adleman's algorithms. [3] This algorithm is basically used to reduce the difference between the actual DNA samples and the ones which are generated. It fixes the DNA generated on silicon or glass. But this algorithm has a limitation. It is believed that if the solution created is not in sync with the size of the DNA strand then its application would be inefficient as it depends on the encoding quantity. Optical scanners were used to examine DNA bit operations to improve the feasibility of the algorithm [4]. This improves the implementation of the algorithm as it allows us to read the output even without the laboratory equipments. This supports the advancements in the field of DNA electronics where electronic equipments can be used to display the outcome. [5] helps in the reduction of errors by using 3D DNA features. This also enables the storage of large quantities of data.

DNA computational model has been used for breaking the DES Data Encryption Standard. Conventional technology took lesser time than DNA computational models. However, the DNA models were cheaper and more potent. In DES a 64-bit message is encrypted using a 56-bit key. Electronic keys are data streams that are used to decode sensitive information. Using a high configuration computer different encrypting components a key to DES can be found in many days. Requirements are that 2^43 data sets of corresponding encrypted and unencrypted messages are needed. [6] Theoretical research suggests that the same message can be broken by using a DNA computer and a redundant search algorithm in a span of 4 months. But the only requirement is a single cipher-text pair. Another related research [7] shows that by using a more sophisticated algorithm the DES can be broken by using only a gram of DNA.


A lot of research has been conducted to solve NP- complete problems by using a DNA based algorithms. In a NP-complete problem as the complication increases the computational problem increases exponentially. NP- complete problem can be tractable or intractable. Adleman's initial work and related research indicated a feasible solution to Hamiltonian pathway problem. More complex NP-complete problems can be solved by DNA computers but at the cost of exponentially increase in volume or material. The property of a DNA to solve tractable NP-complete problems can have applications in real world situations like business management and services. Many rough calculations and more so heuristic algorithms are used to solve the cost problems in NP-complete. The cost problems could be related to scheduling, optimum use of raw materials and routing. These problems might be consistent with the ones which are already computed by using DNA computational schemes.

But still, even today, the technology available is unable to make DNA computing to be used in real time. Thus, it is impossible to solve problems which have to be reported in term of weeks or days. This prohibits the use of DNA computing in specialist systems. In spite of this, DNA can be used to solve problems which have high cost requirements in the beginning. An example of this is chip design. Another useful application of DNA computing is that it helps in planning the travel by optimizing bus and airline routes.

Practical instances suggest that a DNA computer would be implemented in various ways. With respect to hardware, an optimum solution can be obtained by focusing on specific processors rather than trying for a universal DNA computer. Research [8] suggest that a DNA computer can be used to solve problems in which a specific solution is not required and the problem is complex enough to preclude conventional computational techniques. If the use of DNA models proves to be better at solving hard computational problems but biological techniques continue to determine solution times much longer than conventional computer techniques then DNA computers can be used for alias benchmarking of heuristic algorithm. This would increase the use of DNA computers. In such a scheme, the DNA computers could be used to generate the scheduling of a particular product. This step will take a lot of time ( it may take months) before production begins. This combined usage of DNA computers and conventional computers will result in more frequent usage which will cause changes in supply and demand parameters. Although the production will continue at optimum levels.

DNA has a four bit storage component ATGC. It has unlimited capacity for miniaturization of data storage. A conventional CD can store 800MB of data while 1 gram of DNA can store 10^14 MB of data which is equal to 145 trillion CDs.


In this application, conventional computers have no capacity. Unknown stands of DNA are sequenced. This is achieved by amplification and re-encryption of unknown DNA strands into a redundant form. This form can be worked upon by sticker DNA computation models. Various applications are DNA fingerprinting, sequencing, screening of population and mutation detection.


Many papers suggest that the property of DNA to quickly encode information may have applications in nano technology specifically fabrication techniques combining low end computational capabilities and RNA manufacturing abilities. Since both these fields are vaer nascent, experimental and practical applications are highly speculative but promising.


DNA computational model might have high error rates but the nature of the DNA is such that it residentially maintains data integrity. Evolution of genetic algorithm may lead to discovery of principles of error correction. With an advance in DNA computation we will have an understanding of limitations of conventional computation. DNA's property of massive parallelism can lead to development of parallel processing components using more conventional electronic computers. DNA computation shows promise in the fields of molecular modeling and simulations.


Current research has led to emergence of new computation models. DNA computing can be thought of as a library of new computing abstractions. Although a "genuine application" has to be developed, the prospects seem to be bright.


  1. L. Adleman, "Molecular Computation of Solutions to Combinatorial Problems", Science. v.266. November 1994. 1021-1024
  2. L. Adleman, "On Constructing a Molecular Computer", 1st DIMACS workshop on DNA based computers, Princeton. 1995. In DIMACS series. volume 27 (1996). 1-21
  3. T. Eng. B. Serridge, "A Surface-Based DNA Algorithm for Minimal Set Cover", 3rd DIMACS workshop on DNA based computers. June 1997. 74-82
  4. T. Leete. J. Klein. H. Rubin, "Bit Operations Using a DNA Template", 3rd DIMACS workshop on DNA based computers. June 1997. 159-165
  5. N. Jonoska. S. Karl. M. Saito, "Creating 3-Dimensional Graph Structures with DNA", 3rd DIMACS workshop on DNA based computers. June 1997. 192-203
  6. D. Boneh. C. Dunworth. R. Lipton, "Breaking DES Using a Molecular Computer", 1st DIMACS workshop on DNA based computers. Princeton. 1995. In DIMACS series. volume 27 (1996). 37-65
  7. L. Adleman. P. Rothemund. S. Roweis. E. Winfree, "On Applying Molecular Computation To The Data Encryption Standard", 2nd DIMACS workshop on DNA based computers. Princeton. 1996. 28-48
  8. B. Fu. R. Beigel, "On Molecular Approximation Algorithms for NP Optimization Problems", 3rd DIMACS workshop on DNA based computers. June 1997. 93-101