Optimization of GenoCAD Design Based on AMMAS

Abstract

By successively click on prescribed grammar captured in successful genetic designs to assemble a range of genetic parts for example BioBricks, large and complicated genetic systems composed of substantial functional blocks can be constructed. As the number of genetic parts increases, each category of genetic parts includes so many parts that the process of assembling a great deal of genetic parts is costly, time-consuming and error-prone. GenoCAD is a web-based application for synthetic biology to guide users through the design of artificial gene networks, protein expression vectors and other complex genetic constructs by continuously click on predefined grammars according to the notion of genetic parts. However, at the last step of a design in GenoCAD, it’s difficult for users to determine which basic part will be taken in every category. On the basis of statistical language model, a probability distribution over a string S reflecting how frequently a string S will occur and a mathematical model to select basic genetic parts to form a genetic construct can be determined. After converting the parts assembly process into a mathematical model, adaptive maximum-minimum ant system (AMMAS) proposed in this paper can be applied to the mathematical model to figure out an optimal combination of parts of a design with maximum probability automatically within seconds. The adaptive maximum-minimum ant system (AMMAS) can not only optimize the parts selection process of a design but also can devise particular projects performing specific functions based on former successful parts assembly experience. Consequently, redundant operations can be reduced and cost as well as time spent in biological experiments can be minimized drastically.