PrIME

PrIME performs probabilistic inference of orthology and gene-species-tree reconciliations using a Bayesian Markov chain Monte Carlo framework to improve phylogenetic analysis.

Key Features:

• Orthology Analysis: Determines whether homologous genes are orthologs or paralogs by modeling speciation and gene duplication events.
• Probabilistic Modeling: Integrates gene sequences and species-tree constraints within a probabilistic framework instead of relying on a single reconstructed gene tree.
• Bayesian MCMC Framework: Employs a Bayesian Markov chain Monte Carlo approach to sum over possible gene trees and reconciliations with the species tree.
• Comparison and Performance: Has been directly compared to PrIME-GEM (a probabilistic duplication-loss model) and MrBayesMPR (Bayesian inference combined with most parsimonious reconciliation) and shown to outperform them on synthetic and biological datasets, including robustness to incomplete taxon sampling artifacts.

Scientific Applications:

• Genome Annotation: Provides higher-confidence orthology assignments to support inference of gene function during genome annotation.
• Phylogenetic Inference: Improves phylogenetic studies by reconciling gene trees with species trees to yield more reliable evolutionary relationships.
• Functional Genomics: Facilitates prediction of gene function across species by resolving orthologous relationships.

Methodology:

Uses a Bayesian Markov chain Monte Carlo method to sum over possible gene trees and gene-species reconciliations within probabilistic models that integrate gene-sequence evidence and species-tree constraints.