Computing with DNA

Abstract

id e use natural materials to make unnatural objects. WW This is how chemist Nadrian C. Seeman of New York (N.Y) University describes the research that goes on in his laboratory. He and his collaborators have spent the last 15 years working with DNA molecules, assembling short strands into various branched structures, several types of knots, and a number of DNA-edged geometric shapes, including cubes and octahedrons with cutoff corners. Along the way, the researchers have learned how to construct different kinds of branched junctions and how to attach sticky ends to these molecular stalks, enabling them to fashion pieces of DNA into unusual geometries. One of their main goals has been to develop methods of handling molecules to fabricate molecular-scale machinery and electronics (SN: 12/10/94, p. 396). Now, Seeman has eased into a new role as adviser to computer scientists venturing into the biology lab to try out their ideas about computing with DNA molecules. Our experience with these systems has uncovered a large number of experimental pitfalls that may confront individuals working with DNA computing, Seeman says. He provides the kind of handy, practical advice-the tricks of the DNA trade-rarely mentioned in scientific papers, textbooks, or lab manuals. We're trying to take advantage of the rapidly evolving technology for manipulating DNA in the laboratory, says computer scientist Richard J. Lipton of Princeton University. Lipton and others envision computation taking place in test tubes rather than on silicon chips; they see information storage occurring in DNA-laced drops of water instead of on magnetic disks. DNA has a number of qualities that computer scientists believe could make it an effective vehicle for delivering high-performance computing. DNA-based computers, Lipton maintains, would offer advantages in speed, memory capacity, and energy efficiency over conventional electronics for solving certain types of problems. The hope in this new field is that the pattern-matching and polymerization processes of DNA chemistry, multiplied by the enormous number of molecules that fit into a small volume, can handle computations too difficult for conventional silicon-chip-based computers.