Inverse folding of RNA II

In the Oxford Summer School in Computational Biology 2011, one of the projects was aimed at creating an improved method for inverse RNA folding based on a genetic algorithm (GA) approach. This turned out to be sufficiently successful [16] and inspiring, that we will extend on it in this year’s summer school. The inverse RNA folding problem consists of finding an RNA sequence that, as a molecule, folds into a target structure, or structures. We will focus exclusively on what is known as the secondary structure of the molecule, which is the set of base pair interactions formed in the structure. The secondary structure can be inferred from the sequence with reasonable precision, with several paradigms available. This allows the inverse folding problem to be addressed completely in silico (i.e. computationally), using predicted structures as a proxy for the true structure of a designed sequence. The aim of this follow-up project is to address at least some of the following points: 1. enhance the existing GA with functionality for multi-objective fitness and objective functions 2. extend the method to utilise more folding paradigms to create more robust designs 3. extend the software with a graphical user interface and parallelism 4. improve the sequence initialisation and search methods to be able to design perfect sequences for more targets 5. extend the target specification to include structures with pseudoknots, nucleotide constraints for specific positions as well as overall distributions, less specific targets with regions where any structure is allowed or even as more abstract shapes, and possibly even two or more molecules forming complex structures 6. improve complexity results for the inverse RNA folding problem It is not realistic to assume that all these points can be addressed in full, or possibly even at all. The list should work as a source of inspiration, from which you can choose the most relevant and interesting elements to work on. However, the ordering does reflect our a priori prioritisation. This project description first describes the basic background for the inverse folding problem, including a brief summary of our method. It then outlines the points for extensions in more details, also discussing the importance of each proposed extension.