MAPGAPS

MAPGAPS performs global alignment of protein sequences using multiply-aligned profiles to detect, classify, and produce multiple sequence alignments of distantly related and functionally divergent proteins for evolutionary and functional analyses.

Key Features:

• Statistical analysis of sequence patterns: Infers biochemical similarities and differences from patterns of sequence similarity and divergence among evolutionarily related proteins.
• Use of multiply-aligned profiles: Employs multiply-aligned profiles as queries for detecting and classifying related sequences and as templates for multiple sequence alignment.
• Karlin-Altschul statistics and PSI-BLAST heuristics: Applies Karlin-Altschul statistics to enhance sensitivity and incorporates PSI-BLAST heuristics to improve speed.
• Handling of distantly related sequences: Detects and aligns weakly conserved, distantly related proteins, demonstrated by alignment of 33 P-loop GTPases with known structures.
• Comparison with other methods: Outperformed sequence- and structure-based alignment methods such as hmmalign and PROMALS3D in comparative studies by avoiding misalignment of significant regions.
• Scalability: Applied to a dataset of 65 million protein sequences to identify, classify, and align nearly half a million putative P-loop GTPase sequences.

Scientific Applications:

• Phylogenetic analysis: Produces multiple sequence alignments suitable for reconstructing evolutionary relationships among protein families.
• Functional annotation: Detects conserved and weakly conserved motifs to support inference of biochemical function and divergence.
• Structural biology investigations: Aligns sequences with known structures, such as P-loop GTPases, to inform structural interpretation and comparative modeling.

Methodology:

Uses a curated multiple-profile alignment as input, employs multiply-aligned profiles as queries and templates, applies Karlin-Altschul statistics for scoring, and uses PSI-BLAST heuristics to accelerate searches while detecting weakly conserved sequence motifs across distantly related proteins.