The impact of purifying and background selection on the inference of population history: problems and prospects

Johri, Parul; Riall, Kellen; Becher, Hannes; Excoffier, Laurent; Charlesworth, Brian; Jensen, Jeffrey D.

doi:10.1101/2020.04.28.066365

Cited by 26 publications

(46 citation statements)

References 118 publications

(205 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One simple but somewhat unsatisfactory solution is to approximate the effects of background selection across the genome by scaling effective population sizes with a factor drawn from empirical estimates of the effect of background selection across the genome [50]. A better solution would be to estimate the distribution of selection coefficients as part of the model [51]. This requires a generator that can simulate selection, for example SLiM [21], but would be much more computationally intensive than the coalescent simulations in the current approach.…”

Section: Discussionmentioning

confidence: 99%

Automatic inference of demographic parameters using Generative Adversarial Networks

Wang

Kourakos

et al. 2020

Preprint

View full text Add to dashboard Cite

Population genetics relies heavily on simulated data for validation, inference, and intuition. In particular, since real data is always limited, simulated data is crucial for training machine learning methods. Simulation software can accurately model evolutionary processes, but requires many hand-selected input parameters. As a result, simulated data often fails to mirror the properties of real genetic data, which limits the scope of methods that rely on it. In this work, we develop a novel approach to estimating parameters in population genetic models that automatically adapts to data from any population. Our method is based on a generative adversarial network that gradually learns to generate realistic synthetic data. We demonstrate that our method is able to recover input parameters in a simulated isolation-with-migration model. We then apply our method to human data from the 1000 Genomes Project, and show that we can accurately recapitulate the features of real data.

show abstract

Section: Discussionmentioning

confidence: 99%

Automatic inference of demographic parameters using Generative Adversarial Networks

Wang

Kourakos

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Understanding the difference between our results and those of Johri et al (2021) also leads to the fundamental question of the link between a PSMC curve and the IICR. As outlined in Mazet et al (2016) and several subsequent studies, the IICR is a theoretical function associated to a given evolutionary model (and sampling scheme), while the PSMC is an estimation of this theoretical quantity.…”

Section: Iicr Predictions and Psmc Estimationsmentioning

confidence: 59%

“…We found that PSMC estimations fit well our predictions for large chunks (10 6 and 10 5 ), but may highlight more complex and unpredicted patterns for smaller ones (Figure S7). This may explain the poor fit of our predictions with the PSMC results of Johri et al (2021), where the heterogeneity of N e was detectable only at very small scale (≤ 55kb).…”

Section: Iicr Predictions and Psmc Estimationsmentioning

confidence: 59%

Heterogeneity in effective size across the genome: effects on the Inverse Instantaneous Coalescence Rate (IICR) and implications for demographic inference under linked selection

Boitard

Arredondo

Noûs³

et al. 2021

Preprint

View full text Add to dashboard Cite

The relative contribution of selection and neutrality in shaping species genetic diversity is one of the most central and controversial questions in evolutionary theory. Genomic data provide growing evidence that linked selection, i.e. the modification of genetic diversity at neutral sites through linkage with selected sites, might be pervasive over the genome. Several studies proposed that linked selection could be modelled as first approximation by a local reduction (e.g. purifying selection, selective sweeps) or increase (e.g. balancing selection) of effective population size (Ne). At the genome-wide scale, this leads to a large variance of Ne from one region to another, reflecting the heterogeneity of selective constraints and recombination rates between regions. We investigate here the consequences of this variation of Ne on the genome-wide distribution of coalescence times. The underlying motivation concerns the impact of linked selection on demographic inference, because the distribution of coalescence times is at the heart of several important demographic inference approaches. Using the concept of Inverse Instantaneous Coalescence Rate, we demonstrate that in a panmictic population, linked selection always results in a spurious apparent decrease of Ne along time. Balancing selection has a particularly large effect, even when it concerns a very small part of the genome. We quantify the expected magnitude of the spurious decrease of Ne in humans and Drosophila melanogaster, based on Ne distributions inferred from real data in these species. We also find that the effect of linked selection can be significantly reduced by that of population structure.

show abstract

“…Second, background selection, the loss of a linked neutral site from purifying selection on a deleterious allele, can potentially mimic patterns of selective sweeps and thus may impact the results of this study (Charlesworth et al 1993 (Johri, Riall, et al 2020;. The demographic strings calculated from PSMC used in this analysis also broadly agree in population size shape with other demographic estimates generated using other methods (e.g., Becquet and Przeworski (2007); Hey (2010)), therefore, background selection unlikely affects the demographic models used in analysis.…”

Section: Recent Attention Has Focused On Admixture Between Lineages Imentioning

confidence: 99%

Soft sweeps predominate recent positive selection in bonobos (Pan paniscus) and chimpanzees (Pan troglodytes)

Brand

White

Ting

et al. 2020

Preprint

View full text Add to dashboard Cite

Two modes of positive selection have been recognized: 1) hard sweeps that result in the rapid fixation of a beneficial allele typically from a de novo mutation and 2) soft sweeps that are characterized by intermediate frequencies of at least two haplotypes that stem from standing genetic variation or recurrent de novo mutations. While many populations exhibit both hard and soft sweeps throughout the genome, there is increasing evidence that soft sweeps, rather than hard sweeps, are the predominant mode of adaptation in many species, including humans. Here, we use a supervised machine learning approach to assess the extent of completed hard and soft sweeps in the closest living relatives of humans: bonobos and chimpanzees (genus Pan). We trained convolutional neural network classifiers using simulated data and applied these classifiers to population genomic data for 71 individuals representing all five extant Pan lineages, of which we successfully analyzed 60 individuals from four lineages. We found that recent adaptation in Pan is largely the result of soft sweeps, ranging from 73.1 to 97.7% of all identified sweeps. While few hard sweeps were shared among lineages, we found that between 19 and 267 soft sweep windows were shared by at least two lineages. We also identify novel candidate genes subject to recent positive selection. This study emphasizes the importance of shifts in the physical and social environment, rather than novel mutation, in shaping recent adaptations in bonobos and chimpanzees.

show abstract

The impact of purifying and background selection on the inference of population history: problems and prospects

Cited by 26 publications

References 118 publications

Automatic inference of demographic parameters using Generative Adversarial Networks

Automatic inference of demographic parameters using Generative Adversarial Networks

Heterogeneity in effective size across the genome: effects on the Inverse Instantaneous Coalescence Rate (IICR) and implications for demographic inference under linked selection

Soft sweeps predominate recent positive selection in bonobos (Pan paniscus) and chimpanzees (Pan troglodytes)

Contact Info

Product

Resources

About