Automating the DNA computer: solving n-variable 3-SAT problems
DNA'06 Proceedings of the 12th international conference on DNA Computing
| Hi-index | 0.00 |
In attempting to automate the computation of n-variable 3-CNF SAT problems using DNA, two physical architectures were scrutinized, the "in-line" architecture and the "waste-well" architecture. Computer modeling of the effects of non-specific binding predicted that the in-line version would not work for problems of more than 7 variables. According to the model, the "wrong answer" DNA strands would swamp out the "correct answer" DNA strands in the final computation module. And in fact, the in-line architecture never performed a computation higher than 6 variables. To perform a 20 variable instance of the 3-CNF SAT problem a manual version of the waste-well architecture was employed. Surprisingly though, after analysis of the modeling results, it appears that through a simple protocol change, the in-line architecture may have been able to perform higher order computations.