Detecting Recent Positive Selection with a Single Locus Test Bipartitioning the Coalescent Tree

Abstract

Many population genomic studies have been conducted in the past to search for traces of recent events of positive selection. These traces, however, can be obscured by temporal variation of population size or other demographic factors. To reduce the confounding impact of demography, the coalescent tree topology has been used as an additional source of information for detecting recent positive selection in a population or a species. Based on the branching pattern at the root, we partition the hypothetical coalescent tree, inferred from a sequence sample, into two subtrees. The reasoning is that positive selection could impose a strong impact on branch length in one of the two subtrees while demography has the same effect on average on both subtrees. Thus, positive selection should be detectable by comparing statistics calculated for the two subtrees. Simulations demonstrate that the proposed test based on these principles has high power to detect recent positive selection even when DNA polymorphism data from only one locus is available, and that it is robust to the confounding effect of demography. One feature is that all components in the summary statistics ([Formula: see text]) can be computed analytically. Moreover, misinference of derived and ancestral alleles is seen to have only a limited effect on the test, and it therefore avoids a notorious problem when searching for traces of recent positive selection.

A new inference method for detecting an ongoing selective sweep

“…Although the distribution of F c is generally very broad under neutrality, it may be claimed that F c < f r can be expected only when recombination is rare or absent, because only particular branches in a coalescent tree can produce a cluster of chromosomes with specified frequency f r (Fu, 1995;Li, 2011;Ferretti et al, 2017;Yang et al, 2018) and mutations in the common stem lineage are not counted in F c . To illustrate this, let us consider a simple case of n = 4 ( Supplementary Fig.…”

“…Li (2011) proposed an alternative method that uses the maximum frequency of derived alleles in a sample as an indicator of unbalanced coalescent trees. This method relies on the fact that the probability of unbalanced basal branches is extremely low under neutrality and is independent of temporal changes in population size (see also Yang et al, 2018 for an extension to a varying size population, and Ferretti et al, 2017, who provided a systematic analysis of the impact of the waiting times and the branching order of coalescent events on the SFS). One challenge, however, arises from the fact that the required imbalance must be strong in a large sample.…”

Adaptation in structured populations and fuzzy boundaries between hard and soft sweeps

“…A large array of statistical methods is available, and is routinely applied in genome annotation studies, to identify genomic locations which supposedly have experienced recent selective sweeps (reviewed by [8]). These methods can roughly be grouped into three types, or mixtures of them, according to their underlying theoretical considerations: there are statistical tests for sweeps which are based on the site frequency spectrum (SFS) (see [9]), tests which are based on haplotypes and their distribution [10], [11], and tests based on properties of the inferred genealogical tree of a sequence sample [12], [13]. Some of the well established methods such as the likelihood ratio test (LRT) of e.g., [14], [15], linkage-disequilibrium (LD) [16] and Bayesian or Hidden-Markov-Model- (HMM-)based methods [17], [18], [19] can be considered as combinations of these types.…”

Sweeps in time: leveraging the joint distribution of branch lengths

“…It will be interesting to ask how much information is contained in the distribution of topologies, allowing us to extend the GF to the larger sample sizes required by these methods. Several summary statistics have been developed to diagnose the effect of sweeps on genealogical topologies (Li and Wiehe, 2013; Yang et al, 2018). This research is motivated by the fact that statistics like root imbalance are invariant to population size changes.…”

“…Several summary statistics have been developed to diagnose the effect of sweeps on genealogical topologies (Li and Wiehe, 2013;Yang et al, 2018). This research is motivated by the fact that statistics like root imbalance are invariant to population size changes.…”