MAFFT

MAFFT performs multiple sequence alignment (MSA) for DNA and protein sequences, providing scalable and accurate alignment methods for analyses of large sequencing datasets.

Key Features:

• H-INS-i, F-INS-i, G-INS-i: Iterative refinement methods that incorporate pairwise alignment information into their objective functions.
• Iterative refinement and progressive techniques: Multiple algorithmic modes support both progressive alignment and iterative refinement approaches.
• G-large-INS-1: A scalable variant of G-INS-1 that retains high accuracy while being applicable to datasets containing 50,000 or more sequences.
• Large dataset handling: Capable of aligning from hundreds to tens of thousands of sequences on typical computational resources.
• Add unaligned sequences: Ability to add unaligned sequences into an existing alignment.
• Nucleotide direction adjustment: Direction adjustment features specific for DNA alignment.
• Constrained alignments: Support for constrained alignment modes to enforce user-specified relationships.
• Parallel processing: Options to utilize parallel processing to accelerate alignment computations.
• Dot plot: Dot plot feature available for DNA alignment visualization and analysis.
• Benchmark accuracy: Iterative refinement options have demonstrated superior accuracy compared to TCoffee and CLUSTAL W in benchmark tests involving alignments of more than 50 sequences.

Scientific Applications:

• Multiple sequence alignment generation: Production of MSAs for DNA and protein sequences across small to very large datasets.
• Large-scale sequencing analyses: Alignment of extensive sequencing datasets including thousands to 50,000+ sequences for downstream analyses.
• Incremental alignment workflows: Incorporation of new unaligned sequences into existing alignments for iterative dataset expansion.
• DNA-specific analyses: Use of dot plots and nucleotide direction adjustment for DNA alignment inspection and correction.
• Accuracy-sensitive studies: Application of pairwise-informed iterative refinement methods where benchmarked alignment accuracy is critical.

Methodology:

MAFFT implements progressive alignment and iterative refinement techniques (H-INS-i, F-INS-i, G-INS-i) that incorporate pairwise alignment information into objective functions, provides a scalable G-large-INS-1 variant, supports adding unaligned sequences, constrained alignments, nucleotide direction adjustment, dot-plot analysis, and parallel processing.