CoChat
Sign Up
Cited by 7Evolution and Genetic DynamicsGenetic diversity and population structureGenomics and Phylogenetic Studies

Finding Hybrid Incompatibilities Using Genome Sequences from Hybrid Populations

Open access
Blanckaert, Alexandre, Payseur, Bret A·2021·Molecular Biology and Evolution
DOI 10.1093/molbev/msab168PMC PMC8476132
Open this paper in CoChat — free accountOpen PDF

Abstract

Natural hybrid zones offer a powerful framework for understanding the genetic basis of speciation in progress because ongoing hybridization continually creates unfavorable gene combinations. Evidence indicates that postzygotic reproductive isolation is often caused by epistatic interactions between mutations in different genes that evolved independently of one another (hybrid incompatibilities). We examined the potential to detect epistatic selection against incompatibilities from genome sequence data using the site frequency spectrum (SFS) of polymorphisms by conducting individual-based simulations in SLiM. We found that the genome-wide SFS in hybrid populations assumes a diagnostic shape, with the continual input of fixed differences between source populations via migration inducing a mass at intermediate allele frequency. Epistatic selection locally distorts the SFS as non-incompatibility alleles rise in frequency in a manner analogous to a selective sweep. Building on these results, we present a statistical method to identify genomic regions containing incompatibility loci that locates departures in the local SFS compared with the genome-wide SFS. Cross-validation studies demonstrate that our method detects recessive and codominant incompatibilities across a range of scenarios varying in the strength of epistatic selection, migration rate, and hybrid zone age. Our approach takes advantage of whole genome sequence data, does not require knowledge of demographic history, and can be applied to any pair of nascent species that forms a hybrid zone.

AI snapshot

Written from the paper’s full text.

TL;DR

The paper shows that hybrid populations have a diagnostic genome-wide site frequency spectrum (SFS) with a mass at intermediate allele frequency, and that epistatic selection against hybrid incompatibilities locally distorts this SFS. It introduces a multinomial/bootstrap outlier method that can locate incompatibility loci from whole-genome sequence data without specifying demographic history.

Key findings

  • In neutral hybrid populations, the SFS has a large proportion of SNPs at frequency 0.5 due to fixed differences between parental populations; migration maintains this intermediate-frequency mass, while drift spreads it around 0.5.
  • Epistatic selection against incompatible alleles locally distorts the SFS, with the distortion extending approximately 10–20 Mb around incompatibility loci in the default scenario.
  • With power as the optimizing metric, the method detects incompatibility loci with power 0.958 for the default codominant scenario and 0.910 for the default recessive scenario; cross-validation with power/(2*d+1) favored {d=9, thr1=900, thr2=80}, giving 9.5 Mb resolution and power near 50% across many scenarios.
  • High migration (m=0.05) causes failure to detect incompatibility loci; even with very strong selection, detection was only about 40% and resolution was low. Single-locus selection produces similar SFS distortions, so the method alone cannot exclude single-locus selection as an explanation.

Methods

  • Individual-based simulations in SLiM v3.4 modeled one hybrid population (N=10,000) formed from two parental populations (Np=5,000 each) that diverged for 50,000 generations with fixed differences at two loci; migration into the hybrid population was m=0.005 from each parental population by default.
  • Genomes were one chromosome of 500 Mb with mutation rate μ=10^-9 and recombination rate r=5×10^-9 per bp; incompatibility loci were at 175 and 325 Mb, with epistatic fitness effect ε=−0.1 by default, and samples of 50 diploid individuals were taken at generations 100, 1,000, 5,000, and 10,000.
  • The method computed the unfolded SFS in non-overlapping 500 kb windows, calculated the probability of each local SFS given the genome-wide SFS using a multinomial distribution, and used 1,000 bootstrap replicates to identify outlier windows in the lower 1% tail, with thresholds thr1 and thr2 and neighborhood size d.

Limitations

  • The paper states that the method cannot distinguish epistatic selection from single-locus selection when applied in isolation, and that high migration prevents detection of incompatibility loci.
  • The paper states that resolution is limited: incompatibility loci are associated with regions of about 9–15 Mb depending on whether power or resolution is optimized, though most outliers are found much closer to the locus.
  • The paper states that simulations assumed incompatibility alleles were fixed in source populations, ignored genotyping error and low-coverage SFS bias, and did not include more complex migration, structured source populations, asymmetrical migration, migration out of hybrid populations, or demographic departures from mutation-drift equilibrium; the text also makes evident that the model is limited to two-locus BDM incompatibilities and requires experimental validation for confirmed genes.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1093/molbev/msab168

Cites

59 of 59 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

8 of 20 cited sources checked · 5 supported · 0 flagged · accuracy not yet measured

  • Interpretationcitation

    Neutral mutations linked to each compatible allele should spread through the population in a manner partly analogous to a selective sweep.

    “Neutral mutations linked to each compatible allele should also spread through the population in a manner partly analogous to a selective sweep (Maynard Smith and Haigh 1974).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Maynard Smith and Haigh 1974 source

      The Introduction (page unavailable) says alleles linked to a favourable mutation tend to increase in frequency as that mutation becomes fixed; §2 applies this to neutral alleles. This supports the selective-sweep analogy, but the paper does not establish that each compatible allele will spread or carry linked neutral mutations with it. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
  • Interpretationreasoning

    The SFS of polymorphisms in genomic regions containing hybrid incompatibilities is expected to depart from the genome-wide SFS.

    “Therefore, we might expect the site frequency spectrum (SFS) of polymorphisms in genomic regions containing hybrid incompatibilities to depart from the genome-wide SFS (Braverman et al. 1995; Simonsen et al. 1995).”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites Braverman et al. 1995 source

      The introductory passage (page unavailable) predicts that simple genetic hitchhiking skews the site frequency spectrum of linked polymorphisms toward rare variants. It does not examine genomic regions containing hybrid incompatibilities or compare their spectra with a genome-wide SFS, so the claim extends the paper’s prediction to a different setting. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites Simonsen et al. 1995 source
  • Findingown data

    Epistatic selection locally distorts the SFS as non-incompatibility alleles rise in frequency in a manner analogous to a selective sweep.

    “Epistatic selection locally distorts the SFS as non-incompatibility alleles rise in frequency in a manner analogous to a selective sweep.”
    • Also stated in: § Characterization of the SFS in a Hybrid Population · plain
      “Epistatic selection against hybrid genotypes shifts the SFS in the vicinity of incompatibility loci (fig. 2A for the default scenario; see supplementary figs. S2−S17, Supplementary Material online for other scenarios).”
    Quote in the paperSubstantial claim
  • Findingreasoning

    Epistatic selection against hybrid genotypes shifts the SFS in the vicinity of incompatibility loci.

    “Epistatic selection against hybrid genotypes shifts the SFS in the vicinity of incompatibility loci (fig. 2A for the default scenario; see supplementary figs. S2−S17, Supplementary Material online for other scenarios).”
    Quote in the paperSubstantial claim
  • Findingreasoning

    The probability of observing an SFS near an incompatibility locus is many orders of magnitude lower than the probability of observing one that is similar to the global SFS.

    “Nevertheless, the probability of observing an SFS near an incompatibility locus is many orders of magnitude lower than the probability of observing one that is similar to the global SFS (fig. 2B).”
    Quote in the paperSubstantial claim
  • Classified as probably not a substantial claim. 22
  • May combine two claims. 37

  • Classified as probably not a substantial claim.

    Premisecitation

    The Bateson-Dobzhansky-Muller (BDM) model postulates that reproductive isolation arises when geographically separate populations fix incompatible mutations in genes that normally interact.

    “A popular explanation for postzygotic isolation is the Bateson−Dobzhansky−Muller (BDM) model (Bateson 1909; Dobzhansky 1936; Muller 1942). The model postulates that reproductive isolation arises when geographically separate populations fix incompatible mutations in genes that normally interact.”
    Quote in the paperSubstantial claim1 of 3 sources checked
    • cites Bateson 1909

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Dobzhansky 1936 source

      The opening section (pp. 113–114) proposes that complementary genetic factors from two parental forms can interact to make their hybrid sterile, supporting the incompatibility mechanism. Our copy does not specify that the populations were geographically separate, that they fixed mutations, or that the implicated genes normally interact. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites Muller 1942

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The faster-than-linear accumulation of hybrid incompatibilities with divergence time has been recovered in multiple taxa including Solanum, Drosophila, and Mus musculus.

    “this result has been recovered in multiple taxa: Including Solanum (Moyle and Nakazato 2010; Guerrero et al. 2017), Drosophila (Matute et al. 2010), or Mus musculus (Wang et al. 2015).”
    Quote in the paperNot a substantial claim1 of 3 sources checked
    • cites Moyle and Nakazato 2010 source
    • cites Guerrero et al. 2017 source
    • cites Matute et al. 2010 source

  • May combine two claims.

    Premisecitation

    Many known examples of incompatible genes involve strongly diverged species, presenting a risk that they arose after the speciation process was complete.

    “many of the known examples of incompatible genes involve strongly diverged species (Powell et al. 2020), presenting a risk that they arose after the speciation process was complete.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Powell et al. 2020 source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Although a certain number of candidate genes were identified, the authors pointed out that single-locus selection could not be excluded as an alternative explanation since selective sweeps can generate similar patterns.

    “Although a certain number of candidate genes were identified, the authors pointed out that single-locus selection could not be excluded as an alternative explanation since selective sweeps can generate similar patterns.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Beissinger et al. 2016 source

  • May combine two claims.

    Premisecitation

    Genotyping error could reduce power and increase the false-positive rate, especially for low-coverage sequencing, which can lead to biased reconstructions of the SFS.

    “Genotyping error could reduce power and increase the false-positive rate, especially for low-coverage sequencing, which can lead to biased reconstructions of the SFS (Han et al. 2014).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Han et al. 2014 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Beissinger et al. (2016) computed the DIS′2 statistic, a measure that partitions linkage disequilibrium between and within populations.

    “Beissinger et al. (2016) computed the DIS′2 statistic (Ohta 1982a, 1982b), a measure that partitions linkage disequilibrium between and within populations.”
    Quote in the paperNot a substantial claim1 of 3 sources checked
    • cites Beissinger et al. 2016 source
    • cites Ohta 1982a

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Ohta 1982b

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Findingreasoning

    When initially formed, the hybrid population possesses a unique SFS characterized by a large proportion of SNPs at frequency 0.5; this proportion depends on the level of divergence between the two parental populations.

    “Indeed, when initially formed, the hybrid population possesses a unique SFS (fig. 1B), characterized by a large proportion of SNPs at frequency 0.5; this proportion depends on the level of divergence between the two parental populations.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingreasoning

    Although the SFS in the parental populations does not change, the part of the SFS in the hybrid population generated by migration changes as the parental populations continue to diverge and accumulate fixed differences.

    “In addition, it is worth noting that although the SFS in the parental populations does not change, the part of the SFS in the hybrid population generated by migration (equivalent to fig. 1B) changes as the parental populations continue to diverge and accumulate fixed differences.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingnone given

    Due to chance, some regions of the chromosome may have a local SFS that differs from the rest of the chromosome.

    “However, due to chance, some regions of the chromosome may have a local SFS that differs from the rest of the chromosome (e.g., in fig. 2B to the left of locus A, around position 125 Mb).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingreasoning

    The constant influx of fixed differences between parental populations into the hybrid zone via migration helps maintain the proportion of SNPs with frequency 0.5, whereas drift spreads the distribution around 0.5.

    “The constant influx of fixed differences between parental populations into the hybrid zone via migration helps to maintain the proportion of SNPs with frequency 0.5, whereas drift spreads the distribution around 0.5 (fig. 1D).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    A similar distortion of the SFS is generated by single-locus selection but this distortion is stronger (assuming s=ε) because single-locus selection acts independently of the genetic background and therefore is always present.

    “A similar distortion of the SFS is generated by single-locus selection but this distortion is stronger (assuming s=ϵ) because single-locus selection acts independently of the genetic background (supplementary fig. S3, Supplementary Material online) and therefore is always present.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    After 1,000 generations, the isolated hybrid population still displays an excess of alleles at intermediate frequency, though this effect is far more diffuse than in the presence of migration and almost vanishes after 10,000 generations.

    “After 1,000 generations, the isolated hybrid population still displays an excess of alleles at intermediate frequency (fig. 1C), though this effect is far more diffuse than in the presence of migration and almost vanishes after 10,000 generations (supplementary fig. S1, Supplementary Material online).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingnone given

    About half of the mutations at frequency 0.5 originated from population 1 and are in strong linkage disequilibrium with allele A (or b), whereas the other half originated from population 2 and are in strong association with allele a (or B).

    “About half of the mutations at frequency 0.5 originated from population 1 and are in strong linkage disequilibrium with allele A (or b), whereas the other half originated from population 2 and are in strong association with allele a (or B).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    The second most common combination is {d=9,thr1=900,thr2=80}, which is very similar and confirms the result.

    “The second most common combination, {d=9,thr1=900,thr2=80}, is really similar, therefore confirming our result.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In the default scenario, the window can be reduced by approximately two-thirds to 6.5 Mb with an associated decrease by half in power from 0.958 to 0.486.

    “For example, in the default scenario, the window can be reduced by approximately two-thirds to 6.5 Mb with an associated decrease by half in power from 0.958 to 0.486.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingreasoning

    When epistatic selection is strong, the focal outlier is usually detected further away from the positions of incompatibility loci; this is also true for single-locus selection.

    “When epistatic selection is strong, the focal outlier is usually detected further away from the positions of incompatibility loci (fig. 4A; this is also true for single-locus selection in supplementary fig. S25, Supplementary Material online).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingreasoning

    Although there is a relationship between the number of SNPs and the (log) probability of the local SFS, windows with the lowest probabilities are not those with the most SNPs, whether they evolve under neutrality, single-locus selection, or epistatic selection.

    “Although we recover such a relationship between the number of SNPs and the (log) probability of the local SFS, windows with the lowest probabilities are not those with the most SNPs, whether they evolve under neutrality (supplementary fig. S19A, Supplementary Material online), single-locus selection (supplementary fig. S19B, Supplementary Material online), or epistatic selection (supplementary fig”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingreasoning

    Selection distorts the local SFS; alleles with frequencies at or near 0.5 are more strongly affected because many are in tight linkage disequilibrium with incompatibility alleles.

    “Selection distorts the local SFS (fig. 2A; supplementary figs. S3−S17, Supplementary Material online); alleles with frequencies at or near 0.5 are more strongly affected because many of them are in tight linkage disequilibrium with incompatibility alleles.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingreasoning

    Even when selection against hybrids is extremely strong (F1 hybrid fitness reduced by 50% in the codominant case, and by 94% for the double homozygote AABB), incompatibility loci are detected only 40% of the time and with low resolution (in a 16.5 Mb region).

    “Even when selection against hybrids is extremely strong (F1 hybrid fitness is reduced by 50% in the codominant case, and by 94% for the double homozygote AABB), incompatibility loci are detected only 40% of the time and with a rather low resolution (in a 16.5 Mb region).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    A total of 172 outliers are found in 130 of the 400 cases, but in 79 of the 100 replicates.

    “A total of 172 outliers are found in 130 of the 400 cases, but in 79 of the 100 replicates.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    As long as epistatic selection is not too weak, the method detects loci with power near 50% while maintaining a false positive rate below 5% (corrected for multiple testing) for both recessive and codominant incompatibilities.

    “As long as epistatic selection is not too weak, we are able to detect the loci with power near 50%, whereas maintaining a false positive rate below 5% (corrected for multiple testing, see Materials and Methods), for both recessive incompatibilities and codominant incompatibilities.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    After bootstrap analysis, the signal from a neutral region around 125 Mb partially vanishes: it appears in the 1% tail about 40% of the time and is isolated.

    “After bootstrap analysis, the signal partially vanishes: First this region appears in the 1% tail about 40% of the time; second, the signal is isolated.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    When using power as the optimizing metric, the best combination of criteria is {d=16,thr1=400,thr2=40} and the power achieved is far greater than when using power/(2*d+1), reaching above 95% in some cases.

    “When using power as the optimizing metric, the best combination of criteria is {d=16,thr1=400,thr2=40} and the power achieved is far greater than when using power/(2*d+1) as the optimizing metric, reaching above 95% in some cases”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the absence of epistatic selection, the focal loci are detected in 3.25% of cases, which is less than the 6.8% expected assuming the position of outliers is independently and uniformly distributed along the chromosome.

    “In the absence of epistatic selection, the focal loci are detected in 3.25% of the cases, which is less than expected assuming the position of outliers is independently and uniformly distributed along the chromosome (6.8%).”
    Quote in the paperNot a substantial claim
    • The authors state this as an assumption: The position of outliers is independently and uniformly distributed along the chromosome.

  • Probably not a substantial claim: may not pass S4 (based).

    Findingnone given

    The method performs relatively well across a wide range of epistatic selective strength.

    “In addition, the method performs relatively well across a wide range of epistatic selective strength.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    When ignoring the effect of d on power, the method obtains relatively high power across multiple scenarios, finding the incompatibility loci more than 90% of the time, for both recessive and codominant incompatibilities.

    “When we ignore the effect of d on power, we obtain relatively high power across multiple scenarios, finding the incompatibility loci more than 90% of the time, for both recessive and codominant incompatibilities.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Using a common combination of {d,thr1,thr2} and the power/(2*d+1) optimization metric does not reduce the ability to detect the antagonistic scenario in the codominant case but does for the recessive case.

    “Using a common combination of {d,thr1,thr2} and the power/(2*d+1) optimization metric does not reduce our ability to detect this scenario in the codominant case (table 2) but it does for the recessive case.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    About 65% of loci under strong single-locus selection are detected, a rate similar to the strong epistatic selection scenario.

    “About 65% of loci under strong single-locus selection are detected, a rate similar to the strong epistatic selection scenario.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The bootstrap analysis effectively removes some windows with extreme probabilities while retaining windows adjacent to incompatibility loci.

    “The bootstrap analysis effectively removes some windows with extreme probabilities while retaining windows adjacent to incompatibility loci.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    When a heterospecific combination of alleles at two loci decreases hybrid fitness, selection reduces the number of individuals with this genotype, and the population frequencies of the compatible alleles increase.

    “When a heterospecific combination of alleles at two loci decreases hybrid fitness, selection reduces the number of individuals with this genotype. As a result, the population frequencies of the compatible alleles increase.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    With epistatic selection against incompatible alleles A and B, genetic backgrounds containing the ancestral a allele or the ancestral b allele have a marginal fitness advantage, and derived alleles at neutral SNPs linked to a or b increase in frequency away from 0.5, creating a detectable local signature in the SFS.

    “With epistatic selection against incompatible alleles A and B, genetic backgrounds containing the ancestral a allele or the ancestral b allele have a marginal fitness advantage. As a result, derived alleles at neutral SNPs linked to a or b increase in frequency away from 0.5, creating a detectable, local signature in the SFS.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationnone given

    The approach can be applied to any pair of nascent species that forms a hybrid zone.

    “Our approach takes advantage of whole genome sequence data, does not require knowledge of demographic history, and can be applied to any pair of nascent species that forms a hybrid zone.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisereasoning

    Hybrids carrying incompatible combinations suffer reduced viability, reduced fertility, or both.

    “Hybrids carrying incompatible combinations (referred to as Dobzhansky−Muller incompatibilities or hybrid incompatibilities) suffer reduced viability, reduced fertility, or both.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    There is not enough time for the parental haplotype to fully break down, and therefore the distortion of the SFS extends over a rather large region (more than 50 Mb), making it a challenge to pinpoint the incompatibilities themselves.

    “There is not enough time for the parental haplotype to fully break down and therefore the distortion of the SFS extends over a rather large region (more than 50 Mb, see supplementary fig. S14, Supplementary Material online), making it a challenge to pinpoint the incompatibilities themselves.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    The method is robust to variation in hybrid zone age and can be applied without assumptions about this parameter.

    “This result indicates that our method is robust to variation in hybrid zone age and it can be applied without assumptions about this parameter.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Epistatic selection is less effective than single-locus selection because an incompatibility allele is only disfavored when combined with the incompatibility allele at the second locus; this difference is strongest when epistatic interactions are fully recessive.

    “Epistatic selection is less effective than single-locus selection because an incompatibility allele is only disfavored when it is combined with the incompatibility allele at the second locus (this difference is strongest when epistatic interactions are fully recessive).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    A cross-validation analysis was performed using 25% of the simulations, randomly chosen, from scenarios with epistatic selection in which the incompatibility loci were detectable.

    “we performed a cross-validation analysis, using 25% of the simulations (randomly chosen) from scenarios with epistatic selection in which the incompatibility loci were detectable”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    When selection is strong, neutral alleles in linkage disequilibrium with A are pushed to a lower frequency, making it harder to distinguish the SFS under strong epistatic selection from the baseline SFS; therefore, the signal is recovered further away.

    “When selection is strong, neutral alleles in linkage disequilibrium with A are pushed to a lower frequency, making it harder to distinguish the SFS under strong epistatic selection from the baseline SFS. Therefore, the signal is recovered further away.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Incompatibility alleles decrease in frequency at a faster rate compared with recombination when selection is strong; as a result, linkage disequilibrium between incompatibility alleles and neutral alleles extends further and so does the distortion in the SFS.

    “incompatibility alleles decrease in frequency at a faster rate compared with recombination when selection is strong. As a result, linkage disequilibrium between incompatibility alleles and neutral alleles extends further and so does the distortion in the SFS.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationnone given

    The method is applicable to any pair of nascent species that form a natural hybrid zone; breeding organisms in the laboratory is not required.

    “First, the method is applicable to any pair of nascent species that form a natural hybrid zone; breeding organisms in the laboratory is not required.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Due to epistatic selection, allele A decreases in frequency, and so do neutral alleles in linkage disequilibrium with A; conversely, allele a and linked neutral alleles increase in frequency.

    “Due to epistatic selection, allele A decreases in frequency, and so do the neutral alleles in linkage disequilibrium with A. Conversely, allele a and linked neutral alleles increase in frequency.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisenone given

    The probability of observing a given data set under a multinomial distribution depends on the number of observations as well as how they are partitioned.

    “The probability of observing a given data set under a multinomial distribution depends on the number of observations as well as how they are partitioned.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The number of hybrid incompatibilities is expected to grow faster than linearly with divergence time.

    “The number of hybrid incompatibilities is expected to grow faster than linearly with divergence time (Orr 1995)”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Orr 1995 source

  • May combine two claims.

    Interpretationreasoning

    More stringent filters decrease or leave unchanged the false positive rate, but may increase the proportion of false positives among outliers due to exclusion of true positive outliers.

    “although the false positive rate will decrease (or remain unchanged in the worst-case scenario) when the filters become more stringent, the proportion of false positives among outliers may increase, due to the possible exclusion of true positive outliers.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    The too-large distortion is either due to the strength of epistatic selection itself or to the antagonist interaction between single-locus and epistatic selection maintaining one of the two incompatible alleles at high frequency (0.77 after generation 1,000).

    “This “too large” distortion is either due to the strength of epistatic selection itself or the antagonist interaction between single-locus and epistatic selection maintaining one of the two incompatible allele at high frequency (0.77 after generation 1,000).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    The correction for multiple testing is a factor 9/100, meaning that the false positive rate should be below 4.5×10-3.

    “The correction for multiple testing is a factor 9/100, meaning that the false positive rate should be below 4.5×10-3.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Schumer and Brandvain (2016) demonstrated that pairs of incompatibility loci tend to be found in the extreme tail of the distribution of admixture disequilibrium.

    “Schumer and Brandvain (2016) demonstrated that pairs of incompatibility loci tend to be found in the extreme tail of the distribution of admixture disequilibrium”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Schumer and Brandvain 2016 source

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

    Interpretationnone given

    More complex models of migration including structured source populations, asymmetrical migration, and migration out of hybrid populations are likely features of real hybrid zones.

    “More complex models of migration including structured source populations, asymmetrical migration, and migration out of hybrid populations are likely features of real hybrid zones.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given

    Both hybrid populations and source populations will routinely depart from mutation-drift equilibrium due to demographic factors.

    “both hybrid populations and source populations will routinely depart from mutation-drift equilibrium due to demographic factors.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Interpretationreasoning

    The disparity likely reflects the decision to model BDM incompatibilities with asymmetric fitness effects and to consider weaker selection, in addition to the limited time frame during which epistatic selection distorts admixture disequilibrium.

    “We suspect this disparity reflects our decision to model (BDM) incompatibilities with asymmetric fitness effects (whereas Schumer and Brandvain [2016] focused on incompatibilities with symmetric fitness effects) and to consider scenarios with weaker selection, in addition to the limited time frame during which epistatic selection distorts admixture disequilibrium.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Confirmation that a gene participates in an incompatibility will ultimately require experimental validation.

    “Confirmation that a gene participates in an incompatibility will ultimately require experimental validation (see, e.g., Bozdag et al. 2021)”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Bozdag et al. 2021 source

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given

    Incompatibility alleles are expected to be enriched for fixed differences outside the hybrid zone.

    “We expect incompatibility alleles to be enriched for fixed differences outside the hybrid zone”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Reduced introgression can result from single-locus selection against deleterious mutations that accumulate in source populations with small effective sizes.

    “reduced introgression can result from single-locus selection against deleterious mutations that accumulate in source populations with small effective sizes (Harris and Nielsen 2016).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Harris and Nielsen 2016 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Incompatibilities are often polymorphic in nature.

    “incompatibilities are often polymorphic in nature (Cutter 2012).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Cutter 2012 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    The same genome sequence data required for the method can be used to further refine the location of incompatibility loci.

    “The same genome sequence data required for our method can be used to further refine the location of incompatibility loci.”
    Quote in the paperNot a substantial claim
55 of 55

  • The authors state this as an assumption.

    In the simulations, incompatibility alleles were fixed in source populations.

    “our simulations assumed that incompatibility alleles were fixed in source populations”

  • The authors state this as an assumption.

    The measurement of method performance assumed that sequences could be determined with complete accuracy.

    “Our measurement of method performance also assumed that sequences could be determined with complete accuracy.”

  • The authors state this as an assumption.

    The position of outliers is independently and uniformly distributed along the chromosome.

    “assuming the position of outliers is independently and uniformly distributed along the chromosome (6.8%)”
3 of 3

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Reported resolution for weak codominant selection disagrees with Table 2

    Minor

    The weak-epistasis codominant row of Table 2 selects d=18 when optimizing power. At 500 kb per window, its 2*d+1 detection region is 18.5 Mb, not the stated 15.5 Mb.

1 of 1

Cites

Flagged first, then by citations

3 of the 62 references couldn’t be matched to a known paper.

Research works printed but not linked: 1

  1. Muller, 1942, Isolating mechanisms, evolution and temperature, Biol Symp, 6, 71

Cited by

Sorted by citations

Related

Authors
Blanckaert, A., & Payseur, B. A.
Affiliations
University of Wisconsin–Madison (US)
Corresponding author
Blanckaert, Alexandre
ORCID
Venue
Molecular Biology and Evolution
Publisher
Oxford University Press (OUP)
Published
2021-06-07
Volume
38
Issue
10
Pages
4616-4627
ISSN
  • 0737-4038 (ISSN-L)
  • 1537-1719
DOI
10.1093/molbev/msab168
PMC
PMC8476132
OpenAlex
W3170669887
Cited by
7
References
62
Citations by year
2026: 1 · 2025: 1 · 2022: 4 · 2021: 1
FWCI
0.63
Type
journal article
Language
en
Open access
Licence
Locations
  • Molecular Biology and Evolution (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Europe PMC (PubMed Central) (submittedVersion, other-oa)
  • PubMed Central (submittedVersion, cc-by)
Funders
  • National Science Foundation (DEB 1353737)
  • National Institutes of Health (R01GM120051, R35GM139412)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Evolution and Genetic Dynamics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Genetic diversity and population structure
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Phylogenetic Studies
Keywords
hybrid incompatibility, hybrid zone, epistatic selection, postzygotic isolation, site frequency spectrum, speciation, hybridization, migration, whole genome sequencing
MeSH terms
Alleles, Gene Frequency, Hybridization, Genetic, Genomics, Reproductive Isolation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-09.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-09.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review