Computer-assisted planning of the completely synthetic gene experimental design

The construction of a completely synthetic gene offers the most flexible way to synthesize proteins (smaller than ~50,000 molecular weight) with any desired features. The design process of synthetic gene experiments involves a large amount of information, such as the deoxyribonucleotide sequences (primary structure of genes) and amino-acid sequences (the primary structure of proteins), the codon table (the relation between the DNA triplets and amino acids), restriction endonuclease map (mapping of restriction endonuclease sites in a gene sequence), and so on. The ambiguity derived from the reverse translation (conversion of amino acids to proper DNA triplets) further increases the complexity of the design process. A prototype system which is able to manage the information and guide biologists toward a workable experiment design has been developed. The domain knowledge is incorporated into the system by using a knowledge-based expert-system shell, LISP, and C procedural routines on a SUN workstation. An intelligent, graphical user interface has been developed to ease biologists' interaction with the workstation. The system can reduce the experiment design process to hours.