EMMA

EMMA extends multiple sequence alignments by adding unaligned sequences into existing constraint alignments to enable scalable and accurate updates of large MSAs.

Key Features:

• Integration of unaligned sequences: Adds unaligned sequences into a predefined multiple sequence alignment (constraint alignment).
• MAFFT--add foundation: Builds upon MAFFT--add's capability for adding sequences into an existing alignment.
• MAFFT-linsi--add optimization: Optimizes MAFFT-linsi--add, the most precise MAFFT--add mode, to improve alignment accuracy.
• Divide-and-conquer framework: Employs a divide-and-conquer framework to enhance scalability and accuracy.
• Scalability: Handles large-scale constraint alignments, including datasets with hundreds of thousands of sequences.
• Sequence length heterogeneity: Manages alignments that exhibit substantial sequence length heterogeneity.
• Accuracy under realistic conditions: Demonstrates superior alignment accuracy compared to other methods when adding sequences under a variety of realistic conditions.

Scientific Applications:

• Updating large alignments: Updating existing multiple sequence alignments with newly sequenced or additional sequences.
• Incorporating biologically chosen sequences: Incorporating sequences into MSAs based on biological insights or constraints.
• Handling heterogeneous lengths: Managing and refining alignments where sequences vary substantially in length.
• Large-scale MSA refinement: Refining and maintaining alignment fidelity for very large datasets, including hundreds of thousands of sequences.

Methodology:

Employs a divide-and-conquer framework to optimize MAFFT-linsi--add (MAFFT--add) for inserting unaligned sequences into a predefined multiple sequence alignment.