Reciprocal Smallest Distance (RSD)

Reciprocal Smallest Distance (RSD) identifies orthologous genes between pairs of genomes by computing global sequence alignments, estimating evolutionary distances with maximum likelihood, and selecting the smallest reciprocal distances.

Key Features:

• Reciprocal Smallest Distance criterion: Selects orthologs based on the smallest reciprocal evolutionary distance between sequence pairs rather than reciprocal best hits.
• Extension of reciprocal best BLAST hits (RBH): Builds on the RBH concept by incorporating evolutionary distance estimation to reduce RBH confounding.
• Global sequence alignment: Compares full-length sequences using global alignment to ensure comprehensive sequence comparison.
• Maximum likelihood distance estimation: Computes evolutionary distances using maximum likelihood methods for improved accuracy.
• Paralog resolution: Reduces misidentification due to close paralogs by using distance-based criteria.
• Sensitivity to divergent sequences: Detects orthologs that may be missed by conventional best-hit approaches across significant evolutionary divergence.
• Parameter adjustability: Allows tuning of algorithmic parameters to identify increasingly distant orthologs.
• Pairwise genome comparison: Operates on comparisons between pairs of genomes.

Scientific Applications:

• Phylogenetic reconstruction: Supports reconstruction of phylogenetic histories by providing distance-based ortholog assignments.
• Gene diversification and conservation analysis: Enables analysis of gene diversification and conservation patterns across species.
• Comparative genomics of divergent species: Identifies orthologs in highly divergent species where best-hit methods fail.
• Molecular basis of disease studies: Informs studies of the molecular basis of disease through accurate ortholog identification.
• Developmental complexity and evolutionary biology: Facilitates investigations into developmental complexity and broader evolutionary questions by clarifying gene relationships.

Methodology:

Performs global sequence alignment, applies maximum-likelihood estimation to compute evolutionary distances, and selects the smallest reciprocal distances between sequence pairs.