Wenjie Zhu

Research Interests

My research lies at the interface of phylogenetics and population genetics. I develop and apply computational and statistical methods for species tree inference and species delimitation using genomic data.

Species delimitation in the genomic era

Species is the basic unit of biological studies. Many research fields, from ecology to conservation biology, heavily rely on species definitions. Yet defining species boundaries remains challenging for many organisms, particularly those that are morphologically cryptic or recently diverged. Genomic data offer high-resolution information for addressing this challenge. In this project, we extend polymorphism-aware phylogenetic models (PoMos) for species delimitation. PoMos operate at the interface of phylogenetics and population genetics, explicitly modeling allele frequency dynamics while accounting for within-population variation during phylogenetic inference. Combined with Bayes factors, we are able to compare different species delimitation hypotheses under the PoMo framework. European Luciola fireflies are morphologically difficult to distinguish, making species delimitation challenging using traditional approaches. We use genome-wide single nucleotide polymorphisms (SNPs), applying population genetic methods to infer biogeographic history and PoMo-based approaches for species delimitation in this taxonomically challenging group.

Species tree inference under the multispecies coalescent model

A robust species phylogeny is fundamental to understanding evolutionary history, biodiversity patterns, and comparative biology. Molecular data have become the primary source of information for species tree inference. Phylogenomic datasets commonly include hundreds to thousands of genes. However, individual genes often tell conflicting evolutionary stories. This discordance can arise from biological processes such as incomplete lineage sorting (ILS), where ancestral genetic variation persists through speciation events. Gene tree estimation uncertainty, caused by limited phylogenetic signals or systematic errors, further compounds this challenge. The multispecies coalescent (MSC) model provides a framework for inferring species trees while accounting for ILS-induced gene tree discordance. However, existing MSC methods face a fundamental trade-off. Joint full-likelihood approaches are statistically rigorous but computationally prohibitive for large datasets. Computationally efficient summary methods often ignore gene tree uncertainty. My research seeks to develop and evaluate methods that balance computational feasibility with statistical accuracy for robust species tree inference.

Zhu W and Höhna S. Two-step Species Tree Inference under the Multispecies Coalescent Using Full-Likelihood. Evolutionary Journal of the Linnean Society. https://doi.org/10.1093/evolinnean/kzaf018

Zhu W, Chen L, van der Burgt XM, Haba PM, Garaeva D, Lau P, Morales-Briones DF, Veranso-Libalah MC. Rediscovery and phylogenomic placement of Feliciadamia stenocarpa (Feliciadamieae: Melastomataceae): assessing gene tree discordance. Botanical Journal of the Linnean Society. https://doi.org/10.1093/botlinnean/boaf047

Publikationsliste