RNA-RNA Interaction Prediction Using Genetic Algorithm


The GRNAs predicts RNA-RNA interaction structures including kissing hairpins. GRNAs computes the minimum free energy (MFE) structures by using a novel genetic algorithm. Here, you can see our program, GRNAs.exe, to predict minimum free energy interaction structure. Also we use program RNAeval.exe from ViennaRNA Package to calculate minimum free energy.

You can enter two desired RNA sequences by running GRNAs.exe. For each input, there are some interaction structures with their minimum free energies, so that each prediction represents an individual in population. The results are shown by parantheses, dots and brackets. In the other words, the base pairs in each RNA secondary structure are represented by '(' and ')', and each unpaired base is declared by '.'. The base pairs between two RNAs are represented by '[' and ']'.

In RNA-RNA Examples, there are some RNA-RNA pairs on datasets [1, 2]. GRNAs is evaluated in sensitivity, positive predictive value, F-measure [1] and Matthews Correlation Coefficient [3] on these datasets.

[1] Kato Y, Akutsu T, Seki H: A grammatical approach to RNA-RNA interaction prediction. Pattern Recognition 2009, 42: 531-538.
[2] Busch A, Richter AS, Backofen R: IntaRNA: efficient prediction of bacterial sRNA targets incorporating target site accessibility and seed regions. Bioinformatics 2008, 24: 2849-2856.
[3] Seemann SE, Richter AS, Gesell T, Backofen R, Gorodkin J: PETcofold: predicting conserved interactions and structures of two multiple alignments of RNA sequences. Bioinformatics 2011, 2, 211-219.