Solving multiple RNA sequences alignment by multi-objective genetic algorithm method based on Pareto optimality

The multiple RNA sequences alignment (MSA) is one of the most interesting tasks in bioinformatics. Based on different aspects many algorithms have been developed to obtain the best alignment. However, finding the accurate MSA remains a challenge to researchers. In this work, the MSA is cast as a multi-objective optimization problem for which a new algorithm is proposed (RNA-MOO). This method is based on the principle of Pareto front combined with a genetic algorithm. Concerning the objective functions, in addition to the entropy function which gives information about the variability of residues in a specific column of the alignment, we have used two new objective functions: the Weighted Fully Matched Column (WFMC) and the Base Pair Score (BPS). The WFMC function favorises the alignments with conserved regions and consequently penalizes those with less conserved regions. BPS is a function based on the secondary structure of sequences determinate by the RNAafold server. This function gives information about the conservation degree of the secondary structure.