The study by Teng et al (2017) provides a detailed population genomic analysis of the Brown Norway rat and its sibling species, Rattus nitidus, to reconstruct their evolutionary history. The main findings reveal that the speciation event separating the two rat species likely occurred during the drastic climatic changes of the Middle Pleistocene. Following this divergence, the researchers uncovered evidence of widespread and geographically significant gene flow, or introgression, from R. nitidus into Brown Norway rat populations. Some of these introgressed genes, particularly those related to chemical communication, appear to have been adaptive. This could have potentially contributed to the Brown Norway rat’s remarkable success as a global colonizer. Also, the study identified signatures of positive selection in genes related to metabolism and immune response, further explaining the rat’s adaptability.

To arrive at these conclusions, the authors employed a comprehensive suite of computational methods appropriate for whole-genome data. After sequencing 51 Brown Norway rats, they used standard bioinformatics pipelines, including the Genome Analysis Toolkit (GATK), to identify millions of genetic variants. To investigate population relationships, they used a combination of Principal Component Analysis (PCA) and the model-based clustering program Admixture. The demographic history and divergence timing were inferred using the Pairwise Sequentially Markovian Coalescent (PSMC) model and the Bayesian coalescent-based tool G-PhoCS. To specifically test for and localize gene flow, they used Patterson’s D-statistics and the modified f-statistic (ƒd), which are designed to detect imbalances in shared genetic ancestry. In order to identify regions of the genome under positive selection, they used a cross-population composite likelihood ratio (XP-CLR) test.

I would have used the same analytical framework as the authors. The chosen methods represent a logical and hardy progression for addressing the study’s core questions. Using whole-genome sequencing was necessary to capture a comprehensive picture of variation, and the combination of PCA and Admixture is a standard and powerful approach for defining population structure without strong prior assumptions. The use of D-statistics and ƒd was an appropriate strategy. These are the gold-standard methods for detecting and quantifying introgression, a central finding of the paper. The decision to integrate multiple lines of evidence from demographic modeling, introgression tests, and selection scans creates a more compelling and well-supported evolutionary narrative than any single analysis could provide.

One of the main strengths of the study was its use of simulations to provide a null hypothesis for comparison. To determine which genomic regions showed statistically significant evidence of selection, the researchers needed to know what patterns of genetic variation would be expected by chance under the species’ specific demographic history. They used a whole-genome simulation program called ARGON to generate genomic data under a neutral model, with parameters informed by their own demographic inferences from PSMC and G-PhoCS. By running their selection scan (XP-CLR) on this simulated neutral data, they could establish a reliable significance threshold. This allowed them to confidently distinguish true selective sweeps in their real rat data from patterns that could have arisen simply due to random genetic drift within their inferred population history.

References

Salojärvi, J. (2019). Computational tools for population genomics. In O. P. Rajora (Ed.), Population genomics: Concepts, approaches and applications (pp. 127–152). Cham, Switzerland: Springer Nature Switzerland AG.

Teng, H., Zhang, Y., Shi, C., Mao, F., Cai, W., Lu, L., Zhao, F., Sun, Z., & Zhang, J. (2017). Population Genomics Reveals Speciation and Introgression between Brown Norway Rats and Their Sibling Species. Molecular biology and evolution, 34(9), 2214–2228. https://doi.org/10.1093/molbev/msx157