Distance-based phylogenetic inference from typing data: a unifying view.

Area of Science:

• Microbiology
• Bioinformatics
• Computational Biology

Background:

• Typing methods are crucial for infectious disease surveillance, outbreak investigation, and understanding infection natural history.
• The adoption of high-throughput sequencing generates massive datasets, necessitating advanced phylogenetic analysis algorithms.
• Existing algorithms face challenges in correctness and scalability when handling large volumes of typing data.

Purpose of the Study:

• To provide a unified review and perspective on phylogenetic inference algorithms for typing data.
• To identify potential computational bottlenecks and areas for improvement in current algorithms.
• To facilitate a better understanding of algorithms used in analyzing large-scale biological datasets.

Main Methods:

• Review of distance-based algorithms for phylogenetic tree inference.
• Analysis of closest pair joining schemes and hierarchical clustering approaches.
• Examination of evolutionary models influencing cluster proximity and optimization criteria.

Main Results:

• Phylogenetic inference algorithms differ primarily in their definitions of cluster proximity and optimization criteria.
• These criteria often depend on underlying models of evolution.
• A unified view helps in understanding algorithmic differences and similarities.

Conclusions:

• A comprehensive understanding of phylogenetic algorithms is essential for effective analysis of large-scale typing data.
• Identifying computational bottlenecks can lead to improved algorithms for infectious disease surveillance.
• This work aids researchers in selecting and refining methods for complex biological data analysis.