A comparison of methods for estimating substitution rates from ancient DNA sequence data

Sequential Bayesian Phylogenetic Inference

The ideal approach to Bayesian phylogenetic inference is to estimate all parameters of interest jointly in a single hierarchical model. However, this is often not feasible in practice due to the high computational cost. Instead, phylogenetic pipelines generally consist of sequential analyses, whereby a single point estimate from a given analysis is used as input for the next analysis (e.g., a single multiple sequence alignment is used to estimate a gene tree). In this framework, uncertainty is not propagated from step to step, which can lead to inaccurate or spuriously confident results. Here, we formally develop and test a sequential inference approach for Bayesian phylogenetic inference, which uses importance sampling to generate observations for the next step of an analysis pipeline from the posterior distribution produced in the previous step. Our sequential inference approach presented here not only accounts for uncertainty between analysis steps but also allows for greater flexibility in software choice (and hence model availability) and can be computationally more efficient than the traditional joint inference approach when multiple models are being tested. We show that our sequential inference approach is identical in practice to the joint inference approach only if sufficient information in the data is present (a narrow posterior distribution) and/or sufficiently many important samples are used. Conversely, we show that the common practice of using a single point estimate can be biased, for example, a single phylogeny estimate can transform an unrooted phylogeny into a time-calibrated phylogeny. We demonstrate the theory of sequential Bayesian inference using both a toy example and an empirical case study of divergence-time estimation in insects using a relaxed clock model from transcriptome data. In the empirical example, we estimate 3 posterior distributions of branch lengths from the same data (DNA character matrix with a GTR+Γ+I substitution model, an amino acid data matrix with empirical substitution models, and an amino acid data matrix with the PhyloBayes CAT-GTR model). Finally, we apply 3 different node-calibration strategies and show that divergence time estimates are affected by both the data source and underlying substitution process to estimate branch lengths as well as the node-calibration strategies. Thus, our new sequential Bayesian phylogenetic inference provides the opportunity to efficiently test different approaches for divergence time estimation, including branch-length estimation from other software.

Do morphometric data improve phylogenetic reconstruction? A systematic review and assessment

BACKGROUND

Isolating phylogenetic signal from morphological data is crucial for accurately merging fossils into the tree of life and for calibrating molecular dating. However, subjective character definition is a major limitation which can introduce biases that mislead phylogenetic inferences and divergence time estimation. The use of quantitative data, e.g., geometric morphometric (GMM; shape) data can allow for more objective integration of morphological data into phylogenetic inference. This systematic review describes the current state of the field in using continuous morphometric data (e.g., GMM data) for phylogenetic reconstruction and assesses the efficacy of these data compared to discrete characters using the PRISMA-EcoEvo v1.0. reporting guideline, and offers some pathways for approaching this task with GMM data. A comprehensive search string yielded 11,123 phylogenetic studies published in English up to Oct 2023 in the Web of Science database. Title and abstract screening removed 10,975 articles, and full-text screening was performed for 132 articles. Of these, a total of twelve articles met final inclusion criteria and were used for downstream analyses.

RESULTS

Phylogenetic performance was compared between approaches that employed continuous morphometric and discrete morphological data. Overall, the reconstructed phylogenies did not show increased resolution or accuracy (i.e., benchmarked against molecular phylogenies) as continuous data alone or combined with discrete morphological datasets.

CONCLUSIONS

An exhaustive search of the literature for existing empirical continuous data resulted in a total of twelve articles for final inclusion following title/abstract, and full-text screening. Our study was performed under a rigorous framework for systematic reviews, which showed that the lack of available comparisons between discrete and continuous data hinders our understanding of the performance of continuous data. Our study demonstrates the problem surrounding the efficacy of continuous data as remaining relatively intractable despite an exhaustive search, due in part to the difficulty in obtaining relevant comparisons from the literature. Thus, we implore researchers to address this issue with studies that collect discrete and continuous data sets with directly comparable properties (i.e., describing shape, or size).

Temporal dynamics of mildly deleterious nonsynonymous substitutions in mitochondrial gene sequences in rodents and moles

We have previously estimated the evolutionary rate (number of substitutions/site/million years) of mitochondrial cytochrome b gene (Cytb) sequences in rodents and moles to be about 0.11 at more recent divergence times of tens of thousands of years, and to decrease rapidly to about 0.03 at more distant divergence times. Because this time dependency is thought to be caused by the removal of mildly deleterious substitutions in later generations, we focused in this study on the abundance of nonsynonymous substitutions. We collected 23 haplogroups of Cytb with signals of late Quaternary population expansion events from rodents and moles and categorized them into three groups for comparison based on predicted expansion start time: 5,000-15,000 years ago (Group I), ca. 53,000 years ago (Group II) and 130,000-230,000 years ago (Group III). We counted the numbers of nonsynonymous and synonymous substitutions in all haplogroups. The rates of nonsynonymous substitutions were lowest in Groups II and III (0.08-0.22), whereas those in Group I varied markedly. We further classified Group I into two subgroups based on high (0.29-0.43) and low (0.09-0.20) nonsynonymous substitution rates, which were likely to be associated with the start of the expansion within 10,000 years and at around 15,000 years ago, respectively. The Group II and III networks had two- or three-step star-shaped structures and tended to exhibit frequent and less frequent nonsynonymous substitutions on exterior and interior branches, respectively. Based on temporal dynamics, nonsynonymous mitochondrial DNA (mtDNA) substitutions in small mammals accounted for at most 40% of all substitutions during the early evolutionary stage and then rapidly declined, dropping to approximately 15%. The results of this study provide a good explanation of the time-dependent trend in the mtDNA evolution rate predicted in previous work.

Variation in the substitution rates among the human mitochondrial haplogroup U sublineages

Resolving the absolute timescale of phylogenetic trees stipulates reliable estimates for the rate of DNA sequence evolution. For this end, various calibration methods have been developed and studied intensively. Intraspecific rate variation among distinct genetic lineages, however, has gained less attention. Here, we have assessed lineage-specific molecular rates of human mitochondrial DNA (mtDNA) by performing tip-calibrated Bayesian phylogenetic analyses. Tip-calibration, as opposed to traditional nodal time stamps from dated fossil evidence or geological events, is based on sample ages and becoming ever more feasible as ancient DNA data from radiocarbon-dated samples accumulate. We focus on subhaplogroups U2, U4, U5a, and U5b, the data including ancient mtDNA genomes from ¹⁴C-dated samples (n = 234), contemporary genomes (n = 301), and two outgroup sequences from haplogroup R. The obtained molecular rates depended on the data sets (with or without contemporary sequences), suggesting time-dependency. More notable was the rate variation between haplogroups: U4 and U5a stand out having a substantially higher rate than U5b. This is also reflected in the divergence times obtained (U5a: 17,700 years and U5b: 29,700 years), a disparity not reported previously. After ruling out various alternative causes (e.g., selection, sampling, and sequence quality), we propose that the substitution rates have been influenced by demographic histories, widely different among populations where U4/U5a or U5b are frequent. As with the Y-chromosomal subhaplogroup R1b, the mitochondrial U4 and U5a have been associated with remarkable range extensions of the Yamnaya culture in the Bronze Age.

Detection of Ancient Viruses and Long-Term Viral Evolution

The COVID-19 outbreak has reminded us of the importance of viral evolutionary studies as regards comprehending complex viral evolution and preventing future pandemics. A unique approach to understanding viral evolution is the use of ancient viral genomes. Ancient viruses are detectable in various archaeological remains, including ancient people's skeletons and mummified tissues. Those specimens have preserved ancient viral DNA and RNA, which have been vigorously analyzed in the last few decades thanks to the development of sequencing technologies. Reconstructed ancient pathogenic viral genomes have been utilized to estimate the past pandemics of pathogenic viruses within the ancient human population and long-term evolutionary events. Recent studies revealed the existence of non-pathogenic viral genomes in ancient people's bodies. These ancient non-pathogenic viruses might be informative for inferring their relationships with ancient people's diets and lifestyles. Here, we reviewed the past and ongoing studies on ancient pathogenic and non-pathogenic viruses and the usage of ancient viral genomes to understand their long-term viral evolution.

Pedigree derived mutation rate across the entire mitochondrial genome of the Norfolk Island population

Estimates of mutation rates for various regions of the human mitochondrial genome (mtGenome) vary widely, depending on whether they are inferred using a phylogenetic approach or obtained directly from pedigrees. Traditionally, only the control region, or small portions of the coding region have been targeted for analysis due to the cost and effort required to produce whole mtGenome Sanger profiles. Here, we report one of the first pedigree derived mutation rates for the entire human mtGenome. The entire mtGenome from 225 individuals originating from Norfolk Island was analysed to estimate the pedigree derived mutation rate and compared against published mutation rates. These individuals were from 45 maternal lineages spanning 345 generational events. Mutation rates for various portions of the mtGenome were calculated. Nine mutations (including two transitions and seven cases of heteroplasmy) were observed, resulting in a rate of 0.058 mutations/site/million years (95% CI 0.031-0.108). These mutation rates are approximately 16 times higher than estimates derived from phylogenetic analysis with heteroplasmy detected in 13 samples (n = 225, 5.8% individuals). Providing one of the first pedigree derived estimates for the entire mtGenome, this study provides a better understanding of human mtGenome evolution and has relevance to many research fields, including medicine, anthropology and forensics.

Mutation hotspots and spatiotemporal distribution of SARS-CoV-2 lineages in Brazil, February 2020-2021

The COVID-19 pandemic has already reached more than 110 million people and is associated with 2.5 million deaths worldwide. Brazil is the third worst-hit country, with approximately 10.2 million cases and 250 thousand deaths. International efforts have been established to share information about Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) epidemiology and evolution to support the development of effective strategies for public health and disease management. We aimed to analyze the high-quality genome sequences from Brazil from February 2020-2021 to identify mutation hotspots, geographical and temporal distribution of SARS-CoV-2 lineages by using phylogenetics and phylodynamics analyses. We describe heterogeneous sequencing efforts, the progression of the different lineages along time, evaluating mutational spectra and frequency oscillations derived from the prevalence of specific lineages across different Brazilian regions. We found at least seven major (1-7) and two minor clades related to the six most prevalent lineages in the country and described its spatial distribution and dynamics. The emergence and recent frequency shift of lineages (P.1 and P.2) carrying mutations of concern in the spike protein (e. g., E484K, N501Y) draws attention due to their association with immune evasion and enhanced receptor binding affinity. Improvements in genomic surveillance are of paramount importance and should be extended in Brazil to better inform policy makers about better decisions to fight the COVID-19 pandemic.

Considering Pleistocene North American wolves and coyotes in the eastern Canis origin story

The evolutionary origins and hybridization patterns of Canis species in North America have been hotly debated for the past 30 years. Disentangling ancestry and timing of hybridization in Great Lakes wolves, eastern Canadian wolves, red wolves, and eastern coyotes are most often partitioned into a 2-species model that assigns all ancestry to gray wolves and/or coyotes, and a 3-species model that includes a third, North American evolved eastern wolf genome. The proposed models address recent or sometimes late Holocene hybridization events but have largely ignored potential Pleistocene era progenitors and opportunities for hybridization that may have impacted the current mixed genomes in eastern Canada and the United States. Here, we re-analyze contemporary and ancient mitochondrial DNA genomes with Bayesian phylogenetic analyses to more accurately estimate divergence dates among lineages. We combine that with a review of the literature on Late Pleistocene Canis distributions to: (a) identify potential Pleistocene progenitors to southern North American gray wolves and eastern wolves; and (b) illuminate opportunities for ancient hybridization events. Specifically, we propose that Beringian gray wolves (C. lupus) and extinct large wolf-like coyotes (C. latrans orcutti) are likely progenitors to Mexican and Plains gray wolves and eastern wolves, respectively, and may represent a potentially unrecognized source of introgressed genomic variation within contemporary Canis genomes. These events speak to the potential origins of contemporary genomes and provide a new perspective on Canis ancestry, but do not negate current conservation priorities of dwindling wolf populations with unique genomic signatures and key ecologically critical roles.

Genomic epidemiology of SARS-CoV-2 in Esteio, Rio Grande do Sul, Brazil

BACKGROUND

Brazil is the third country most affected by Coronavirus disease-2019 (COVID-19), but viral evolution in municipality resolution is still poorly understood in Brazil and it is crucial to understand the epidemiology of viral spread. We aimed to track molecular evolution and spread of Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in Esteio (Southern Brazil) using phylogenetics and phylodynamics inferences from 21 new genomes in global and regional context. Importantly, the case fatality rate (CFR) in Esteio (3.26%) is slightly higher compared to the Rio Grande do Sul (RS) state (2.56%) and the entire Brazil (2.74%).

RESULTS

We provided a comprehensive view of mutations from a representative sampling from May to October 2020, highlighting two frequent mutations in spike glycoprotein (D614G and V1176F), an emergent mutation (E484K) in spike Receptor Binding Domain (RBD) characteristic of the B.1.351 and P.1 lineages, and the adjacent replacement of 2 amino acids in Nucleocapsid phosphoprotein (R203K and G204R). E484K was found in two genomes from mid-October, which is the earliest description of this mutation in Southern Brazil. Lineages containing this substitution must be subject of intense surveillance due to its association with immune evasion. We also found two epidemiologically-related clusters, including one from patients of the same neighborhood. Phylogenetics and phylodynamics analysis demonstrates multiple introductions of the Brazilian most prevalent lineages (B.1.1.33 and B.1.1.248) and the establishment of Brazilian lineages ignited from the Southeast to other Brazilian regions.

CONCLUSIONS

Our data show the value of correlating clinical, epidemiological and genomic information for the understanding of viral evolution and its spatial distribution over time. This is of paramount importance to better inform policy making strategies to fight COVID-19.

Bayesian Evaluation of Temporal Signal in Measurably Evolving Populations

Phylogenetic methods can use the sampling times of molecular sequence data to calibrate the molecular clock, enabling the estimation of evolutionary rates and timescales for rapidly evolving pathogens and data sets containing ancient DNA samples. A key aspect of such calibrations is whether a sufficient amount of molecular evolution has occurred over the sampling time window, that is, whether the data can be treated as having come from a measurably evolving population. Here, we investigate the performance of a fully Bayesian evaluation of temporal signal (BETS) in sequence data. The method involves comparing the fit to the data of two models: a model in which the data are accompanied by the actual (heterochronous) sampling times, and a model in which the samples are constrained to be contemporaneous (isochronous). We conducted simulations under a wide range of conditions to demonstrate that BETS accurately classifies data sets according to whether they contain temporal signal or not, even when there is substantial among-lineage rate variation. We explore the behavior of this classification in analyses of five empirical data sets: modern samples of A/H1N1 influenza virus, the bacterium Bordetella pertussis, coronaviruses from mammalian hosts, ancient DNA from Hepatitis B virus, and mitochondrial genomes of dog species. Our results indicate that BETS is an effective alternative to other tests of temporal signal. In particular, this method has the key advantage of allowing a coherent assessment of the entire model, including the molecular clock and tree prior which are essential aspects of Bayesian phylodynamic analyses.

Human molecular evolutionary rate, time dependency and transient polymorphism effects viewed through ancient and modern mitochondrial DNA genomes

Human evolutionary genetics gives a chronological framework to interpret the human history. It is based on the molecular clock hypothesis that suppose a straightforward relationship between the mutation rate and the substitution rate with independence of other factors as demography dynamics. Analyzing ancient and modern human complete mitochondrial genomes we show here that, along the time, the substitution rate can be significantly slower or faster than the average germline mutation rate confirming a time dependence effect mainly attributable to changes in the effective population size of the human populations, with an exponential growth in recent times. We also detect that transient polymorphisms play a slowdown role in the evolutionary rate deduced from haplogroup intraspecific trees. Finally, we propose the use of the most divergent lineages within haplogroups as a practical approach to correct these molecular clock mismatches.

Gene Tree Estimation Error with Ultraconserved Elements: An Empirical Study on Pseudapis Bees

Summarizing individual gene trees to species phylogenies using two-step coalescent methods is now a standard strategy in the field of phylogenomics. However, practical implementations of summary methods suffer from gene tree estimation error, which is caused by various biological and analytical factors. Greatly understudied is the choice of gene tree inference method and downstream effects on species tree estimation for empirical data sets. To better understand the impact of this method choice on gene and species tree accuracy, we compare gene trees estimated through four widely used programs under different model-selection criteria: PhyloBayes, MrBayes, IQ-Tree, and RAxML. We study their performance in the phylogenomic framework of $>$800 ultraconserved elements from the bee subfamily Nomiinae (Halictidae). Our taxon sampling focuses on the genus Pseudapis, a distinct lineage with diverse morphological features, but contentious morphology-based taxonomic classifications and no molecular phylogenetic guidance. We approximate topological accuracy of gene trees by assessing their ability to recover two uncontroversial, monophyletic groups, and compare branch lengths of individual trees using the stemminess metric (the relative length of internal branches). We further examine different strategies of removing uninformative loci and the collapsing of weakly supported nodes into polytomies. We then summarize gene trees with ASTRAL and compare resulting species phylogenies, including comparisons to concatenation-based estimates. Gene trees obtained with the reversible jump model search in MrBayes were most concordant on average and all Bayesian methods yielded gene trees with better stemminess values. The only gene tree estimation approach whose ASTRAL summary trees consistently produced the most likely correct topology, however, was IQ-Tree with automated model designation (ModelFinder program). We discuss these findings and provide practical advice on gene tree estimation for summary methods. Lastly, we establish the first phylogeny-informed classification for Pseudapis s. l. and map the distribution of distinct morphological features of the group. [ASTRAL; Bees; concordance; gene tree estimation error; IQ-Tree; MrBayes, Nomiinae; PhyloBayes; RAxML; phylogenomics; stemminess].

The geography of evolutionary divergence in the highly endemic avifauna from the Sierra Madre del Sur, Mexico

Background

Mesoamerica is a remarkable region with a high geological and ecological complexity. Within northern Mesoamerica, the biotic province of the Sierra Madre del Sur (SMS) in southwestern Mexico harbors exceptionally high avian endemism and diversity. Herein, we searched for spatially and temporally concordant phylogeographic patterns, in four bird genera from three distinct avian orders co-distributed across Mesoamerica and investigated their causes through hypothesis testing regarding historical processes. Selected species include endemic and differentiated populations across the montane forests of Mesoamerica, and particularly within the SMS.

Results

We gathered mitochondrial DNA sequences for at least one locus from 177 individuals across all species. We assessed genetic structure, demographic history, and defined a framework for the coalescent simulations used in biogeographic hypothesis testing temporal and spatial co-variance. Our analyses suggested shared phylogeographic breaks in areas corresponding to the SMS populations, and between the main montane systems in Mesoamerica, with the Central Valley of Oaxaca and the Nicaragua Depression being the most frequently shared breaks among analyzed taxa. Nevertheless, dating analyses and divergence patterns observed were consistent with the hypothesis of broad vicariance across Mesoamerica derived from mechanisms operating at distinct times across taxa in the SMS.

Conclusions

Our study provides a framework for understanding the evolutionary origins and historical factors enhancing speciation in well-defined regions within Mesoamerica, indicating that the evolutionary history of extant biota inhabiting montane forests is complex and often idiosyncratic.

Environmental temperatures shape thermal physiology as well as diversification and genome-wide substitution rates in lizards

Climatic conditions changing over time and space shape the evolution of organisms at multiple levels, including temperate lizards in the family Lacertidae. Here we reconstruct a dated phylogenetic tree of 262 lacertid species based on a supermatrix relying on novel phylogenomic datasets and fossil calibrations. Diversification of lacertids was accompanied by an increasing disparity among occupied bioclimatic niches, especially in the last 10 Ma, during a period of progressive global cooling. Temperate species also underwent a genome-wide slowdown in molecular substitution rates compared to tropical and desert-adapted lacertids. Evaporative water loss and preferred temperature are correlated with bioclimatic parameters, indicating physiological adaptations to climate. Tropical, but also some populations of cool-adapted species experience maximum temperatures close to their preferred temperatures. We hypothesize these species-specific physiological preferences may constitute a handicap to prevail under rapid global warming, and contribute to explaining local lizard extinctions in cool and humid climates.

The molecular clock of Mycobacterium tuberculosis

The molecular clock and its phylogenetic applications to genomic data have changed how we study and understand one of the major human pathogens, Mycobacterium tuberculosis (MTB), the etiologic agent of tuberculosis. Genome sequences of MTB strains sampled at different times are increasingly used to infer when a particular outbreak begun, when a drug-resistant clone appeared and expanded, or when a strain was introduced into a specific region. Despite the growing importance of the molecular clock in tuberculosis research, there is a lack of consensus as to whether MTB displays a clocklike behavior and about its rate of evolution. Here we performed a systematic study of the molecular clock of MTB on a large genomic data set (6,285 strains), covering different epidemiological settings and most of the known global diversity. We found that sampling times below 15-20 years were often insufficient to calibrate the clock of MTB. For data sets where such calibration was possible, we obtained a clock rate between 1x10-8 and 5x10-7 nucleotide changes per-site-per-year (0.04-2.2 SNPs per-genome-per-year), with substantial differences between clades. These estimates were not strongly dependent on the time of the calibration points as they changed only marginally when we used epidemiological isolates (sampled in the last 40 years) or three ancient DNA samples (about 1,000 years old) to calibrate the tree. Additionally, the uncertainty and the discrepancies in the results of different methods were sometimes large, highlighting the importance of using different methods, and of considering carefully their assumptions and limitations.