Bd-RPC

Bd-RPC performs rapid phylogenetic placement of novel viral samples to improve accuracy and efficiency in viral phylogenetic inference.

Key Features:

• Frequency Vector Recoding: Approximates phylogenetic distances by recoding nucleotide bases into frequency vectors for efficient comparison of viral sequences.
• Phylogenetic Simulated Annealing Search: Optimizes concordance between the recoded distance matrix and a phylogenetic tree using a simulated annealing search.
• Indel Heuristic for Sequence Recognition: Incorporates an indel (insertion/deletion) heuristic to recognize foreign sequences and improve placement robustness.
• Comparative Performance: Demonstrated superior precision relative to PAGAN2, EPA-ng, and TreeBeST across Alphacoronavirus, Alphaherpesvirinae, and Betacoronavirus as measured by Split and Robinson-Foulds distances.

Scientific Applications:

• Phylogenetic placement and tree construction: Place new viral samples into existing phylogenies to study virus evolution and transmission dynamics.
• Outbreak investigation and surveillance: Trace evolutionary transitions and support analyses of infectious disease outbreaks.

Methodology:

Applies Frequency Vector Recoding to approximate sequence distances, uses a Phylogenetic Simulated Annealing Search to align the recoded distance matrix with a phylogenetic tree, and employs an indel heuristic for foreign-sequence recognition; performance comparisons used Split and Robinson-Foulds distances.