A computational approach to building gene silencing modules

The past decades pointed out RNA has alot of functions besides being an information carrier (mRNA). Small RNAs (e.g. antisense RNA) form an essential part of different prokaryotic regulatory mechanisms by, for example, blocking the ribosome binding site (RBS). As such sRNA can be used in synthetically constructed biological devices to silence a gene on demand. Tools and know how for model-based design of RNA molecules that efficiently block the RBS of a specific gene are however still underdeveloped. Here we present a method to design silencing modules that can efficiently block the translational process. This approach uses knowledge on antisense RNA to model the physical nature of this biological interaction. The available literature was used to identify potential characteristics of a good silencing sequence. Based on this information a bioinformatics framework was developed to enable a computational characterization of a potential silencing sequence. Herein, several dynamic programming algorithms are used to accurately predict these RNA-RNA interactions. The influence of the different defined features of the candidate sequences, which were semi-rationally generated and send through a preliminary in silico filter, was investigated. The performance of a group of selected candidate sequences are tested in vivo to determine their silencing capacities. As a test case, mRNA containing a red fluorescent protein was constructed using biofab parts. Based on these results the importance of the features is evaluated. Ultimately, this computational approach can be used for the design of tailor made silencing modules with excellent performance