Basically, during the summer we have successfully developed the following features for the first version of DNAmazing:
Although DNAmazing has performed well some basic functions of CAD programs, due to the a different approach that we have taken (lithography-like inputting, automatic generation of folding path, automatic generation of staples sequences with sticky ends), there certainly have been some issues related to the initial results and the efficiency of the program.
Regarding the processing time of the program,we observed that some large structures ( larger than 20x20 squares) may take long time for the program to handle the Hamiltonian algorithm, yet the results for smaller structures are very nice. In fact, medium smiling faces (12x12 squares) take only 5 to 10 seconds to get the final results; whereas tradition CAD programs may take us at least 10 minutes to get the first step done: drawing of scaffold strands. This can be improved by employing parallel computing technique to accelerate the speed of the DNAmazing.The scaffolding ways produced by DNAmazing tend to be symmetric as a filtering process was developed to optimize and select the best possible results from Hamiltonian circuit. This feature has not been observed in other programs because users design the scaffold folding pathways themselves. However, we observed some strange results for some exceptional structures with very complicated holes. We believe that the only way to "mature" the algorithm is to carry out more tests.