DNA Computing

During the last years, we have studied various DNA based models. These models generally encode information by DNA and manipulate information by enzymes. The emphasis has been on autonomous models (i.e., models that work without hand-crafting). The applications of autonomous DNA models lie in the analysis of  DNA in vitro and eventually in the logical control of cell behaviour in vivo.

Our main inventions have been an autonomous DNA model for finite-state machines, an automaton called computational genes able to logically control cell activities, the first autonomous solution of Adleman's first experiment, a bunch of new sticker-based algorithms, and a method that allows to implement sticker-based algorithms in silico onto dedicated hardware (FPGA).

 Literature:

• I. Martinez-Perez, W. Brandt, M. Wild, K.-H. Zimmermann: Bioinspired parallel algorithms for maximum clique problem on FPGA architectures. J. Sig. Proc. Syst., 1939-8115 (online), 2009.
• I. Martinez-Perez, Z. Ignatova, K.-H. Zimmermann: An autonomous DNA model for finite state automata. Int. J. Bioinform. Res. App., vol. 5, no. 1, 81-96, 2009.
• I. Martinez-Perez, Z. Ignatova, K.-H. Zimmermann: Exploiting the features of finite state automata for biomolecular computing. J Recent Patents DNA & Gene Sequences, vol. 3, no. 2, 130-138, 2009.
• I. Martinez-Perez, Z. Ignatova, K.-H. Zimmermann: DNA Computing Models. Springer, 300 pages, 2008.
• I. Martinez-Perez, Z. Ignatova, K.-H. Zimmermann: Computational genes. Wikipedia, 2008.
• I. Martinez-Perez: Biomolecular Computing Models for Graph Problems and Finite State Automata. PhD thesis, mbv Berlin, 2007.
• I. Martinez-Perez, Z. Ignatova, K.-H. Zimmermann: Computational genes: A tool for molecular diagnosis and therapy of aberrant mutational phenotype..BMC Bioinformatics, vol. 8, 365, 2007.
• I. Martinez-Perez, Z. Ignatova, K.-H. Zimmermann: An autonomous DNA model for finite state automata. Techn. Report, 06.1, TUHH, 2006
• I. Martinez-Perez, Z. Ignatova, K.-H. Zimmermann: Solving the maximum clique problem via DNA haiprin formation. Techn. Report, 06.3, TUHH, 2006.
• I. Martinez-Perez, Z. Ignatova, Z. Gong, K.-H. Zimmermann: Solving the Hamiltonian path problem via DNA hairpin formation. Int. J. Bioinform. Res. App., vol. 1, 389-398, 2006.
• I. Martinez-Perez, Z. Ignatova, K.-H. Zimmermann: An autonomous DNA model for stochastic finite state automata. Techn. Report, 06.2, TUHH, 2006.

A computational gene is an autonomous molecular automaton that is able to assemble an anti-drug or a wild-type gene if the mRNA of an aberrant gene has been detected. Computational genes have been successfully tested in vitro.

Patent:

• I. Martinez-Perez, Z. Ignatova, K.-H. Zimmermann: Rechengen/Computer Gene/Gene de Calcul, DE/EN/US, 2007-09. 

A careful design of DNA strands is crucial for several biological applications such as microarray techniques, Polymerase Chain Reaction (PCR), and DNA computing. For this, the important criterion under laboratory conditions is the hybridisation energy of two DNA strands. During the last decade, a thermodynamic model was developed that allows for the calculation of the DNA/DNA hybridisation energy and recently also the cross-hybridisation energy of structural motifs.

We have developed a new algorithm for the secondary structure prediction of DNA/DNA cross-hybridisation complexes called HYBGRAPH. The method is based on Gibbs free energy minimisation and the paradigm of dynamic programming.

Literature:

• S. Torgasin, K.-H. Zimmermann: Algorithm for thermodynamically based prediction of DNA/DNA crosshybridisation. Int. J. Bioinform. Res. App., vol. 6, no. 1, 82-97, 2010.

 DNA computing is an integral part of our teaching activities (rotative seminar in the summer term).