Comparison of Single Genome and Allele Frequency Data Reveals Discordant Demographic Histories

Abstract

Inference of demographic history from genetic data is a primary goal of population genetics of model and nonmodel organisms. Whole genome-based approaches such as the pairwise/multiple sequentially Markovian coalescent methods use genomic data from one to four individuals to infer the demographic history of an entire population, while site frequency spectrum (SFS)-based methods use the distribution of allele frequencies in a sample to reconstruct the same historical events. Although both methods are extensively used in empirical studies and perform well on data simulated under simple models, there have been only limited comparisons of them in more complex and realistic settings. Here we use published demographic models based on data from three human populations (Yoruba, descendants of northwest-Europeans, and Han Chinese) as an empirical test case to study the behavior of both inference procedures. We find that several of the demographic histories inferred by the whole genome-based methods do not predict the genome-wide distribution of heterozygosity, nor do they predict the empirical SFS. However, using simulated data, we also find that the whole genome methods can reconstruct the complex demographic models inferred by SFS-based methods, suggesting that the discordant patterns of genetic variation are not attributable to a lack of statistical power, but may reflect unmodeled complexities in the underlying demography. More generally, our findings indicate that demographic inference from a small number of genomes, routine in genomic studies of nonmodel organisms, should be interpreted cautiously, as these models cannot recapitulate other summaries of the data.

Genotype‐free estimation of allele frequencies reduces bias and improves demographic inference from RADSeq data

“…The timing of population divergence as inferred here thus corresponds remarkably well with the start of differences in population size trajectories inferred using PSMC analysis (Nadachowska-Brzyska et al, 2016). Estimates of effective population sizes, both pre-and post-split, vary more widely between the two approaches, consistent with previously reported discrepancies between estimates of effective population sizes between inference methods (e.g., Beichman, Phung, & Lohmueller, 2017; Mazet, Rodríguez, Grusea, Boitard, & Chikhi, 2016).…”

A practical introduction to sequentially Markovian coalescent methods for estimating demographic history from genomic data

“…Finally, there are reasons to believe that the sequentially Markovian coalescent might perform poorly on realistic data. For example, when genetic data are produced by simulation under demographic models inferred by MSMC from human genomes, they fail to resemble the empirical data in important ways (Beichman, Phung, & Lohmueller, 2017). Other methods that use data from many individuals, such as δaδi (Gutenkunst, Hernandez, Williamson, & Bustamante, 2009) and SMC++ (Terhorst, Kamm, & Song, 2017), perform substantially better in this regard.…”

A community-maintained standard library of population genetic models

“…Importantly, there are clear examples of different methods yielding fundamentally different conclusions. For example, Markovian coalescent methods applied to human genomes have suggested large ancient (> 100,000 years ago) ancestral population sizes and bottlenecks that have not been detected by other methods based on allele frequency spectra (see Beichman et al, 2017 ). These distinct methods differ in how they model, summarize, and optimize fit to genetic variation data, suggesting that such design choices can greatly affect the performance of the inference.…”