PhylDiag

PhylDiag identifies conserved synteny blocks between two eukaryotic genomes to define ancestral gene arrangements while accounting for gene-level events such as deletions, tandem duplications, and lineage-specific de novo gene births.

Key Features:

• Gene-tree-based homology: Uses gene trees to define gene homologies and to account for gene deletions as events that can disrupt synteny.
• Orientation and duplicate handling: Incorporates gene orientations and handles tandem duplicates and lineage-specific de novo gene births in synteny detection.
• Pairwise genome comparison: Operates by comparing two genomes together with their corresponding gene trees.
• Gap(max) parameter: Allows gaps up to a user-defined gap(max) parameter that is theoretically estimated to improve synteny block precision.
• Statistical validation: Applies rigorous statistical validation to assess the significance of identified synteny blocks and reduce random matches.
• Benchmarking and metrics: Benchmarked against i-ADHoRe 3.0 and Cyntenator and compared using metrics for measuring 2D-distances in homology matrices on real and simulated datasets.
• Robustness to genomic disruptions: Detects small synteny blocks despite insertions, deletions, incorrect annotations, or micro-inversions.
• Post-processing strategies: Implements four post-processing strategies to correct micro-rearrangements, recognize mono-genic conserved segments, generate non-overlapping segments, and repair incorrect synteny ruptures.

Scientific Applications:

• Reconstruction of ancestral arrangements: Infers ancestral gene order and orientation by identifying conserved synteny blocks between extant genomes.
• Quantification of conserved synteny: Measures and quantifies conserved genomic segments, including small conserved blocks resistant to rearrangements.
• Evaluation of genomic events: Analyzes the impact of gene-level events such as deletions, tandem duplications, and de novo gene births on genome structure.
• Method benchmarking: Serves as a reference for benchmarking synteny detection methods using simulated evolution scenarios and real datasets.

Methodology:

Uses gene trees to define homologies, applies a theoretically estimated gap(max) parameter, conducts statistical validation of synteny blocks, benchmarks results using 2D-distance metrics against i-ADHoRe 3.0 and Cyntenator on real and simulated datasets, and applies four post-processing refinement strategies.