CoChat
Sign Up
Cited by 5Genomics and Phylogenetic StudiesEvolution and Genetic DynamicsEvolution and Paleontology Studies

The DOI matches its Crossref record; 92 of the 92 references it cites were found in a registry or an index; 10 of the 31 sources its claims cite were checked for support.

Practical Guidance and Workflows for Identifying Fast Evolving Non-Coding Genomic Elements Using PhyloAcc

Open access
Thomas, Gregg W C, Gemmell, Patrick, Shakya, Subir B, Hu, Zhirui, Liu, Jun S, Sackton, Timothy B, Edwards, Scott V·2024·Integrative And Comparative Biology
DOI 10.1093/icb/icae056PMC PMC11579529
Open this paper in CoChat — free accountOpen PDF

Abstract

Synopsis: Comparative genomics provides ample ways to study genome evolution and its relationship to phenotypic traits. By developing and testing alternate models of evolution throughout a phylogeny, one can estimate rates of molecular evolution along different lineages in a phylogeny and link these rates with observations in extant species, such as convergent phenotypes. Pipelines for such work can help identify when and where genomic changes may be associated with, or possibly influence, phenotypic traits. We recently developed a set of models called PhyloAcc, using a Bayesian framework to estimate rates of nucleotide substitution on different branches of a phylogenetic tree and evaluate their association with pre-defined or estimated phenotypic traits. PhyloAcc-ST and PhyloAcc-GT both allow users to define a priori a set of target lineages and then compare different models to identify loci accelerating in one or more target lineages. Whereas ST considers only one species tree across all input loci, GT considers alternate topologies for every locus. PhyloAcc-C simultaneously models molecular rates and rates of continuous trait evolution, allowing the user to ask whether the two are associated. Here, we describe these models and provide tips and workflows on how to prepare the input data and run PhyloAcc.

Overview

AI snapshot

Written from the paper’s full text.

TL;DR

This paper provides practical guidance and Snakemake workflows for applying the PhyloAcc family of Bayesian models (ST, GT, C) to identify fast-evolving non-coding genomic elements, illustrated with the Zoonomia 241-mammal alignment and echolocation. In the example, 622 conserved elements on human chromosome 1 fit a model of accelerated evolution on echolocating lineages using Bayes factor cutoffs of 5.

Key findings

  • PhyloAcc models share three latent conservation states—neutral, conserved, and accelerated—and compare scenarios M0 (no acceleration), M1 (acceleration only on focal lineages), and M2 (acceleration on any branch) via Bayes factors; PhyloAcc-ST assumes the species tree, PhyloAcc-GT allows gene tree heterogeneity, and PhyloAcc-C links molecular rates to continuous trait evolution.
  • The conserved-element prediction workflow uses phyloP scores corrected by Benjamini–Hochberg FDR at 0.05, a minimum conserved element size of 200 bp, and merging of adjacent windows; depending on the cutoff, it predicted between 331,208 and 5,202 conserved elements, with the example proceeding with 331,208 elements, then subsampling to chromosome 1 (33,017 elements) and requiring all 241 species, leaving 8,928 elements to test.
  • In the echolocation example on the Zoonomia 241-mammal alignment, using Bayes factor cutoffs of 5, 622 of 8,928 conserved elements on chromosome 1 fit a model allowing accelerated rates only in one or more echolocating lineages.
  • Compute-time estimates for the Zoonomia workflow were: neutral-site extraction a little over 2 days, neutral model estimation an additional 3 days, conserved-element estimation roughly 3–4 days, and PhyloAcc-ST 3 days on chromosome 1; the entire pipeline completed in under 2 weeks.

Methods

  • The paper is practical guidance and workflow documentation for the PhyloAcc family, describing input requirements, command-line and R interfaces, Snakemake pipelines, post-processing, and interpretation of Bayes factors.
  • The empirical example uses the Zoonomia 241-way mammal MAF alignment with human as reference, phyloP conservation scores, and a neutral substitution model estimated from four-fold degenerate sites with phyloFit; conserved elements are predicted from clusters of conserved phyloP sites, extracted with a custom MAF indexer, deduplicated, and filtered to elements present in all 241 species.
  • PhyloAcc-ST was run on human chromosome 1 conserved elements, comparing marginal likelihoods for M0, M1, and M2 via Bayes factors, with post-processing summarizing loci favoring M1 or M2.

Limitations

  • The paper states that Bayes factors used by PhyloAcc are sensitive to within-model priors and are somewhat controversial, so sensitivity analyses under different prior settings are recommended.
  • The paper states that PhyloAcc assumes a small number of rate categories are sufficient to describe alignment data, which may not capture directional, purifying, or relaxed selection manifesting differently across branches; users are advised to examine highlighted elements and parameter estimates and to re-estimate branch lengths with more flexible clocks.
  • The paper also notes that phastCons may place too many constraints on conserved-element prediction and scale poorly, that BLAST-based retrieval does not lend itself to discovering new conserved elements and can retrieve paralogs, and that PhyloAcc does not currently implement codon models for coding sequences; evident from the text, the empirical demonstration is limited to PhyloAcc-ST on human chromosome 1 rather than the full genome or GT/C models, and no functional validation of the 622 elements is reported.

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

Integrity signals

Registry

Crossref · 10.1093/icb/icae056

Cites

92 of 92 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

10 of 31 cited sources checked · 5 supported · 0 flagged · accuracy not yet measured

  • Premisecitation

    Trimming too much data may remove important phylogenetic signal.

    “tr imming t oo muc h data may in fact remove im portan t phylogenet ic sig na l ( Tan et a l. 2015 )”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Tan et al. 2015 source
  • Premisecitation

    Most other methods proposed for linking rates of genomic and phenotypic evolution are not available as easily accessible, standalone software.

    “m ost oth er m eth o ds prop osed for this purpose are not available a s ea sil y accessib le, stan dalon e software ( Hi l ler et a l. 2012 ;Marcovitz et al. 2016 ;Ber g er et al. 2018 )”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites Hiller et al. 2012 source

      The Introduction describes the paper’s forward-genomics method, and Results and Discussion (‘Many phenotypes are changed in independent lineages’; page unavailable) says users can run searches through a web portal. A web portal is not standalone software, but the paper does not establish that its method lacks a standalone implementation or that most other methods do. (LLM, conf=0.73)

      Experimental — accuracy not yet measured
    • cites Marcovitz et al. 2016 source

      The Discussion (page not provided) describes a reverse-genomics method linking genomic changes with phenotypic traits, so the paper is an example of a method proposed for this purpose. Neither that passage nor the passages retrieved about software and code availability establish whether the method is available as easily accessible, standalone software. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites Berger et al. 2018 source

      Checked before — its earlier verdict is held back until a new check runs.

  • Findingcitation

    Relaxing the assumption of a single topology for all loci improved on the performance of PhyloAcc, particularly in situations such as the loss of flight in paleognaths that also experience high levels of ILS.

    “R el axing the as s umptio n o f a single topology for a l l loci improve d on t he per for mance of PhyloAcc, p art icu l arly in situ ations such as the loss of flight in p a le og nat hs t hat also exper ien ce high leve ls of ILS ( C lout ier et al . 2019 ; Sac kt on et al. 2019).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Cloutier et al. 2019 source

      Checked before — its earlier verdict is held back until a new check runs.

    • cites Sackton et al. 2019 source
  • Premisenone given

    The number of conserved elements easily obtainable by BLAST is not as high as with methods that identify them from aligned genomes.

    “Addi tio nally, the number of con serv ed e lem ents e asily obt ainable by bla st i s not a s high a s wit h met hods t ha t iden tify them from a lig ne d genomes.”
    Quote in the paperNot a substantial claim
  • Premisereasoning

    Because the variance rate multipliers characterize the rate but not the direction of phenotypic change, large increases or decreases in a trait are treated in the same way as rapid trait evolution.

    “Note tha t, beca use the va ria nce rate mu lt iplier s c haract er ize t h e rate but n ot the dire ct io n o f ph en otypic chan g e, lar g e increases or decreases in a trait ar e tr eated in t he s ame wayas rapid t rait evolut ion.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim. 25
  • It may state two separate claims as one. 14
  • The quote only partly states this claim. 1

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Premisecitation

    Conserved non-coding regions may be linked to regulation of gene expression.

    “we want to use PhyloAcc to infer substi tu tio n rate shifts in con serv ed no n-coding regio ns, which may be lin ke d to regu lat ion of gene expression ( Pollard et al. 2006 ;Booker et al. 2016 ;Hol loway et a l. 2016 ;Kostka et a l. 2018 ).”
    Quote in the paperNot a substantial claim3 of 3 sources checked
    • cites Pollard et al. 2006 source
    • cites Booker et al. 2016 source
    • cites Holloway et al. 2016 source

  • It may state two separate claims as one.

    Premisecitation

    The Zoonomia project provides the alignment in MAF format with human as the reference species and conservation scores for every site in the human genome inferred with phyloP.

    “Th e Zoon omia proj e ct provides this a lig nment in MAF fo rmat wi th human as the r efer ence species.They also p rovide co nservatio n sco res fo r every si te in the human gen om e that were inferred with phyloP.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Zoonomia 2020 source

      The Methods > Alignment section says the authors exported a MAF using human as the reference, calculated phyloP conservation scores, and made the alignment and conservation annotations available. It does not establish that a score is provided for every site in the human genome. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • It may state two separate claims as one.

    Premisecitation

    The BFs that PhyloAcc uses to determine which model best fits the data are known to be sensitive to within-model priors and are therefore somewhat controversial, even for very simple models.

    “o ne potent ia l is s ue i s th at th e B Fs that PhyloAcc uses to determine which m ode l best fits the data are known to be sen sitiv e to within-m ode l priors ( Kass and Raftery 1995 ;Sinharay and Stern 2002 ), and are therefore s ome wha t con troversial ( Ten deiro an d Kiers 2019 ; va n Ravenzwaa ij a nd Wag enmakers 2022 ), ev en for v ery simple m ode ls.”
    Quote in the paperNot a substantial claim1 of 3 sources checked
    • cites Kass and Raftery 1995 source
    • cites Sinharay and Stern 2002 source
    • cites Tendeiro and Kiers 2019 source

      The indexed Abstract describes a wide range of issues and shortcomings in Bayes-factor testing, supporting the characterization of the approach as somewhat controversial. The available passages do not establish that these issues arise even for very simple models. The within-model-prior sensitivity claim is attached to other citations in the sentence. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Premisecitation

    The first applications of PhyloAcc to genomewide datasets are beginning to appear.

    “Inde e d , the fir st a pplica tio ns o f PhyloAcc to geno mew ide d ata sets are beginning to appear ( Sun et al. 2022 ;Xu et al. 2022 ;Cicconardi et al. 2023 ;Eastment et al. 2024 )”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites Sun et al. 2022 source

      The Methods section (‘Evolution analysis of regulatory elements in 16 limb-related genes’) says the authors used PhyloAcc on alignments built from mammalian genomes. But they screened candidate regions near 16 limb-related genes, and the Results (‘Analysis of cis-regulatory elements’) report 26 CNEs in those regions. This supports an application to genomic data, not a genomewide analysis. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites Xu et al. 2022

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

    • cites Cicconardi et al. 2023 Eastment et al. 2024 source

  • The quote only partly states this claim.

    Findingown data

    Within a subset of 8928 conserved elements present in all 241 species on chromosome 1 and using Bayes Factor cutoffs of 5, there are 622 conserved elements that fit a model allowing accelerated rates only in one or more echolocating lineages.

    “We found that within a subset of 8928 con serv ed e lem en ts presen t in a l l 241 species on chrom osom e 1 an d usin g Bay es Factor cu to ffs o f 5, ther e ar e 622 con serv ed e lem en ts tha t fit a 3.”
    • Also stated in: § Running PhyloAcc-ST and GT and interpr eting r esults · plain
      “m ode l a l lowing accelerate d rates on ly in o ne o r mo re e cholocat in g lineag es ( Fig. 3 ).”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: PhyloAcc-ST assumes that the species tree and gene tree are identical.

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Findingown data

    Conserved element estimation from conserved sites using phyloP took roughly 3-4 days.

    “Con serv ed e lem en t estima tio n fro m co n serv ed sites using phyloP also took roughly 3-4 days.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Findingreasoning

    The entire pipeline was completed in under 2 weeks.

    “In a l l, this me ans t he entire p i peline was co mpleted in under 2 w eeks.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Findingown data

    For our example of chromosome 1 of the Zoonomia data, we required that all 241 species have some sequence for each element, leaving us with 8928 elements to test.

    “F or our exam ple o f chro moso me 1 o f the Zoo no mi a d ata, we a lso re quire d that a l l 241 spe cies have s ome s equence for each element, leaving us with 8928 e lem ents t o t est ( Fig. S3 ).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Findingown data

    Running neutral model extraction on each chromosome in parallel takes a little over 2 days.

    “Running on each chro moso me in p ara l lel, t his t akes a litt le over 2 days.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Findingnone given

    For our brief example, we proceeded with the maximum number predicted (331,208 elements), but subsequently sub-sampled to chromosome 1 only (33,017 elements).

    “and for our brief example, we proce e de d with the maximum number pre dicte d (331,208 e lem ents), but su bsequently sub-sampled to chro moso me 1 only (33,017 elements).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Findingown data

    PhyloAcc-ST itself took 3 days on chromosome 1, using batches of 20 loci with 24 processes allocated for each batch and submitting 20 batches at once.

    “Fina l ly, PhyloAcc-ST itself took 3 days on chrom osom e 1, using bat c hes of 20 loci with 24 processes a l locate d for each bat c h and sub mi tt ing 20 b at c hes at once.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Findingown data

    Most predicted conserved sites were also high depth.

    “We also considered a lig nment depth (number of species present) at a given si te bu t foun d that m ost pre dicte d con serv ed sites were also high depth ( Fig. S1 ).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper may give no reason why it holds. It may state two separate claims as one.

    Premisenone given

    Although several tools exist for extracting sequences from MAF, none of them scale well with the size of the alignment nor do they properly handle the presence of duplicate (paralogous) sequences in the MAF.

    “Alt hough s e veral tool s exi s t for s uch tasks, n on e of th em scale we ll with the size of the a lig nment nor do they properly handle the presence of duplicate (p ara logous) se quen ces in th e MAF.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper may give no reason why it holds.

    Premisenone given

    There were relatively few scalable and statistically robust tools to examine convergently evolved genotype-phenotype associations in a phylogenetic context.

    “th er e wer e r e lative ly fe w s c al able and s tatis tically ro bus t tools to examine conv er g ently ev olv ed g enotypeph en otyp e asso ciations in a phylogenetic context.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Premisecitation

    Recent updates to PhyloAcc have made the software much easier to use on large datasets.

    “Addi tio nally, recent u pdates to PhyloAcc have made the software much easier to use on lar g e datas ets ( G emm e l l et a l.2024 ; Yan et a l.2023 ).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Gemmell et al. 2024 source

      Checked before — its earlier verdict is held back until a new check runs.

    • cites Yan et al. 2023 source

      Checked before — its earlier verdict is held back until a new check runs.

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds. It may state two separate claims as one.

    Premisenone given

    PhyloAcc-ST and its descendant methods can be run on any input loci, or in combination with other methods.

    “And whi le employe d orig ina l ly on CNEEs, PhyloAcc-ST and its descendant methods can be run on any input loci, or in combination with other methods.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Premisenone given

    The Zoonomia alignment includes 12 Cetaceans, 5 echolocating bats in the suborder Yinpterochiroptera, and 20 echolocating bats in the suborder Yangochiroptera.

    “Th e Zoon omia a lig nm ent in cludes 12 Cet ace ans, 5 e cholocat ing b ats in th e su bor der Yinpter ochir o p tera, and 20 e cholocat ing bats in the suborder Yan g ochiro p tera.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Interpretationreasoning

    With the index, we no longer need to read through the entire MAF or store large portions of it in memory to look up alignment blocks.

    “With the index, we now no lon g er ne e d to read through the entire MAF or store lar g e po rtio ns o f i t in m em o ry to look u p a lig nment blocks.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Premisenone given

    Approaches that translate nucleotide sequences to amino acids, align, and back-translate have the advantage of being much faster computationally than codon-aware aligners but may also be more complicated technically.

    “Approaches like these have the advantage of being much faster co mpu tatio n ally th an codon-aware a lig ners but may a lso be more com plica ted t ec hnically”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.

    Premisereasoning

    PhyloAcc allows us to study convergence of putative regulatory elements that may play a role in the evolution of echolocation.

    “PhyloAcc a l lows us to study conv er g ence o f pu tat ive regu latory e lem ents th at m a y pla y a role in the evol u tio n o f e cholocat ion.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Premisecitation

    PhyloAcc-ST assumed a single tree identical to the species tree for all CNEEs examined and did not specifically model the evolution of the binary trait itself.

    “th e m eth od as s ume d a single t re e ident ica l to the spe cies t re e for a l l CNEEs examine d and did not spe cifica l ly m ode l th e evol u tio n o f the b inary trai t i tself ( Hu et al . 2019 ).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Hu et al. 2019

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

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Premisenone given

    PhyloAcc is one of a handful of methods explicitly designed to link rates of genomic and phenotypic evolution.

    “PhyloAcc is but one of a handful of m eth ods exp licitl y desig ne d to lin k rates o f geno mic and phenotyp ic evol u tio n”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim.

    Premisecitation

    PhyloAcc-ST is a Bayesian method for inferring shifts in substitution rate on designated target species annotated for a binary trait.

    “we first p rod uced PhyloAcc (with t he or ig ina l m ode l n ow r eferr ed to as PhyloAcc-ST, for S pe cies T re e), a Bayesian m eth od for infer r ing shif ts in s ubs ti tu tio n rate on designated tar g et species annotated for a bina ry tra it”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Hu et al. 2019

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

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds. It may state two separate claims as one.

    Premisenone given

    Whereas methods like phyloP can identify very short (<50 bp) stretches of conserved sites, obtaining short loci by BLAST can become problematic.

    “Fina l ly, where as met hods lik e phyloP ca n identify very s h ort ( < 50 bp) stret c hes o f co n serv ed sites, obtainin g s h ort loci by B LAST can beco me p roblematic.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Premisecitation

    Gene tree heterogeneity has been known to induce apparent accelerations due to excess substitutions on gene tree branches not conforming to the species tree, yet is rarely acknowledged in tests of molecular rates among lineages.

    “Gene t re e h eterogen eity h a s been kn own to in duce a pparen t accelera tio ns d ue to excess s ubs ti tu tio ns o n gene t re e bran ch es n ot co nfo rming to the spe cies t re e, yet is rare ly ackn owle dge d in tests of mole cu la r rates a m ong lin eages ( Men des an d Hahn 2016 ).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Mendes and Hahn 2016 source

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Premisecitation

    PhyloAcc and its descendants are already being used in a variety of contexts and animal clades.

    “PhyloAcc an d its descendants provide a n importa nt adva nce towa rd re alizing t he f u l l potent ia l of PhyloG2P a nd a r e alr eady being used in a variety of contexts and animal clades ( Sac kt on et al . 2019 ;Sun et al . 2022 ;Wuit c hik et al . 2022 ;Xu et al. 2022 ;Cicconardi et a l. 2023 ;East ment et al. 2024 ).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Sackton et al. 2019 source
    • cites Sun et al. 2022 source

      Checked before — its earlier verdict is held back until a new check runs.

    • cites Wuitchik et al. 2022 source

      Checked before — its earlier verdict is held back until a new check runs.

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Premisecitation

    Edwards and others (2017) and Wu and others (2024) both used BLAST to retrieve thousands of CNEEs from avian genomes.

    “F or exam p le, Ed wards and others (2017) and Wu and o thers (2024) bo th used BLAST to retrieve th ousan ds of CNEEs from av i an gen om es.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Edwards and others (2017)

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

    • cites Wu and others (2024)

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

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds. It may state two separate claims as one.

    Premisenone given

    phastCons may place too many constraints on conserved element prediction and is difficult to scale to large alignments.

    “How ev er, ph a stCons m ay place too m a ny constra ints o n co n serv ed e lem ent pre dict ion and is difficult to sc ale to l ar g e a lig nmen ts.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Premisenone given

    The major challenge with the BLAST approach is that it does not lend itself to discovery of new conserved elements and can lead to higher rates of retrieving paralogous loci.

    “Th e major cha l len g e wit h t hi s approach i s th at it does n ot len d itse lf to discovery of new con serv ed e lem ents an d can lead to higher rates of retrieving paralogous loci.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Premisecitation

    Molecular convergence of echolocation between toothed whales and bats has usually been constrained to gene sequences.

    “Alth ough m ole cu lar conv er g en ce of ech olo cation b etween to oth ed whales an d bats h a s be en studie d p reviously, i t has usua l ly be en const raine d to gene sequences ( Davies et al . 2012 ;S hen et a l. 2012 ;Liu et a l. 2014 ;Thom a s and Hahn 2015 ;Zou and Zhang 2015 ;Lambert et al. 2017 ;Marcovitz et al. 2019 )”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Davies et al. 2012 source

      Checked before — its earlier verdict is held back until a new check runs.

    • cites Shen et al. 2012 source

      Checked before — its earlier verdict is held back until a new check runs.

    • cites Liu et al. 2014 source

      Checked before — its earlier verdict is held back until a new check runs.

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Premisecitation

    PhyloAcc-GT builds on the PhyloAcc-ST model by allowing gene tree heterogeneity across loci.

    “a method that builds on the PhyloAcc-ST m ode l by a l lowin g g ene t re e h eterogen eity across loci ( Yan et a l. 2023 ).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Yan et al. 2023 source

      Checked before — its earlier verdict is held back until a new check runs.

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.

    Interpretationnone given

    The precise times for similar datasets will depend on hardware and parallelization.

    “Note that even f or simila r dat asets, t he pre cise t imes wi l l depend on ha rdwa re a nd pa ralleliza tion.”
    Quote in the paperNot a substantial claim
32 of 32

  • The authors state this as an assumption.

    The PhyloAcc family of models assumes that a small number of rate categories are sufficient to describe alignment data.

    “a limitation of the PhyloAcc family of models is t hat t hey a l l as s ume a sma l l n umber of ra te ca tegories are sufficient to describe a lig nmen t da t a”

  • The authors state this as an assumption.

    PhyloAcc-ST assumes that the species tree and gene tree are identical.

    “PhyloAcc-ST assumes the species tree (outline) and gene tree (inner topology) are identical”
2 of 2

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?

  • PhyloG2P studies have revealed genetic replication in noncoding regions when it was not found in coding regions.

    “Similarly, PhyloG2P studies have revealed the presence of genetic replication in noncoding regions when it was not found in coding regions (Langer et al. 2018; Feigin et al. 2019; Thomas et al. 2024; Shakya et al. 2025).”

    From From Trees to Traits: A Review of Advances in PhyloG2P Methods and Future Directions (2025) · cites this paper as Thomas et al. 2024

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.

  • Opposite criteria given for assigning loci to the gene-tree model

    Minor

    The first passage says loci are assigned to the gene-tree model when discordance is below a cutoff, but the second says the model is run when the gene tree is significantly different from the species tree. These describe opposite selection criteria.

1 of 1

Citations

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Metadata

Authors
Thomas, G. W. C., Gemmell, P., Shakya, S. B., Hu, Z., Liu, J. S., Sackton, T. B., & Edwards, S. V.
Affiliations
  • Harvard University (US)
  • Gladstone Institutes (US)
Corresponding author
  • Thomas, Gregg W C
  • Edwards, Scott V
ORCID
Venue
Integrative And Comparative Biology
Publisher
Oxford University Press (OUP)
Published
2024-05-30
Volume
64
Issue
5
Pages
1513-1525
ISSN
  • 1540-7063 (ISSN-L)
  • 1557-7023
DOI
10.1093/icb/icae056
PMC
PMC11579529
OpenAlex
W4399250809
Semantic Scholar
41c873a6b901eaf322623dc5c7b8dd302f291726
Cited by
5
References
92
Citations by year
2026: 2 · 2025: 2 · 2024: 1
FWCI
0.58
Type
journal article
Language
en
Open access
Licence
academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model (vor)
Locations
  • Integrative and Comparative Biology (publishedVersion)
  • PubMed Central (publishedVersion)
  • PubMed (publishedVersion)
Funders
National Institutes of Health (NHGRI R01HG011485)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Phylogenetic Studies
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Evolution and Genetic Dynamics
  • Physical Sciences › Earth and Planetary Sciences › Paleontology › Evolution and Paleontology Studies
Keywords
molecular evolution rate, nucleotide substitution rates, comparative genomics, convergent evolution, species tree estimation, continuous trait evolution, genome evolution, phenotypic traits
MeSH terms
Animals, Bayes Theorem, Models, Genetic, Phylogeny, Software, Evolution, Molecular, Genomics, Workflow

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.

A language model drew the claims from the paper on 2026-11-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-11-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (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