Gene Selection and Evolutionary Modeling Affect Phylogenomic Inference of Neuropterida Based on Transcriptome Data.

Area of Science:

• * Insect evolutionary biology
• * Phylogenomics and bioinformatics
• * Systematics of Holometabola

Background:

• * Neuropterida, comprising Megaloptera, Neuroptera, and Raphidioptera, has debated higher-level evolutionary relationships.
• * Previous studies have not definitively resolved the phylogenetic positions of families within Neuropterida.
• * Understanding these relationships is crucial for reconstructing the evolutionary history of this insect superorder.

Purpose of the Study:

• * To investigate the evolutionary history and higher-level relationships within Neuropterida using phylogenomic data.
• * To assess the impact of different analytical models on phylogenetic reconstruction.
• * To clarify the monophyly of Neuroptera and its placement within Holometabola.

Main Methods:

• * Sequencing and analysis of a new transcriptome for *Ctenochauliodes similis* (Megaloptera).
• * Phylogenomic data matrix construction with 1392 ortholog genes from 22 species.
• * Phylogenetic reconstruction using concatenation and coalescent methods with site-homogeneous and site-heterogeneous models.

Main Results:

• * Site-homogeneous models suggested paraphyletic Neuroptera, with Coniopterygidae as sister to other Neuropterida.
• * Site-heterogeneous models and analyses of genes with high phylogenetic signal recovered monophyletic Neuroptera.
• * Data and model selection were shown to significantly influence phylogenomic outcomes.

Conclusions:

• * The choice of phylogenetic model is critical for accurate evolutionary inference in large datasets.
• * Neuroptera is likely monophyletic, contrary to findings from some site-homogeneous models.
• * This study highlights the importance of robust model selection in phylogenomics.