Connectivity and movement patterns of populations are influenced by past and present environmental and biotic factors, which are reflected in genetic relatedness among populations. Methods that estimate the "commute time" between populations based on electrical resistance (i.e., isolation-by-resistance [IBR]) have been widely applied to either infer movement patterns directly from environmental factors or detect possible barriers to gene flow given empirical genetic relatedness. Yet, the commute time is only equivalent to the coalescence time between populations under symmetric migration with isotropic landscapes. Asymmetric gene flow is relatively common when populations are expanding, retreating, or with source-sink dynamics. In a From the Cover paper in this issue of Molecular Ecology Resources, Lundgren and Ralph (Molecular Ecology Resources, 19, 2019) describe a Bayesian method to infer bidirectional gene flow rates and population sizes without the assumption of symmetry. The method shows great accuracy in connectivity estimations under symmetric, as well as asymmetric gene flow scenarios where resistance methods fail. However, computational complexity limits the method to a few populations, preventing its application to finescale environmental maps. Also, as a discrete-deme static model, the inferred differences in gene flow rates are sensitive to population discretization and cannot be directly used to differentiate among processes (e.g., past expansion vs. current barrier). Here, we discuss scenarios where the new method can best be utilized and provide potential directions to identify the underlying processes causing deviations in gene flow patterns.

中文翻译：

---

人群何时不像电路那样连接？识别从合并时代开始的基因流偏向。

人口的连通性和流动方式受过去和现在的环境和生物因素影响，这反映在人口之间的遗传相关性中。基于电阻（即电阻隔离[IBR]）估算种群之间“通勤时间”的方法已广泛应用于直接从环境因素推断运动模式或在经验遗传相关的情况下检测基因流的可能障碍。然而，通勤时间仅等于各向同性景观下对称迁移下种群之间的合并时间。当种群正在扩展，后退或具有源库动态时，不对称基因流相对普遍。在本期《分子生态资源》的封面文章中，Lundgren和Ralph（Molecular Ecology Resources，19，2019）描述了一种贝叶斯方法，可以在不假设对称性的情况下推断双向基因流速和群体大小。该方法在对称方法以及抗性方法失败的非对称基因流情况下的连通性估计中显示出很高的准确性。但是，计算复杂性将方法限制在少数人群中，从而使其无法应用于精细比例的环境图。同样，作为离散行为静态模型，推断的基因流速差异对群体离散敏感，不能直接用于区分过程（例如，过去的扩展与当前的障碍）。这里，