UnderlyingApproach

UnderlyingApproach computes alignment-free distance measures from irredundant common subwords to enable phylogenetic analysis of whole genomes that may lack shared genes, exhibit low sequence similarity due to rearrangements and inversions, or differ in length.

Key Features:

• Alignment-Free Methodology: Employs a distance function based on subword compositions using "independent" irredundant common subwords to circumvent alignment-based limitations.
• Subword Composition Analysis: Derives matching statistics from irredundant common subwords to capture the statistical occurrence of shared sequences while preventing redundant counting of shared regions.
• Efficient Construction and Linear Complexity: Constructs a set of patterns called "underlying" that is linear in size relative to the input and devoid of overlaps, enabling scalable computation.
• Scoring Function Based on Underlying Patterns: Builds a scoring function from underlying subwords to prioritize significant subword regions and improve accuracy and computational efficiency.

Scientific Applications:

• Phylogenetic Tree Reconstruction: Applies the alignment-free distance measure to reconstruct phylogenetic trees and has demonstrated superior performance compared to existing state-of-the-art methods.
• Biological Insights from Subword Analysis: Uses selected subwords that often carry significant biological information to provide meaningful biological insights from genome comparisons.

Methodology:

Leverages a distance-like measure derived from irredundant common subwords; derives matching statistics from these subwords; constructs non-overlapping "underlying" patterns that are linear in size; filters redundant subwords to represent each genomic region uniquely; and builds a scoring function from the underlying subwords.