CoChat
Sign Up
PreprintCited by 28Forensic and Genetic ResearchPleistocene-Era Hominins and ArchaeologyEvolution and Genetic Dynamics

A Chronological Atlas of Natural Selection in the Human Genome during the Past Half-million Years

Open access
Zhou, Hang, Hu, Sile, Matveev, Rostislav, Yu, Qianhui, Li, Jing, Khaitovich, Philipp, Jin, Li, Lachmann, Michael +3 more·2015·bioRxiv (Cold Spring Harbor Laboratory)
DOI 10.1101/018929
Open this paper in CoChat — free accountOpen PDF

Abstract

The spatiotemporal distribution of recent human adaptation is a long standing question. We developed a new coalescent-based method that collectively assigned human genome regions to modes of neutrality or to positive, negative, or balancing selection. Most importantly, the selection times were estimated for all positive selection signals, which ranged over the last half million years, penetrating the emergence of anatomically modern human (AMH). These selection time estimates were further supported by analyses of the genome sequences from three ancient AMHs and the Neanderthals. A series of brain function-related genes were found to carry signals of ancient selective sweeps, which may have defined the evolution of cognitive abilities either before Neanderthal divergence or during the emergence of AMH. Particularly, signals of brain evolution in AMH are strongly related to Alzheimer’s disease pathways. In conclusion, this study reports a chronological atlas of natural selection in Human.

AI snapshot

Written from the paper’s full text.

TL;DR

The paper presents a coalescent-based method using PSMC/UPGMA on 1000 Genomes phase 1 data to classify human genomic regions as neutral, positive, negative, or balancing selection and to estimate selection times over the past half-million years. It reports candidate positive-selection regions in CEU, CHB, and YRI, ancient brain-related signals linked to Alzheimer’s disease pathways, and recent stress-response signals in YRI.

Key findings

  • The study identified 117, 230, and 485 candidate positive-selection (PS) regions in CEU, CHB, and YRI, respectively, occupying 1.13%–2.94% of the genome. Negative selection was higher in YRI (~10%) than in CEU (~2.5%) and CHB (~1.9%); balancing selection was abundant in YRI but much less frequent in CEU and CHB, with the MHC regions harboring the strongest balancing-selection signals in all three populations.
  • PS signals were significantly enriched for genes expressed in the brain and sperm; balancing-selection signals were enriched for MHC class II (P = 3.67 × 10⁻⁸), cellular defense response (P = 3.86 × 10⁻⁷), and cell-cell adhesion (P = 2.1 × 10⁻⁵).
  • Selection-time estimates included CEU TmaxD of 835 ga (~20.9 kya) for aFM and 706 ga (~17.7 kya) for aHG, and YRI TmaxD of 1,898 ga (47.5 kya) for aEA. Signals older than 2,218 ga (~55.5 kya) were enriched for brain function (odds ratio = 7.53, P = 8.08 × 10⁻⁶), while recent YRI stress-response signals after 444 ga (~11.1 kya) were enriched (odds ratio = 4.21, P = 8.37 × 10⁻⁴).
  • Ancient AMH brain-related PS genes including SPON1, SORL1, MAPT, ELAVL4, and SNCA were reported to interact within an Alzheimer’s disease pathogenesis sub-network; recent CEU aFM-like signals included GATM, GALK2, and SORD.

Methods

  • Coalescent trees were reconstructed from 30 haploid genomes each from CEU, CHB, and YRI in the 1000 Genomes Project phase 1 dataset using PSMC pairwise TMRCA estimates and UPGMA; rescaled trees were analyzed with a likelihood framework (H0, H1, H2) to assign neutrality, negative selection, balancing selection, or positive selection and to estimate selection times and coefficients.
  • Simulations using MSMS and SFS_CODE evaluated power and false-positive rates, with comparisons to Tajima’s D, Fay and Wu’s H, and iHS. Empirical ancient DNA analyses used Ust’-Ishim (~45,000 years old), Stuttgart (~7,000 years old), Loschbour (~8,000 years old), and a Neanderthal consensus; haplotype distance D aDNA,MHG and TmaxD were calculated with Chi-square tests and permutations.
  • Functional enrichment used DAVID and PANTHER, and temporal enrichment used Fisher’s exact test with 10,000 permutations to calculate false discovery rates.

Limitations

  • The paper states that selection coefficients can best be estimated only roughly as strong or weak, although selection times could be estimated quite accurately in simulations. It also notes that ancient PS selection times showed obvious underestimation and required correction.
  • The paper states that CEU TmaxD estimates were substantially older than radiocarbon-dated aDNA specimen times, potentially due to overestimation of selection time in empirical CEU data, early parting of the aDNA lineage from direct CEU ancestors, or strong ancestral population structure. It also notes that ancient brain-function enrichment may be confounded because many ancient signals were manually annotated based on literature review.
  • The text makes evident additional limitations: aDNA haplotype-distance relationships can be disrupted by incomplete sweeps or later outbreeding; only three ancient AMH genomes and a Neanderthal consensus were used; and the method depends on PSMC/UPGMA tree reconstruction and a correction assuming the overall genome coalescent profile is neutral.

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

Integrity signals

Registry

Crossref · 10.1101/018929

Cites

74 of 74 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 34 cited sources checked · 0 supported · 0 flagged

  • Findingnone given

    The method simultaneously estimates selection coefficient and selection time.

    “Most importantly, our method simultaneously estimates selection coefficient and selection time.”
    Quote in the paperSubstantial claim
  • Findingown data

    Simulations showed that the method has high power to detect negative selection and balancing selection for a wide range of conditions.

    “Simulations showed that our method has high power to detect NS and BS for a wide range of conditions (Supplementary Fig. 5).”
    Quote in the paperSubstantial claim
  • Findingown data

    Simulations indicate that selection coefficients can best be estimated only roughly as strong or weak.

    “Simulations indicate that selection coefficients can best be estimated only roughly as strong or weak (Supplementary Fig. 7);”
    Quote in the paperNot a substantial claim
  • Findingown data

    For positive selection simulations based on realistic demographic models, RPS and APS tests showed high combined power toward either recent (<2,500 generations ago) or ancient (2,500-24,000 generations ago) events.

    “For the PS simulations based on realistic demographic models (Supplementary Fig. 6 and Online Methods), RPS and APS tests showed high combined power toward either recent [< 2,500 generations ago (ga), Fig. 1a] or ancient (2,500-24,000 ga, Fig. 1b) events.”
    Quote in the paperSubstantial claim
  • Findingown data

    Patterns in CEU and CHB were similar to those observed in YRI.

    “Patterns in CEU and CHB were similar to those observed in YRI (Supplementary Fig. 10).”
    Quote in the paperNot a substantial claim
  • May combine two claims. 35
  • Classified as probably not a substantial claim. 24
  • The quote only partly states this claim. 1

  • May combine two claims.

    Findingown data

    In YRI, the correlation between predicted and true selection times showed good linearity that persisted into the ancient time range (2,500-24,000 generations ago), although the correlation decreased slightly to 0.62.

    “Particularly in YRI, the correlation showed good linearity (Fig. 1f) persisted into the ancient time range (2,500-24,000 ga), although the correlation between the predicted and true values decreased slightly (0.62, P = 1.17 × 10 -137 , Fig. 1g).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The median α score was lowest in promoter regions (0.755) and CDS (0.769), and progressively elevated in UTRs (0.800), enhancers (0.875), introns (0.932), and intergenic regions (1.008), revealing a strong negative correlation with expected functional essentiality.

    “The median α score was lowest in the promoter regions (0.755) and CDS (0.769), and progressively elevated in the untranslated regions (UTRs, 0.800), enhancers (0.875), introns (0.932), and intergenic regions (1.008), revealing a strong negative correlation with the expected functional essentiality of a genomic region (Fig. 2d).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Other enriched categories include alcohol-metabolism in CHB and genes expressed in hair roots in CEU.

    “Other interesting categories include alcohol-metabolism in CHB and genes expressed in hair roots in CEU (Supplementary Tables 4-6 and Online Methods).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Functional enrichment analyses of balancing selection signals revealed involvement of MHC class II (P = 3.67 × 10^-8), cellular defense response (P = 3.86 × 10^-7), and cell-cell adhesion (P = 2.1 × 10^-5).

    “Functional enrichment analyses of the BS signals revealed substantial involvement of multiple biological processes (Fig. 2e and Supplementary Tables 10-13, Online Methods), including: MHC class II (P = 3.67 × 10 -8 , Bonferroni correction); cellular defense response (P = 3.86 × 10 -7 , Bonferroni correction); and cell-cell adhesion (P = 2.1 × 10 -5 , Bonferroni correction).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    RPS and APS have moderate power (26%-87%) toward recent soft sweep events, outperforming the other tests.

    “Interestingly, RPS and APS also have moderate power (26%-87%) toward recent soft sweep events, clearly outperforming the other tests (Fig. 1d).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    71.3% and 62.2% of the balancing selection signals in CEU and CHB were also annotated as balancing selection in YRI.

    “Notably, 71.3% and 62.2% of the BS signals in CEU and CHB were also annotated as BS in YRI.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The study identified 117, 230, and 485 candidate positive selection regions in CEU, CHB, and YRI, respectively, occupying 1.13%-2.94% of the genome.

    “We identified 117, 230, and 485 candidate PS regions in CEU, CHB, and YRI, respectively (Fig. 2a,b and Supplementary Tables 123), which occupied 1.13%-2.94% of the genome.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Abundant signals of balancing selection were detected in YRI, whereas much fewer balancing selection signals were observed in CEU and CHB.

    “Abundant signals of BS were detected in YRI, whereas much fewer BS signals were observed in CEU and CHB (Fig. 2a,b and Supplementary Tables 789).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Compared to several existing tests, including Tajima's D, Fay and Wu's H and iHS, only Tajima's D outperformed RPS and APS, and only around 1,000-1,500 generations ago.

    “Compared to several existing tests, including Tajima's D 37 , Fay and Wu's H 38 and iHS 2 , only Tajima's D outperformed RPS and APS, and only around 1,000-1,500 ga (Fig. 1a,c).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [37] source

      Not checked in the free run.

    • cites [38] 2000, Hitchhiking under positive Darwinian selection, Genetics source

      Not checked in the free run.

    • cites [2] source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    In CEU and YRI, positive selection signals were enriched for genes expressed in the pituitary gland; in CHB and YRI, for genes expressed in the appendix.

    “In addition, in CEU and YRI enrichment was found for genes expressed in the pituitary gland, while enrichment for genes expressed in the appendix was observed in CHB and YRI.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The types of estimates were highly consistent either between the reconstructed and true trees or for the true trees for the different demographic scenarios.

    “Results show that the types of estimates were highly consistent either between the reconstructed and true trees (Supplementary Fig. 4), or for the true trees for the different demographic scenarios (Supplementary Fig. 15).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Selection times were estimated for all positive selection signals and ranged over the last half million years, penetrating the emergence of anatomically modern human (AMH).

    “Most importantly, the selection times were estimated for all positive selection signals, which ranged over the last half million years, penetrating the emergence of anatomically modern human (AMH).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The neutral simulations estimated the false positive rates of RPS and APS to be 2% and 3.5%, respectively.

    “The neutral simulations estimated the false positive rates of RPS and APS to be 2% and 3.5%, respectively.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Findingown data

    A more stringent threshold was applied to balancing selection due to an excessive upper tail in the empirical distribution.

    “A more stringent threshold was applied to BS due to an excessive upper tail in the empirical distribution.”
    Quote in the paperSubstantial claim

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

    Findingnone given

    The specific biological processes in the stress response category include response to wounding, inflammatory response, and defense response to other organism.

    “The specific biological processes in this category include response to wounding, inflammatory response and defense response to other organism.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingcitation

    Numerous well-established positive selection signals were re-captured, including SLC24A5, LCT, EDAR, ADH gene cluster, CCR5, and LARGE; their estimated times of selection were mostly consistent with previous reports.

    “Numerous well established PS signals were re-captured such as SLC24A5, LCT, EDAR, ADH gene cluster, CCR5, and LARGE; and their estimated times of selection were mostly consistent with previous reports 13,[15][16][17] .”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [13] source

      Not checked in the free run.

    • cites [15][16][17]

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    The haplotype profile of SLC24A5 in CEU revealed high affinity to aFM (D aFM,MHG = 2) and substantial distance to aHG (D aHG,MHG = 28).

    “Interestingly, the haplotype profile of SLC24A5 in CEU revealed a high affinity to aFM (D aFM,MHG = 2) and a substantial distance to aHG (D aHG,MHG = 28), as suggests that skin lightening associated with SLC24A5 originated from Near East, and likely was introduced into ancient Europeans via farming transition.”
    Quote in the paperSubstantial claim

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

    Findingcitation

    SPON1, MAPT, SNCA, SORL1, SHC3, and ELAVL4 were mapped to a protein-protein interaction (PPI) network.

    “SPON1, MAPT, SNCA, SORL1, SHC3, and ELAVL4 were mapped to a protein-protein interaction (PPI) network 69 (Figure 5).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [69] source

      Not checked in the free run.

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

    Findingown data

    In YRI, PS signals were classified, as shown in Supplementary Table 20.

    “Supplementary Table 20 shows the classified PS signals in YRI.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation
    § A Chronological Atlas of Natural Selection in the Human Genome during the Past Half-million Years

    A new class of methods bases population genetic inference on reconstructed coalescent information and has promising test power and resolution for inferring selection.

    “In the last few years, a new class of methods has emerged, which base the population genetic inference on reconstructed coalescent information 23,33,34 .These methods manifested promising test power and resolution for inferring selection, as the genome-wide coalescent trees provide a complete record of genealogies, mutations, and recombination 33 .”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [23] source

      Not checked in the free run.

    • cites [33] source

      Not checked in the free run.

    • cites [34] source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation
    § A Chronological Atlas of Natural Selection in the Human Genome during the Past Half-million Years

    Positive selection events older than 250,000 years ago might not leave detectable signatures given conventional test frameworks.

    “It was argued that PS events older than 250,000 ya might not leave detectable signatures given the conventional test frameworks 20,21 .”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [20] 2002, The signature of positive selection at randomly chosen loci, Genetics source

      Not checked in the free run.

    • cites [21] source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductreasoning

    The raw trees were controlled for reconstruction errors and rescaled into a coalescent time scale, eliminating effects of population size variation.

    “The raw trees were consequently controlled for reconstruction errors and rescaled into a coalescent time scale, where effects of population size variation were eliminated (Online Methods).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisereasoning

    Rescaled neutral trees follow a standard constant-size coalescent distribution, so selection events can be detected as altered coalescent patterns.

    “Rescaled neutral trees follow a standard constant-size coalescent distribution; therefore, various selection events can be detected as altered coalescent patterns.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    These estimated times are substantially older than the radiocarbon-dated times of the aDNA specimens, potentially due to a tendency to overestimate selection time in empirical CEU data.

    “These estimated times are substantially older than the radiocarbondated times of the aDNA specimens, potentially due to a tendency to overestimate the selection time in empirical CEU data.”
    Quote in the paperNot a substantial claim

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

    Study design and conductcitation

    For the Neanderthal sequence, all sites with genotype quality < 40 and mapping quality < 30 were removed.

    “For the Neanderthal sequence, we removed all sites with genotype quality < 40, and mapping quality < 30, as done previously 67 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [67] source

      Not checked in the free run.

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

    Premisecitation

    SLC24A5 plays a pivotal role in skin pigmentation lightening in Europeans.

    “SLC24A5 was shown to play a pivotal role in skin pigmentation lightening in Europeans 10 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [10] source

      Not checked in the free run.

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

    Study design and conductown data

    The efficacy of the correction was evaluated in simulations on 10 models of population size change over time.

    “The efficacy of correction was evaluated in simulations on 10 models of population size change over time (Supplementary Fig. 14 and Supplementary Note 3.1).”
    Quote in the paperNot a substantial claim

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

    Premisecitation

    ELAVL4 encodes HuD, a neuron-specific RNA-binding protein that regulates spatiotemporal activation of neuronal mRNAs and affects neuronal development and plasticity, learning, and memory.

    “ELAVL4 (2,556 ga, ~64 kya) encodes HuD, a neuron-specific RNA-binding protein that regulates the spatiotemporal activation of neuronal mRNAs, and affects neuronal development and plasticity, learning, and memory 48 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [48] source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    AUTS2 and SLTM are both involved in autism spectrum disorders, affecting communication and social interaction abilities.

    “AUTS2 (18,027 ga, ~450 kya) and SLTM (17,355 ga, ~434 kya) are both involved in autism spectrum disorders (ASD), affecting communication and social interaction abilities 41,42 .”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [41] 2010, Variants in several genomic regions associated with asperger disorder, Autism Res source

      Not checked in the free run.

    • cites [42] source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    Skin lightening associated with SLC24A5 originated from the Near East and was likely introduced into ancient Europeans via farming transition.

    “Interestingly, the haplotype profile of SLC24A5 in CEU revealed a high affinity to aFM (D aFM,MHG = 2) and a substantial distance to aHG (D aHG,MHG = 28), as suggests that skin lightening associated with SLC24A5 originated from Near East, and likely was introduced into ancient Europeans via farming transition.”
    Quote in the paperSubstantial claim

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

    Interpretationnone given

    One possible reason for the lower NS proportion in non-African populations is that ancient signals of coalescent compression were eliminated by bottlenecks in non-African populations.

    “One possible reason is that ancient signals of coalescent compression were eliminated by bottlenecks in non-African populations.”
    Quote in the paperNot a substantial claim

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

    Study design and conductown data

    The study compared TMRCA estimation and population inferences between low-coverage 1000G and high-coverage Complete Genomics data for a few individuals.

    “To investigate the potential impact of such low coverage, for a few individuals that have been sequenced both at low coverage in 1000G and high coverage in Complete Genomics 56 , we compared the TMRCA estimation and population inferences between the two (Supplementary Note 1.2).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [56] source

      Not checked in the free run.

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

    Interpretationnone given

    Under neutrality and assuming exact tree inference, the estimator of coalescent scaling coefficient α (α*) follows a distribution with a median approximately at 1.

    “Under neutrality and assuming exact tree inference, the estimator of coalescent scaling coefficient α (α * ) follows 1 1, 1 n n ⎛ ⎞ Γ - ⎜ ⎟ - ⎝ ⎠ , with a median approximately at 1.”
    Quote in the paperNot a substantial claim
    • The authors state this as an assumption: Exact tree inference is assumed when deriving the estimator of coalescent scaling coefficient α.

  • May combine two claims.

    Interpretationown data

    The brain function-related genes carrying ancient selective sweep signals may have defined the evolution of cognitive abilities either before Neanderthal divergence or during the emergence of AMH.

    “which may have defined the evolution of cognitive abilities either before Neanderthal divergence or during the emergence of AMH.”
    Quote in the paperSubstantial claim

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

    Premisecitation

    MAPT encodes the tau protein that assembles and stabilizes the microtubule framework of neurons; non-specific aggregation of tau is the hallmark of AD.

    “MAPT (3,078 ga, ~77 kya) encodes the tau protein that assembles and stabilizes the microtubule framework of neurons.Non-specific aggregation of tau is the hallmark of AD 47 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [47] source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    For a population of varying population size, with initial effective population size N0 and effective population size Ne(t), and n individuals sampled at present, it is possible to define the rescaled coalescent time as in Equation 1.

    “For a population of varying population size, with initial effective population size N 0 and effective population size N e (t), assuming n individuals sampled at present, Griffiths and Tavaré 57,58 showed that it is possible to define the rescaled coalescent time as follows:”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Griffiths and Tavaré 57,58

      Not checked in the free run.

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

    Interpretationnone given

    Positive selection signals in the aFM-like class might have contributed to early agriculture transition.

    “PS signals in the aFM-like class might have contributed to early agriculture transition.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    Sites included for analysis were required to have a minimum root-mean-square mapping quality of MQ ≥ 30.

    “Sites included for analysis were required to have a minimum root-mean-square mapping quality of MQ ≥ 30.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). The quote only partly states this claim.

    Interpretationnone given

    In the constant-size demography scenario with Ne of 10,000, setting sample sizes for Tajima's D, Fay and Wu's H, and iHS tests at 120 would render better performance for tests sensitive to small sample sizes such as iHS.

    “In this scenario, the sample sizes for Tajima's D, Fay and Wu's H, and iHS tests were set at 120, which would render better performance for tests sensitive to small sample sizes such as iHS.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationnone given

    Successful detection of selection relies on proper tree construction and rescaling, and UPGMA and PSMC may both introduce estimation errors during this process.

    “Successful detection of selection relies on proper tree construction and rescaling; UPGMA and PSMC may both introduce estimation errors during this process.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    Thorough filtering was performed on haploid genome sequences to ensure high-quality data.

    “To ensure high-quality data, thorough filtering was performed on the haploid genome sequences.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    The sudden increase of positive selection signals in stress response in YRI seems to strongly coincide with the emergence of agriculture, likely due to expanded spectra of pathogens and parasites brought by dense dwelling, irrigation and livestock raising.

    “For the more recent history, the sudden increase of PS signals in stress response in YRI seems to strongly coincide with the emergence of agriculture, likely due to the expanded spectra of pathogens and parasites brought by dense dwelling, irrigation and livestock raising.”
    Quote in the paperSubstantial claim

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

    Premisecitation

    SNCA encodes alpha-synuclein, plays an important role in neurotransmitter release and inter-neuronal signaling, and is associated with AD.

    “SNCA (2,234 ga, ~56 kya) encodes the alpha-synuclein protein and plays an important role in the release of neurotransmitters and inter-neuronal signaling, and is also associated with AD 49 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [49] source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    SHC3 is almost brain-specific, highly expressed in the cerebral cortex and frontal and temporal lobes, regulates neuronal survival, and protects the CNS against environmental stresses.

    “SHC3 (7,586 ga, ~189 kya) is almost brain-specific, highly expressed in the cerebral cortex and frontal and temporal lobes, regulates neuronal survival, and protects the CNS against environmental stresses 44 .”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [44] source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    RNF180 regulates brain levels of monoamine oxidase A (MAO-A) and affects emotional and social behaviors via the serotonin pathway.

    “RNF180 (16,601 ga, ~415 kya) regulates the brain levels of monoamine oxidase A (MAO-A) and affects emotional and social behaviors via the serotonin pathway 43 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [43] source

      Not checked in the free run.

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

    Premisecitation

    SORL1 is a neuronal endocytic receptor that regulates the recycling of APP from the cell surface.

    “SORL1 (3,285 ga, ~82 kya) is a neuronal endocytic receptor that regulates the recycling of APP from the cell surface 46 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [46] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S4 (based). May combine two claims.

    Premisereasoning

    Under neutrality, τi and τj are independent given i ≠ j, and therefore τi is exchangeable across different trees for the same ith interval without affecting the global neutral distribution.

    “Under neutrality, τ i and τ j are independent, given i ≠ j.Therefore τ i is exchangeable across different trees for the same ith interval, without affecting the global neutral distribution.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Among aFM-aHG common signals, DGKZ, SLC44A1, PITPNB, and ACSL6 are involved in lipid metabolism, and CDH8 is related to the response to cold and may be involved in climate adaptation.

    “Among these, DGKZ, SLC44A1, PITPNB, and ACSL6 are involved in lipid metabolism, and CDH8 is related to the response to cold 50 and may be involved in climate adaptation.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [50] source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductown data

    The corrected time estimation was thoroughly evaluated in simulations and applied on empirical data.

    “The corrected time estimation was thoroughly evaluated in simulations (Supplementary Fig. 1e-1g and Supplementary Fig. 8) and applied on empirical data.”
    Quote in the paperSubstantial claim

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

    Study design and conductreasoning

    The study simulated an increased density of polymorphic sites by elevating the mutation rate to test the power to detect balancing selection (BS).

    “To test the power to detect BS, we simulated an increased density of polymorphic sites by elevating the mutation rate.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Increased genetic diversity that results in more deeply structured coalescent trees is the signature of balancing selection (BS).

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

    Study design and conductnone given

    The distribution of distance between the estimated and true centers of selection is shown in Supplementary Figure 16.

    “The distribution of distance between the estimated and true centers of selection is shown in Supplementary Figure 16.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationnone given

    When PS events were more ancient, whole tree compression was not so evident, but the time dependence of the coalescent rate persisted.

    “when the PS events were more ancient, the whole tree compression was not so evident, but the time dependence of the coalescent rate persisted”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Genome regions affected by recent positive selection have extensive LD.

    “In view of extensive LD in genome regions affected by recent positive selection 2,5”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [2] source

      Not checked in the free run.

    • cites [5] source

      Not checked in the free run.

53 of 53

  • The authors state this as an assumption.

    Negative and balancing selection may be described by trees of globally enhanced or decreased coalescent rates.

    “In brief, NS/BS may be described by trees of globally enhanced/decreased coalescent rates.”

  • The authors state this as an assumption.

    Positive selection exhibits a sudden increase in the coalescent rate, and the time of this increase approximates the starting time of selection.

    “On the other hand, PS exhibits a sudden increase in the coalescent rate, the time of which should approximate the starting time of selection (Online Methods).”

  • The authors state this as an assumption.

    Rescaled neutral trees follow a standard constant-size coalescent distribution.

    “Rescaled neutral trees follow a standard constant-size coalescent distribution; therefore, various selection events can be detected as altered coalescent patterns.”

  • The paper relies on this without support.

    Selection time estimation was restricted to a few well-studied cases, mainly due to high computational cost and lack of generalized methods.

    “However, selection time estimation was restricted to a few wellstudied cases, mainly due to the high computational cost and lack of generalized methods.”

  • The paper relies on this without support.

    Soft sweep hinders a comprehensive understanding of past positive selection events due to low test power.

    “Finally, soft sweep, a mode of positive selection where selection acts on multiple standing variants, also hinders a comprehensive understanding of the past PS events due to low test power.”

  • The paper relies on this without support.

    Identifying the actual targets of selection and determining when, how, and why selection has acted on a particular target remain daunting tasks.

    “Still, while genome scans provide lists of candidate loci, identifying the actual targets of selection and determining when, how, and why selection has acted on a particular target remain daunting tasks.”

  • The paper relies on this without support.

    Galactose is primarily found in dairy food, grains, and vegetables, whereas sorbitol mainly exists in fruits.

    “Interestingly, galactose is primarily found in dairy food, grains, and vegetables, whereas sorbitol mainly exists in fruits.”

  • The paper relies on this without support.

    Creatine is essential for energy buffering in vertebrate cells, especially in muscles, and may also be supplied from carnivorous diets.

    “Creatine is essential for energy buffering in vertebrate cells, especially in muscles, and it may also be supplied from carnivorous diets.”

  • The paper relies on this without support.

    GATM encodes glycine amidinotransferase, a key enzyme for the synthesis of creatine in the human body.

    “Among the aFM-like signals, GATM (1,020 ga, ~25 kya) encodes glycine amidinotransferase, a key enzyme for the synthesis of creatine in the human body.”

  • The paper relies on this without support.

    GALK2 encodes galactokinase, an enzyme responsible for the conversion of galactose to glucose.

    “GALK2 is an aFM-like candidate gene (623 ga, ~15.6 kya) that encodes galactokinase, which is an enzyme responsible for the conversion of galactose to glucose.”

  • The paper relies on this without support.

    SORD encodes sorbitol dehydrogenase, which catalyzes the conversion of sorbitol to fructose.

    “SORD encodes sorbitol dehydrogenase, which catalyzes the conversion of sorbitol to fructose.”

  • The authors state this as an assumption.

    Under neutrality, coalescent occurs at a constant rate of (i choose 2) when there are i ancestors.

    “Under neutrality, coalescent occurs at a constant rate of 2 i ⎛ ⎞ ⎜ ⎟ ⎝ ⎠ when there are i ancestors.”

  • The authors state this as an assumption.

    The overall genome coalescent profile is neutral and therefore should fit the H0 distribution.

    “It assumes that the overall genome coalescent profile is neutral and therefore should fit the H 0 distribution.”

  • The authors state this as an assumption.

    Time rescaling is assumed for the rest of the section.

    “Time rescaling is assumed for the rest of this section.”

  • The authors state this as an assumption.

    Exact tree inference is assumed when deriving the estimator of coalescent scaling coefficient α.

    “Under neutrality and assuming exact tree inference”

  • The paper relies on this without support.

    Hitchhiking during a selective sweep results in progressively decreasing diversity toward the selection center.

    “Since hitchhiking during a selective sweep results in progressively decreasing diversity toward the selection center”

  • The paper relies on this without support.

    A recent PS is expected to affect both temporal (H2 signature) and global coalescent rates (H1 signature).

    “A recent PS is expected to affect both temporal (H 2 signature) and global coalescent rates (H 1 signature).”

  • The paper relies on this without support.

    Increased genetic diversity that results in more deeply structured coalescent trees is the signature of balancing selection (BS).

    “Increased genetic diversity that results in more deeply structured coalescent trees is the signature of BS.”
18 of 18

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • In our dataset, several genes carrying HHMCs and associated with basal ganglia functions stand out, like SLITRK1 and NOVA1.

    “in our dataset, several genes carrying HHMCs and associated with basal ganglia functions (critical for language and cognition) stand out, like SLITRK1114 and NOVA136,115–118 (Supplementary Information 4).”

    From A catalog of single nucleotide changes distinguishing modern humans from archaic hominins (2019) · cites this paper as [36]

  • The HHMC in NCOA6 is located nearby and three 5′-UTR variants within a putatively selected region, with an estimated time of selection at around 150 kya.

    “The HHMC is located nearby and three 5′-UTR variants within a putatively selected region36, with an estimated time of selection at around 150 kya (assuming a slow mutation rate).”

    From A catalog of single nucleotide changes distinguishing modern humans from archaic hominins (2019) · cites this paper as [36]

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.

  • BS threshold is also assigned to the α < 1 class

    Minor

    The BS rule applies D b 10 when α > 1, but the subsequent definition says D b 10 is the threshold for α < 1.

  • Identical threshold conditions classify a tree as both PS and NS

    Minor

    The stated RPS rule calls a tree a PS signal when D 21 exceeds D rp 21, then calls it NS under that same threshold condition. The two classifications cannot be distinguished by the rule as written.

2 of 2

Cites

Flagged first, then by citations

Research works printed but not linked: 2

  1. Mathieson, I. et al. Eight thousand years of natural selection in Europe. BioRxiv doi: http://dx.doi.org/10.1101/016477(2015). · DOI
  2. 10.1111/j.l096-0031.1989.tb00562.x · DOI

Cited by

Sorted by citations

Related

Authors
Zhou, H., Hu, S., Matveev, R., Yu, Q., Li, J., Khaitovich, P., Jin, L., Lachmann, M., Stoneking, M., Fu, Q., & Tang, K.
Affiliations
  • Chinese Academy of Sciences (CN)
  • Max Planck Institute for Mathematics in the Sciences (DE)
  • Santa Fe Institute (US)
  • Max Planck Institute for Evolutionary Anthropology (DE)
  • Broad Institute (US)
ORCID
Venue
bioRxiv (Cold Spring Harbor Laboratory)
Publisher
openRxiv
Published
2015-05-05
DOI
10.1101/018929
OpenAlex
W2129877637
Cited by
28
References
74
Citations by year
2024: 1 · 2023: 1 · 2022: 1 · 2021: 3 · 2020: 5 · 2019: 2 · 2017: 3 · 2016: 8 · 2015: 4
Type
posted content
Language
en
Open access
Licence
Locations
bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc-nd)
Funders
  • National Natural Science Foundation of China (31371267)
  • Max-Planck-Gesellschaft
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Forensic and Genetic Research
  • Social Sciences › Anthropology › Pleistocene-Era Hominins and Archaeology
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Evolution and Genetic Dynamics
Keywords
natural selection, human genome, positive selection, negative selection, balancing selection, ancient human genomes, Neanderthals, anatomically modern humans, brain evolution, cognitive evolution

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 2026-10-10.

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

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