No abstract

The emerging field of DNA computing draws from biochemistry, math and computer science to create new ways to solve problems. DNA‐based computers use combinatorial libraries of DNA strands to store information while manipulating these strands to solve problems. In order for DNA computing to be successful, these libraries must be able to be constructed in an easy and reliable manner. We have constructed libraries of 28 combinations and have developed a method for building libraries of almost unlimited size. We first assembled of an initial library (a “block”) that represents every possible combination of two sequences in five positions (five variables). Oligonucleotides were primer‐extended to create 25 possible sequences of 88‐bp. Another 3‐variable block of 62‐bp was similarly created. The blocks contained restriction sites. They were digested and ligated to create 28 combinations of 150‐bp library strands. Currently, we are verifying that each of the 28 possible strand combinations is present in the library. This library should be useful in solving mathematical satisfiability (SAT) problems. This research is supported by an NSF‐UBM Award.

DNA is being investigated as a tool for biomolecular computing. Some of these biomolecular computing methods require the creation of combinatorial libraries. We have developed a method for creating combinatorial libraries of nearly any size. We demonstrate creation of library of 28 or 256 distinct DNA strands of 128bp, made by linking eight 16‐mer oligomers together. Each 16‐base oligomer has one of two possible sequences. One sequence is designated as “true” and the other as “false”. In computing terms, each library strand has eight variables with values of true or false (corresponding to 1 or 0 in a binary system). Every possible combination of the eight linked true or false strands (variables) yields 28 or 256 distinct library strands. The library was constructed by a series of primer extension and ligation steps. We used a variety of PCR‐based methods and gel electrophoresis to demonstrate that the sequences in the library have the expected length and composition. This work was funded by NSF, the Camille and Henry Dreyfus Foundation, the Geneseo Student Association and the Geneseo Association.

DNA‐based computers use combinatorial libraries of DNA strands to store information while manipulating these strands to solve computationally hard problems in parallel. In order for DNA computing to be successful, these libraries must be constructed in an easy and reliable manner.We show the construction of two libraries of 2exp3 and 2exp5 strand combinations and then we link them to produce a library of 2exp8 or 256 combinations of 150‐bp DNA strands. We then verify that the expected strands are present. This library should be useful in solving mathematical satisfiability (SAT) problems.