Robust inference of population size histories from genomic sequencing data

Abstract

Unraveling the complex demographic histories of natural populations is a central problem in population genetics. Understanding past demographic events is of general anthropological interest, but is also an important step in establishing accurate null models when identifying adaptive or disease-associated genetic variation. An important class of tools for inferring past population size changes from genomic sequence data are Coalescent Hidden Markov Models (CHMMs). These models make efficient use of the linkage information in population genomic datasets by using the local genealogies relating sampled individuals as latent states that evolve along the chromosome in an HMM framework. Extending these models to large sample sizes is challenging, since the number of possible latent states increases rapidly.Here, we present our method CHIMP (CHMM History-Inference Maximum-Likelihood Procedure), a novel CHMM method for inferring the size history of a population. It can be applied to large samples (hundreds of haplotypes) and only requires unphased genomes as input. The two implementations of CHIMP that we present here use either the height of the genealogical tree (T MRCA ) or the total branch length, respectively, as the latent variable at each position in the genome. The requisite transition and emission probabilities are obtained by numerically solving certain systems of differential equations derived from the ancestral process with recombination. The parameters of the population size history are subsequently inferred using an Expectation-Maximization algorithm. In addition, we implement a composite likelihood scheme to allow the method to scale to large sample sizes.We demonstrate the efficiency and accuracy of our method in a variety of benchmark tests using simulated data and present comparisons to other state-of-the-art methods. Specifically, our implementation using T MRCA as the latent variable shows comparable performance and provides accurate estimates of effective population sizes in intermediate and ancient times. Our method is agnostic to the phasing of the data, which makes it a promising alternative in scenarios where high quality data is not available, and has potential applications for pseudo-haploid data.

Joint inference of evolutionary transitions to self-fertilization and demographic history using whole-genome sequences

“…Our results suggest that population-size decline could be linked to the transition to selfing and that our demographic inference using teSMC should be more reliable than those from eSMC (and other SMC approach ignoring the selfing transition). A way forward would be to improve our inference framework using the new theoretical results by Upadhya and Steinrücken, 2022 , which focuses on modeling the time to the most recent common ancestor of a sample size bigger than two. This study demonstrates that the inference in the far past can be improved, at the cost of losing accuracy in recent times.…”

Biases in ARG-based inference of historical population size in populations experiencing selection

“…In addition, extreme rates of recombination (cM/Mb ≤ 0.01, cM/Mb ≥ 5) have been shown to result in mis-inference of historical population size as calculated from nucleotide diversity using Relate (Novo et al 2022). However, Upadhya and Steinrücken (2022) suggested that demographic inference performance with Relate improves when recombination rates are variable due to the presence of low recombination rate regions from which exact local genealogies can be estimated, which could in part explain the moderate improvement provided by modelling mutation and recombination rate heterogeneity in D. melanogaster simulated populations (Figure S3). It is unclear whether recombination and mutation rate heterogeneity may have more substantial effects in different species, as recent reports suggest masking low-recombination regions in songbird species may improve demographic inference with Relate (Ishigohoka and Liedvogel 2024).…”

“…Systematic ARG estimation using exhaustive likelihood approaches is computationally impractical (McVean and Cardin 2005), and as a result, Relate and similar methods use heuristics and HMMs to minimise redundant calculations (Li and Stephens 2003), greatly improving scalability with large sample sizes at the cost of precision. Relate (Speidel et al 2019) and tsinfer/tsdate (Kelleher et al 2019; Wohns et al 2022) share biases when estimating recent and ancient coalescence times (Brandt et al 2022; Upadhya and Steinrücken 2022), where accuracy drops when compared to e.g . the less scalable, more precise ARGweaver (Rasmussen et al 2014), likely due to factors such as under-sampling of marginal trees (Deng et al 2021).…”

Simultaneous Inference of Past Demography and Selection from the Ancestral Recombination Graph under the Beta Coalescent

“…With the rise of SMC approaches [53], most current methods leverage information from whole genome sequences by simultaneously reconstructing a portion of the ARG in order to infer past demographic history [53, 77, 88, 89], migration rates [46, 90], variation in recombination and mutation along the genome [5, 4], as well as ecological life history traits such as selfing or seed banking [80, 86]. However in previous studies authors discussed uncoupling both steps, namely reconstructing the ARG and then inferring parameters from its distribution [81, 31, 68].…”