QOMA

QOMA maximizes the Sum-of-Pairs (SP) score for multiple alignment of protein sequences using a graph-based local optimization strategy.

Key Features:

• Objective: Maximizes the Sum-of-Pairs (SP) score to assess and improve multiple protein sequence alignments.
• Input: Operates on protein sequences starting from a preliminary multiple alignment.
• Graph Representation: Represents the alignment as a K-partite graph to enable structured optimization.
• Local Optimization: Applies a sliding window approach for targeted local optimizations that iteratively adjust the alignment.
• Adjustable Parameters: Exposes Window Size and Graph Sparsity parameters to balance computational time and optimization scope.
• Order Independence: Does not depend on the order in which sequences are aligned, unlike progressive alignment methods.
• Benchmark Performance: Reported to achieve higher SP scores on BAliBASE compared with ClustalW, Probcons, Muscle, T-Coffee, and DCA, with larger improvements for distant proteins.

Scientific Applications:

• Phylogenetic analysis: Produces high-quality alignments suitable for evolutionary tree inference.
• Conserved motif identification: Improves detection of conserved motifs across multiple protein sequences.
• Evolutionary studies: Supports analyses of sequence divergence and conservation among distant proteins.
• Functional annotation: Aids transfer of functional information via improved alignments.
• Structural biology investigations: Provides alignments that can inform comparative modeling and structure-based analyses.

Methodology:

Generate a preliminary alignment, represent the alignment with a K-partite graph, perform sliding-window local optimizations on the graph with iterative adjustments to enhance the SP score, and control behavior via Window Size and Graph Sparsity parameters.