noisy

noisy identifies and removes phylogenetically uninformative and homoplastic sites from nucleic acid multiple sequence alignments to mitigate alignment errors and improve phylogenetic reconstruction.

Key Features:

• Identification of uninformative sites: Assesses the distribution of character states along a cyclic ordering of taxa within a multiple sequence alignment to detect homoplastic and randomized sites.
• Site filtering to improve trees: Removes identified uninformative sites to increase stability and indices of tree quality in downstream phylogenetic reconstruction.
• Applicability conditions: Particularly effective for datasets with large evolutionary distances, variable substitution rates, low average bootstrap support, and when at least 12 to 15 taxa are present.

Scientific Applications:

• Phylogenetic reconstruction refinement: Reduces misleading signal from homoplasies and alignment errors in nucleic acid-based phylogenies.
• Analyses of deep divergences and heterogeneous datasets: Enhances accuracy and stability of trees for datasets with large evolutionary distances and heterogeneous substitution rates.

Methodology:

Assesses character state distributions within a cyclic ordering of taxa in a multiple sequence alignment and removes sites identified as homoplastic or strongly randomized.