LaRA 2: parallel and vectorized program for sequence-structure alignment of RNA sequences.

Area of Science:

• Bioinformatics
• Computational Biology
• Molecular Biology

Background:

• Non-coding RNA function is dictated by spatial conformation (secondary structure) via Watson-Crick interactions.
• Accurate RNA structural alignment is crucial for discovering novel RNA families and their functions.
• Existing algorithms are computationally intensive, especially for long sequences, and often ignore pseudoknots.

Purpose of the Study:

• To develop a significantly faster and more accurate software for pairwise and multiple structural alignments of RNA sequences.
• To incorporate the handling of arbitrary pseudoknots, which are often omitted in current software.
• To leverage modern hardware through parallel execution techniques.

Main Methods:

• Development of LaRA 2, a SeqAn-based software, as an improved re-implementation of the LaRA tool.
• Implementation of multi-threading and vectorization for parallel processing.
• Introduction of a novel heuristic for calculating a lower bound of the solution.

Main Results:

• LaRA 2 demonstrates significantly improved speed compared to comparable software for RNA structural alignment.
• The software achieves up to a 130-fold increase in speed over its predecessor.
• LaRA 2 accurately aligns structured RNA sequences and handles arbitrary pseudoknots.

Conclusions:

• LaRA 2 provides an efficient tool for analyzing large sets of RNA secondary structures rapidly.
• The structural alignments generated by LaRA 2 can be used to derive structural motifs for genomic database searches.
• This advancement facilitates quicker identification and functional annotation of RNA molecules.