MultiPhyl

MultiPhyl performs high-throughput maximum likelihood phylogenetic analyses by distributing computations across heterogeneous desktop machines to construct and evaluate phylogenetic trees from large numbers of amino acid or nucleotide alignments.

Key Features:

• Distributed computing framework: Utilizes idle processing power of multiple heterogeneous desktop computers to run large-scale phylogenetic calculations.
• High-throughput analysis: Processes hundreds to thousands of amino acid or nucleotide alignments concurrently for large phylogenomic datasets.
• Phylogenetic inference capabilities: Supports model selection, tree searching, and bootstrapping across many gene families.
• Extensive model support: Implements 88 amino acid maximum likelihood models and 56 nucleotide maximum likelihood models.
• Near-linear speedup: Achieves near-linear performance scaling by efficiently utilizing idle clock cycles of participating machines.
• Cross-platform compatibility: Runs across different operating systems to accommodate heterogeneous computing environments.

Scientific Applications:

• Phylogenetic tree construction: Generates maximum likelihood phylogenetic trees from extensive amino acid or nucleotide alignments.
• Model selection and testing: Performs comprehensive selection and comparison among 88 amino acid and 56 nucleotide ML models across many gene families.
• Bootstrapping analysis: Conducts bootstrapping to assess support and reliability of inferred phylogenetic trees on large datasets.

Methodology:

Distributed computation across heterogeneous desktop machines executes high-throughput maximum likelihood analyses including model selection, tree searching, and bootstrapping, with support for 88 amino acid and 56 nucleotide ML models and near-linear speedup from leveraging idle clock cycles.