Future Perspectives on Molecular Computing (detail)

1.Computational Theory and Simulation of Molecular Reactions

In the field of molecular computing, as well as in our project, the computational power of molecular reactions has been studied mainly by formal language theory and complexity theory. However, since molecular reactions follow physico-chemical rules, these discrete methods are not enough to analyze their computational power.

We think that it is important to combine the discrete methods in computer science and the continuous methods in physical chemistry based on differential equations.

In addition to theoretical analysis, computer simulation of molecular reactions is also necessary. Computer simulation has two purposes. One is to improve theoretical analysis, and the other is to verify the design of molecular reactions in molecular computers. The latter corresponds to circuit simulation in the case of designing VLSI.

We also consider that the simulation technology itself is worth studying. Bio-molecular reactions having huge combinatory complexity result in a huge number of differential equations that hold among the molecules in the reactions. Since it is impossible to formulate and solve all the differential equations, one must make approximations by applying some discrete methods (e.g., comparing DNA sequences as strings). Therefore, as in theoretical analysis, it is important to combine discrete methods and continuous methods in computer simulation.

If the simulation technology is established, it will be applied not only to the design of molecular computers, but also to the prediction and interpretation of any kind of molecular reactions involving DNA, RNA, proteins, etc.

2. Self-organization by Bio-molecules

Life is the result of self-organization of molecular reactions in the course of evolution. In order to understand life, it is necessary to analyze the self-organizing power of molecular reactions. One of the approaches to the understanding of self-organization is to construct an artificial system that self-organizes.

Therefore, constructing self-organizing systems out of bio-molecules, i.e., wet artificial life, is expected to lead to the understanding of the origin of life. Investigation on autonomous computation should be included in this direction of research. Compartmentation is one of the key issues for constructing self-organizing systems.

3. Solving Combinatorial Problems by Bio-molecules

Solving combinatorial problems by virtue of the massive parallelism of molecular computing is still an important research direction in this field. There is still much room to be improved in the Adleman-Lipton Paradigm. In particular, scaling up the existing methods and solving large problems should be tried. Applying autonomous computation to enhance the paradigm is also an interesting research issue.

4. Genetic Analysis

Genetic analysis is one of the most promising application areas of molecular computing, in particular, DNA computing. At the beginning of our project, we investigated the implementation of the basic operators of relational databases, because we intended to build databases of genes by DNA molecules themselves. Using the techniques of DNA computing, it might be possible to efficiently retrieve gene databases.

Landweber and Lipton suggest that by transforming a DNA sequence from a chromosome into a pool of DNA molecules in a test tube, it is possible to analyze the sequence using the operators established in DNA computing.

Suyama in our project also suggests the use of DNA computing to analyze genetic information. In particular, he plans to combine the technology of DNA chips with that of DNA computing to analyze gene expressions. He claims that DNA computing makes it possible to perform sophisticated gene expression analyses without sequencing cDNA.

5. Nanotechnology Using Bio-molecules

Nanotechnoloogy is also a promising application area of molecular computing. Constructing complex nanoscale structures requires computation at the molecular level. Since it is difficult to manipulate individual molecules, we should rely on the self-organizing power of molecules. Needless to say, Winfree and Seeman's work is expected to be applied to nanoscale construction.

Recently, DNA-based methods for assembling nanoparticles were reported. In those methods, small single-stranded oligomers are attached to nanoparticles, such as colloidal gold particles, and self-assembly of nanoparticles is guided by small oligomers with sticky ends, or a long single-stranded DNA molecule on which nanoparticles are aligned.

Such research shows the possibility that if one can construct complex structures out of DNA molecules, one can also organize other kinds of molecules guided by the structures of DNA.