A simple protocol for the inference of RNA global pairwise alignments

Area of Science:

• Bioinformatics
• Computational Biology
• Molecular Biology

Background:

• Accurate RNA alignment is essential for RNA annotation and characterization.
• Existing RNA alignment methods have varying performance based on RNA features, making method selection challenging.
• No single method consistently outperforms others for all RNA alignment tasks.

Purpose of the Study:

• To develop a workflow to aid in selecting the most suitable method for RNA pairwise alignment.
• To evaluate the performance of different RNA alignment algorithms under various conditions.

Main Methods:

• Tested six RNA alignment algorithms employing diverse approaches.
• Utilized curated RNA secondary structure annotations and RNA alignment datasets.
• Simulated scenarios with unknown secondary structures by predicting them.

Main Results:

• Algorithm performance is influenced by RNA primary sequence identity.
• The availability and accuracy of secondary structure information significantly impact alignment outcomes.
• The proposed workflow effectively guides method selection based on input data characteristics.

Conclusions:

• The optimal choice of RNA pairwise alignment method is contingent upon sequence identity and secondary structure data.
• This workflow provides a practical approach for researchers to improve RNA annotation accuracy.
• Further refinement of alignment strategies considering sequence and structure is warranted.