MASTR

MASTR performs simultaneous multiple sequence alignment and secondary structure prediction of non-coding RNAs (ncRNAs) to identify evolutionarily conserved structural features and inform structure–function relationships.

Key Features:

• Simultaneous alignment and structure prediction: Combines multiple sequence alignment and RNA secondary structure prediction, explicitly considering both sequence conservation and structural features.
• Markov Chain Monte Carlo (MCMC) with simulated annealing: Employs MCMC within a simulated annealing framework to iteratively improve alignments and predicted structures.
• Combined cost function optimization: Minimizes a cost function that integrates sequence conservation, covariation, and basepairing probabilities.
• Competitive accuracy and efficiency: Reports performance competitive with existing programs in terms of prediction accuracy and computational efficiency.

Scientific Applications:

• Evolutionary studies: Identifies conserved RNA structural features across species to inform evolutionary analyses.
• Functional annotation of ncRNAs: Supports annotation of non-coding RNAs by revealing conserved structural motifs linked to function.
• Comparative genomics: Facilitates comparative genomics by aligning sequences with consideration of secondary structure.
• Motif discovery: Detects conserved structural motifs within sets of related RNA molecules or across taxa.

Methodology:

Uses Markov Chain Monte Carlo (MCMC) within a simulated annealing framework to explore and iteratively refine the space of alignments and secondary structures guided by a combined cost function integrating sequence conservation, covariation, and basepairing probabilities.