RNA-RNA interaction prediction using genetic algorithm.

Area of Science:

• Computational Biology
• Molecular Biology
• Bioinformatics

Background:

• RNA-RNA interactions regulate gene expression and cell development.
• Predicting these interactions involves finding optimal secondary structures for two RNAs.
• Existing algorithms often have high computational demands.

Purpose of the Study:

• To introduce a novel genetic algorithm, GRNAs, for predicting RNA-RNA interaction structures.
• To improve accuracy and reduce computational time in RNA-RNA interaction prediction.

Main Methods:

• GRNAs utilizes a genetic algorithm approach where individuals represent secondary structures of interacting RNAs.
• The minimum free energy serves as the fitness function.
• Crossover and mutation operations are employed to find optimal structures.

Main Results:

• GRNAs demonstrates appropriate accuracy on standard datasets.
• The algorithm achieves lower time complexity compared to state-of-the-art methods.
• Effectiveness and validity are confirmed through comparisons with other algorithms.

Conclusions:

• GRNAs is effective for joint secondary structure prediction of interacting RNAs.
• The algorithm offers comparable time efficiency per iteration to existing methods.
• The proposed method is a valid and efficient tool for RNA-RNA interaction analysis.