Enhancing Phylogenetic Analysis: Integrating Nonparametric Methods for Addressing Abrupt Evolutionary Shifts

A bstract Phylogenetic comparative methods are indispensable for analyzing cross-species data while accounting for evolutionary relationships. Traditional methods like phylogenetically independent contrasts (PIC) and phylogenetic generalized least squares (PGLS) are well-suited for gradual evolution under Brownian motion (BM) or similar assumptions. However, their effectiveness diminishes under abrupt evolutionary shifts or heterogeneous evolutionary processes. We introduce a hybrid framework, outlier-guided correlation (OGC) for large datasets and outlier- and distribution-guided correlation (ODGC) for small datasets, collectively termed O(D)GC, which integrates Pearson and Spearman correlation analyses to handle outliers and non-normal data dynamically. Using simulations across diverse evolutionary scenarios, we compared PIC-O(D)GC with a comprehensive range of methods, including eight robust regression approaches (PIC-MM, PIC-L1, PIC-S, PIC-M, and their PGLS counterparts); PGLS optimized using five evolutionary models: BM, lambda, Ornstein–Uhlenbeck random (OU-random), OU-fixed, and Early-burst; Corphylo (an OU-based method); PIC-Pearson; and two advanced models, phylogenetic generalized linear mixed models (PGLMM) and multi-response phylogenetic mixed models (MR-PMM). Our results demonstrate that PIC-O(D)GC and PIC-MM consistently exhibit superior performance under conditions with evolutionary shifts, maintaining low false positive rates and high accuracy. In no-shift scenarios, PIC-O(D)GC and PIC-MM perform comparably to other reliable methods, underscoring their adaptability and robustness across varied datasets. Furthermore, PIC-O(D)GC offers computational efficiency and flexibility, making it particularly suitable for large-scale datasets. These findings underscore the value of adaptive, nonparametric frameworks like PIC-O(D)GC in complementing traditional phylogenetic approaches. By bridging computational simplicity with methodological robustness, O(D)GC provides a versatile tool for trait correlation analysis in both stable and dynamic evolutionary contexts.