Ultraconserved Elements and Machine Learning Classifiers Enable Robust Phylogenetics and Taxonomy in Model and...

Area of Science:

• Zoology
• Genomics
• Evolutionary Biology

Background:

• Nematodes exhibit vast diversity, but only a fraction of species are described due to size and cryptic diversity.
• Traditional phylogenetic methods struggle with nematode evolutionary history resolution.

Purpose of the Study:

• To develop ultraconserved elements (UCEs) probe sets for two nematode families (Panagrolaimidae and Rhabditidae).
• To establish a scalable and cost-effective phylogenomic framework for nematodes.

Main Methods:

• Designed UCE probe sets targeting thousands of loci for Panagrolaimidae and Rhabditidae.
• Conducted in vitro testing to assess locus recovery and phylogenetic reconstruction accuracy.
• Utilized machine learning (XGBoost) to determine the minimum loci for genus-level classification.

Main Results:

• Successfully recovered numerous loci, enabling robust phylogenetic reconstruction for Panagrolaimidae.
• Confirmed previous analyses, with one strain reclassified as Neocephalobus halophilus BSS8.
• Identified minimal loci (46 for Rhabditidae, 39 for Panagrolaimidae) for accurate genus classification via machine learning.

Conclusions:

• UCEs provide a scalable and cost-effective phylogenomic framework for nematodes.
• This approach enhances taxonomic resolution and evolutionary inference across the phylum.
• The method is suitable for biodiversity assessments and field-based sequencing.