Selfing-aware demographic inference highlights local population structure and bottlenecks in Hawaiian Caenorhabditis elegans

Demographic inference, a method used to analyze the evolutionary history of species and populations, relies on a model known as Sequentially Markovian coalescent (SMC). This model is particularly useful for understanding selfing, a common reproductive mode observed in both invertebrates and plants. However, the genetic impacts of self-fertilization, such as decreased overall genetic diversity and effective population size (Ne), are typically overlooked in demographic inference, which can lead to inaccuracies.

To address this issue, researchers applied a selfing-aware demographic inference method called eSMC2 to Hawaiian populations of Caenorhabditis elegans, a nematode species that predominantly self-fertilizes. Utilizing an expanded dataset containing a wider range of Hawaiian strains than previous studies, the researchers were able to identify previously reported genetic groups, including a significant portion of strains originating from the Big Island.

They also detected population substructure among these Big Island subpopulations, emphasizing the need to identify these subgroups to prevent misinterpretations of demography. The study found high selfing rates in different Big Island subpopulations, consistent with earlier findings, and inferred consistent population declines. Importantly, the demographic estimates from eSMC2 were smaller compared to those obtained from another method called PSMC, which does not account for selfing.

However, rescaling PSMC parameters to incorporate self-fertilization resulted in a similar demographic trajectory as the eSMC2 inference. This study demonstrates the importance of considering selfing effects when using SMC-based methods and reveals that the Hawaiian group experienced population declines similar to those observed in other regions outside Hawaii.