Improved accuracy of multiple ncRNA alignment by incorporating structural information into a MAFFT-based framework

Area of Science:

• Bioinformatics
• Computational Biology
• Molecular Biology

Background:

• The discovery of functional non-coding RNAs (ncRNAs) has increased the importance of RNA structural alignment.
• While pairwise RNA structural alignment has improved, multiple structural alignment methods lag behind sequence-based approaches.
• Accurate multiple RNA structural alignment is crucial for understanding ncRNA function but remains a challenge.

Purpose of the Study:

• To develop an accurate and efficient multiple RNA alignment framework.
• To address the limitations of extending existing algorithms like Sankoff for multiple sequence alignment.
• To provide a novel approach for incorporating structural information into multiple RNA alignments.

Main Methods:

• Developed X-INS-i, an iterative multiple RNA alignment framework integrated into the MAFFT program.
• Incorporated structural information via external pairwise structural alignment tools (e.g., SCARNA, LaRA).
• Introduced a new objective function, Four-way Consistency, based on base-pairing probabilities at each alignment stage.

Main Results:

• X-INS-i demonstrated superior performance in the BRAliBASE benchmark, outperforming existing methods using the sum-of-pairs score (SPS).
• The method's accuracy in predicting common secondary structures is comparable to or exceeds leading tools like RNA Sampler.
• Achieved improved accuracy and time complexity compared to straightforward extensions of the Sankoff algorithm.

Conclusions:

• X-INS-i offers a significant advancement in multiple RNA structural alignment.
• The framework's flexibility allows integration with various pairwise alignment algorithms and base-pairing probability predictors.
• The developed method provides a robust basis for predicting common secondary structures and is publicly available.