GUIDANCE2

GUIDANCE2 assesses and quantifies unreliable regions in multiple sequence alignments (MSAs) to improve the reliability of phylogenetic and comparative genomics analyses.

Key Features:

• Unreliable-region detection: Detects and scores unreliable alignment regions within MSAs.
• Indel uncertainty modeling: Accounts for uncertainty in insertion and deletion (indel) formation during sequence evolution.
• Guide tree uncertainty: Evaluates alignment reliability by accounting for inaccuracies in phylogenetic guide trees used in progressive alignment algorithms.
• Co‑optimal pairwise alignments: Acknowledges multiple co‑optimal pairwise alignments that can underlie MSA construction.
• Alternative MSAs: Produces a set of alternative MSAs to enable exploration of different alignment scenarios.
• Benchmark validation: Validated using extensive simulations and empirical benchmarks, demonstrating improved detection of unreliable regions relative to seven other methods.
• Impact on downstream analyses: Improves reliability of downstream MSA-based analyses, including detection of positive selection.

Scientific Applications:

• Phylogenetic inference: Improves accuracy of phylogenetic analyses by identifying and scoring unreliable MSA regions.
• Comparative genomics: Supports comparative genomics by assessing alignment reliability across homologous sequences.
• Positive selection detection: Enhances detection of positive selection by removing or downweighting unreliable alignment regions.

Methodology:

Accounts for uncertainty in indel formation, evaluates the effects of phylogenetic guide tree assumptions used in progressive alignment algorithms, acknowledges co‑optimal pairwise alignments, generates alternative MSAs, and is validated via extensive simulations and empirical benchmarks.