E-book: Phylogenetics in the Genomic Era (Scornavacca et al. 2021)

This book was not peer-reviewed by PCI Genomics. It has undergone an internal review by the editors.

Accurate reconstructions of the relationships amongst species and the genes encoded in their genomes are an essential foundation for almost all evolutionary inferences emerging from downstream analyses. Molecular phylogenetics has developed as a field over many decades to build suites of models and methods to reconstruct reliable trees that explain, support, or refute such inferences. The genomic era has brought new challenges and opportunities to the field, opening up new areas of research and algorithm development to take advantage of the accumulating large-scale data. Such ‘big-data’ phylogenetics has come to be known as phylogenomics, which broadly aims to connect molecular and evolutionary biology research to address questions centred on relationships amongst taxa, mechanisms of molecular evolution, and the biological functions of genes and other genomic elements. This book brings together experts in the field to present a comprehensive synthesis of Phylogenetics in the Genomic Era, covering key conceptual and methodological aspects of how to build accurate phylogenies and how to apply them in molecular and evolutionary research. The paragraphs below briefly summarise the five constituent parts of the book, highlighting the key concepts, methods, and applications that each part addresses. Being organised in an accessible style, while presenting details to provide depth where necessary, and including guides describing real-world examples of major phylogenomic tools, this collection represents an invaluable resource, particularly for students and newcomers to the field of phylogenomics.

Part 1: Phylogenetic analyses in the genomic era

Modelling how sequences evolve is a fundamental cornerstone of phylogenetic reconstructions. This part of the book introduces the reader to phylogenetic inference methods and algorithmic optimisations in the contexts of Markov, Maximum Likelihood, and Bayesian models of sequence evolution. The main concepts and theoretical considerations are mapped out for probabilistic Markov models, efficient tree building with Maximum Likelihood methods, and the flexibility and robustness of Bayesian approaches. These are supported with practical examples of phylogenomic applications using the popular tools RAxML and PhyloBayes. By considering theoretical, algorithmic, and practical aspects, these chapters provide readers with a holistic overview of the challenges and recent advances in developing scalable phylogenetic analyses in the genomic era.

Part 2: Data quality, model adequacy

This part focuses on the importance of considering the appropriateness of the evolutionary models used and the accuracy of the underlying molecular and genomic data. Both these aspects can profoundly affect the results when applying current phylogenomic methods to make inferences about complex biological and evolutionary processes. A clear example is presented for methods for building multiple sequence alignments and subsequent filtering approaches that can greatly impact phylogeny inference. The importance of error detection in (meta)barcode sequencing data is also highlighted, with solutions offered by the MACSE_BARCODE pipeline for accurate taxonomic assignments. Orthology datasets are essential markers for phylogenomic inferences, but the overview of concepts and methods presented shows that they too face challenges with respect to model selection and data quality. Finally, an innovative approach using ancestral gene order reconstructions provides new perspectives on how to assess gene tree accuracy for phylogenomic analyses. By emphasising through examples the importance of using appropriate evolutionary models and assessing input data quality, these chapters alert readers to key limitations that the field as a whole strives to address.

Part 3: Resolving phylogenomic conflicts

Conflicting phylogenetic signals are commonplace and may derive from statistical or systematic bias. This part of the book addresses possible causes of conflict, discordance between gene trees and species trees and how processes that lead to such conflicts can be described by phylogenetic models. Furthermore, it provides an overview of various models and methods with examples in phylogenomics including their pros and cons. Outlined in detail is the multispecies coalescent model (MSC) and its applications in phylogenomics. An interesting aspect is that different phylogenetic signals leading to conflict are in fact a key source of information rather than a problem that can – and should – be used to point to events like introgression or hybridisation, highlighting possible future trends in this research area. Last but not least, this part of the book also addresses inferring species trees by concatenating single multiple sequence alignments (gene alignments) versus inferring the species tree based on ensembles of single gene trees pointing out advantages and disadvantages of both approaches. As an important take home message from these chapters, it is recommended to be flexible and identify the most appropriate approach for each dataset to be analysed since this may tremendously differ depending on the dataset, setting, taxa, and phylogenetic level addressed by the researcher.

Part 4: Functional evolutionary genomics

In this part of the book the focus shifts to functional considerations of phylogenomics approaches both in terms of molecular evolution and adaptation and with respect to gene expression. The utility of multi-species analysis is clearly presented in the context of annotating functional genomic elements through quantifying evolutionary constraint and protein-coding potential. An historical perspective on characterising rates of change highlights how phylogenomic datasets help to understand the modes of molecular evolution across the genome, over time, and between lineages. These are contextualised with respect to the specific aim of detecting signatures of adaptation from protein-coding DNA alignments using the example of the MutSelDP-ω∗ model. This is extended with the presentation of the generally rare case of adaptive sequence convergence, where consideration of appropriate models and knowledge of gene functions and phenotypic effects are needed. Constrained or relaxed, selection pressures on sequence or copy-number affect genomic elements in different ways, making the very concept of function difficult to pin down despite it being fundamental to relate the genome to the phenotype and organismal fitness. Here gene expression provides a measurable intermediate, for which the Expression Comparison tool from the Bgee suite allows exploration of expression patterns across multiple animal species taking into account anatomical homology. Overall, phylogenomics applications in functional evolutionary genomics build on a rich theoretical history from molecular analyses where integration with knowledge of gene functions is challenging but critical.

Part 5: Phylogenomic applications

Rather than attempting to review the full extent of applications linked to phylogenomics, this part of the book focuses on providing detailed specific insights into selected examples and methods concerning i) estimating divergence times, and ii) species delimitation in the era of ‘omics’ data. With respect to estimating divergence times, an exemplary overview is provided for fossil data recovered from geological records, either using fossil data as calibration points with an extant-species-inferred phylogeny, or using a fossilised birth-death process as a mechanistic model that accounts for lineage diversification. Included is a tutorial for a joint approach to infer phylogenies and estimate divergence times using the RevBayes software with various models implemented for different applications and datasets incorporating molecular and morphological data. An interesting excursion is outlined focusing on timescale estimates with respect to viral evolution introducing BEAGLE, a high-performance likelihood-calculation platform that can be used on multi-core systems. As a second major subject, species delimitation is addressed since currently the increasing amount of available genomic data enables extensive inferences, for instance about the degree of genetic isolation among species and ancient and recent introgression events. Describing the history of molecular species delimitation up to the current genomic era and presenting widely used computational methods incorporating single- and multi-locus genomic data, pros and cons are addressed. Finally, a proposal for a new method for delimiting species based on empirical criteria is outlined. In the closing chapter of this part of the book, BPP (Bayesian Markov chain Monte Carlo program) for analysing multi-locus sequence data under the multispecies coalescent (MSC) model with and without introgression is introduced, including a tutorial. These examples together provide accessible details on key conceptual and methodological aspects related to the application of phylogenetics in the genomic era.

References

Scornavacca C, Delsuc F, Galtier N (2021) Phylogenetics in the Genomic Era. https://hal.inria.fr/PGE/

Phylodynamic approaches enable viral genetic variation to be tracked over time, providing insight into pathogen phylogenetic relationships and epidemiological dynamics. These are important methods for monitoring viral spread, and identifying important parameters such as transmission rate, geographic origin and duration of infection [1]. This knowledge makes it possible to adjust public health measures in real-time and was important in the case of the COVID-19 pandemic [2]. However, these approaches can be complicated to use when combining a very large number of sequences. This was particularly true during the COVID-19 pandemic, when sequencing data representing millions of entire viral genomes was generated, with associated metadata enabling their precise identification.

Danesh et al. [3] present a bioinformatics pipeline, CovFlow, for selecting relevant sequences according to user-defined criteria to produce files that can be used directly for phylodynamic analyses. The selection of sequences first involves a quality filter on the size of the sequences and the absence of unresolved bases before being able to make choices based on the associated metadata. Once the sequences are selected, they are aligned and a time-scaled phylogenetic tree is inferred. An output file in a format directly usable by BEAST 2 [4] is finally generated.

To illustrate the use of the pipeline, Danesh et al. [3] present an analysis of the Delta variant in two regions of France. They observed a delay in the start of the epidemic depending on the region. In addition, they identified genetic variation linked to the start of the school year and the extension of vaccination, as well as the arrival of a new variant. This tool will be of major interest to researchers analysing SARS-CoV-2 sequencing data, and a number of future developments are planned by the authors.

References

[1] Baele G, Dellicour S, Suchard MA, Lemey P, Vrancken B. 2018. Recent advances in computational phylodynamics. Curr Opin Virol. 31:24-32. https://doi.org/10.1016/j.coviro.2018.08.009

[2] Attwood SW, Hill SC, Aanensen DM, Connor TR, Pybus OG. 2022. Phylogenetic and phylodynamic approaches to understanding and combating the early SARS-CoV-2 pandemic. Nat Rev Genet. 23:547-562. https://doi.org/10.1038/s41576-022-00483-8

[3] Danesh G, Boennec C, Verdurme L, Roussel M, Trombert-Paolantoni S, Visseaux B, Haim-Boukobza S, Alizon S. 2023. COVFlow: phylodynamics analyses of viruses from selected SARS-CoV-2 genome sequences. bioRxiv, ver. 7 peer-reviewed and recommended by Peer Community in Genomics. https://doi.org/10.1101/2022.06.17.496544

[4] Bouckaert R, Heled J, Kühnert D, Vaughan T, Wu C-H et al. 2014. BEAST 2: a software platform for Bayesian evolutionary analysis. PLoS Comput Biol 10: e1003537. https://doi.org/10.1371/journal.pcbi.1003537

One important goal of modern genetics is to establish functional associations between genotype and phenotype. Single nucleotide polymorphisms (SNPs) are numerous and widely distributed in the genome and can be obtained from nucleic acid sequencing (1). SNPs allow for the investigation of genetic diversity, which is critical for increasing crop resilience to the challenges posed by global climate change. The associations between SNPs and phenotypes can be captured in genome-wide association studies. SNPs can also be used in combination with machine learning, which is becoming more popular for predicting complex phenotypic traits like yield and biotic and abiotic stress tolerance from genotypic data (2). The availability of many SNP datasets is important in machine learning predictions because this approach requires big data to build a comprehensive model of the association between genotype and phenotype.

Aubert and colleagues have studied, as part of the PeaMUST project, the genetic diversity of 240 Pisum accessions (3). They sequenced exome-enriched genomic libraries, a technique that enables the identification of high-density, high-quality SNPs at a low cost (4). This technique involves capturing and sequencing only the exonic regions of the genome, which are the protein-coding regions. A total of 2,285,342 SNPs were obtained in this study. The analysis of these SNPs with the annotations of the genome sequence of one of the studied pea accessions (5) identified a number of SNPs that could have an impact on gene activity. Additional analyses revealed 647,220 SNPs that were unique to individual pea accessions, which might contribute to the fitness and diversity of accessions in different habitats. Phylogenetic and clustering analyses demonstrated that the SNPs could distinguish Pisum germplasms based on their agronomic and evolutionary histories. These results point out the power of selected SNPs as markers for identifying Pisum individuals.

Overall, this study found high-quality SNPs that are meaningful in a biological context. This dataset was derived from a large set of germplasm and is thus particularly useful for studying genotype-phenotype associations, as well as the diversity within Pisum species. These SNPs could also be used in breeding programs to develop new pea varieties that are resilient to abiotic and biotic stressors.  

References

1.         Fallah M, Jean M, Boucher St-Amour VT, O’Donoughue L, Belzile F. The construction of a high-density consensus genetic map for soybean based on SNP markers derived from genotyping-by-sequencing. Genome. 2022 Aug;65(8):413–25.

https://doi.org/10.1139/gen-2021-005

2.         Gill M, Anderson R, Hu H, Bennamoun M, Petereit J, Valliyodan B, et al. Machine learning models outperform deep learning models, provide interpretation and facilitate feature selection for soybean trait prediction. BMC Plant Biology. 2022 Apr 8;22(1):180.

https://doi.org/10.1186/s12870-022-03559-z

3.         Aubert G, Kreplak J, Leveugle M, Duborjal H, Klein A, Boucherot K, et al. SNP discovery by exome capture and resequencing in a pea genetic resource collection., biorxiv, ver. 4, peer-reviewed and recommended by Peer Community in Genomics.

https://doi.org/10.1101/2022.08.03.502586 

4.         Warr A, Robert C, Hume D, Archibald A, Deeb N, Watson M. Exome sequencing: current and future perspectives. G3 Genes|Genomes|Genetics. 2015 Aug 1;5(8):1543–50.

https://doi.org/10.1534/g3.115.018564

5.         Kreplak J, Madoui MA, Cápal P, Novák P, Labadie K, Aubert G, et al. A reference genome for pea provides insight into legume genome evolution. Nat Genet. 2019 Sep;51(9):1411–22.

https://doi.org/10.1038/s41588-019-0480-1

This manuscript describes the ongoing work and plans of Biogenome Portugal: a new network of researchers in the Portuguese biodiversity genomics community. The aims of this network are to jointly train scientists in ecology and evolution, generate new knowledge and understanding of Portuguese biodiversity, and better engage with the public and with international researchers, so as to advance conservation efforts in the region. In collaboration across disciplines and institutions, they are also contributing to the European Reference Genome Atlas (ERGA): a massive scientific effort, seeking to eventually produce reference-quality genomes for all species in the European continent (Mc Cartney et al. 2023).

The manuscript centers around six iconic and/or severely threatened species, whose range extends across parts of what is today considered Portuguese territory. Via the Portugal chapter of ERGA (ERGA-Portugal), the researchers will generate high-quality genome sequences from these species. The species are the Iberian hare, the Azores laurel, the Black wheatear, the Portuguese crowberry, the Cave ground beetle and the Iberian minnowcarp. In ignorance of human-made political borders, some of these species also occupy large parts of the rest of the Iberian peninsula, highlighting the importance of transnational collaboration in biodiversity efforts. The researchers extracted samples from members of each of these species, and are building reference genome sequences from them. In some cases, these sequences will also be co-analyzed with additional population genomic data from the same species or genetic data from cohabiting species. The researchers aim to answer a variety of ecological and evolutionary questions using this information, including how genetic diversity is being affected by the destruction of their habitat, and how they are being forced to adapt as a consequence of the climate emergency.

The authors did a very good job in providing a justification for the choice of pilot species, a thorough methodological overview of current work, and well thought-out plans for future analyses once the genome sequences are available for study. The authors also describe plans for networking and training activities to foster a well-connected Portuguese biodiversity genomics community.

Applying a genomic analysis lens is important for understanding the ever faster process of devastation of our natural world. Governments and corporations around the globe are destroying nature at ever larger scales (Diaz et al. 2019). They are also destabilizing the climatic conditions on which life has existed for thousands of years (Trisos et al. 2020). Thus, genetic diversity is decreasing faster than ever in human history, even when it comes to non-threatened species (Exposito-Alonso et al. 2022), and these decreases are disrupting ecological processes worldwide (Richardson et al. 2023). This, in turn, is threatening the conditions on which the stability of our societies rest (Gardner and Bullock 2021). The efforts of Biogenome Portal and ERGA-Portugal will go a long way in helping us understand in greater detail how this process is unfolding in Portuguese territories.

References

Díaz, Sandra, et al. "Pervasive human-driven decline of life on Earth points to the need for transformative change." Science 366.6471 (2019): eaax3100. https://doi.org/10.1126/science.aax3100

Exposito-Alonso, Moises, et al. "Genetic diversity loss in the Anthropocene." Science 377.6613 (2022): 1431-1435. https://doi.org/10.1126/science.abn5642

Gardner, Charlie J., and James M. Bullock. "In the climate emergency, conservation must become survival ecology." Frontiers in Conservation Science 2 (2021): 659912. https://doi.org/10.3389/fcosc.2021.659912

Mc Cartney, Ann M., et al. "The European Reference Genome Atlas: piloting a decentralised approach to equitable biodiversity genomics." bioRxiv (2023): 2023-09, ver. 2 peer-reviewed and recommended by Peer Community in Genomics. https://doi.org/10.32942/X20W3Q

Richardson, Katherine, et al. "Earth beyond six of nine planetary boundaries." Science Advances 9.37 (2023): eadh2458. https://doi.org/10.1126/sciadv.adh2458

Trisos, Christopher H., Cory Merow, and Alex L. Pigot. "The projected timing of abrupt ecological disruption from climate change." Nature 580.7804 (2020): 496-501. https://doi.org/10.1038/s41586-020-2189-9

Xenoturbella is a genus of morphologically simple bilaterians inhabiting benthic environments. Until very recently, only one species was known from the genus, Xenoturbella bocki Westblad 1949 [1]. Less than a decade ago, five more species were discovered (X. churro, X. monstrosa, X. profunda, X. hollandorum [2] and X. japonica [3]). These enigmatic animals lack an anus, a coelom, reproductive organs, nephrocytes and a centralized nervous system [1]. The systematic classification of the genus has substantially changed in the last decades, with first being considered as its own phylum (Xenoturbellida) and then being clustered together with acoels and nemertodermatids into the phylum Xenacoelomorpha [4,5]. The phylogenetic position of the xenacoelomorphs has been recalcitrant to resolution, with its position ranging from being the sister group to Nephrozoa (ie, protostomes and deuterostomes [6]) to the sister group to Ambulacraria (ie, Hemichordata and Echinodermata) in a clade called Xenambulacraria [4]. Recent studies based on expanded datasets and more refined analyses support either topology [7,8]. Either way, it is clear that additional studies on Xenoturbella could provide important insights into the origins of bilaterian traits such as the anus, the nephrons and the evolution of a centralized nervous system. 

Small but mighty genome - In this work [9], the authors present the chromosome-level genome of X. bocki - the first one for xenoturbellids - and explore their genomic idiosyncrasies in the context of other animal phyla. The first thing they discuss is the complexity of the genome, with X. bocki having a similar number of genes to other bilaterians (despite its small size of 111Mb), retained ancestral metazoan synteny, conserved clusters of Hox genes, largely complete signaling pathways and most bilaterian miRNAs present. This is not a surprise, though, as we know that the relationship between genomic and morphological complexity is far from straightforward - for instance, protist lineages closely related to animals share many gene families with us [10], and it is not the presence or absence of these gene families but their evolutionary dynamics what defines complexity in each animal phyla (eg [11]). However, the relationship between both is far from well-understood, and having a high-quality genome is the first crucial step towards a holistic understanding of genome evolution, allowing us to ask questions about how and when genes are regulated, how they interact in 3D space, or how their epigenetic landscape is shaped, for instance.

Xenacoelomorphs: deuterostomes or not? - The authors also discuss the phylogenetic position of xenacoelomorphs (including the newly generated high-quality genome of X. bocki) based on a gene presence/absence matrix. Although there is much more to be done to robustly assess the phylogenetic position of the phylum, these analyses represent a first attempt to investigate what the phylogeny looks like after the addition of the new high-quality data. The new analyses reflected once more the previously recovered phylogenies mentioned above, but this time with a twist: X. bocki was recovered as the sister group to echinoderms, yet acoels appeared as sister to all deuterostomes, hence not recovering Xenacoelomorpha as monophyletic. Thus, it is clear that much remains to be explored to disentangle the phylogenetic position of these mysterious lineages, where more sophisticated methodologies such as synteny-based orthology inference or models of evolution accounting for heterotachy probably have an important role to play. 

In any case, we are approaching a qualitative jump in how we understand phylogenomics thanks to efforts derived from the availability of chromosome-level genome assemblies for a growing number of species. Exciting times are ahead for us, evolutionary biologists, to explore what high-quality genomes - in combination with multiomics datasets - will reveal about animal evolution. I am personally really looking forward to it.  

References

1. Westblad E. (1949). Xenoturbella bocki n.g., n.sp., a peculiar, primitive Turbellarian type. Arkiv för Zoologi 1, 3-29 (1949).

2. Rouse, G. W., Wilson, N. G., Carvajal, J. I. & Vrijenhoek, R. C. New deep-sea species of Xenoturbella and the position of Xenacoelomorpha. Nature 530, 94–97 (2016). https://doi.org/10.1038/nature16545

3. Nakano, H. et al. Correction to: A new species of Xenoturbella from the western Pacific Ocean and the evolution of Xenoturbella. BMC Evol. Biol. 18, 1–2 (2018). https://doi.org/10.1186/s12862-018-1190-5​https://doi.org/10.1186/s12862-018-1190-5​

4. Philippe, H. et al. Acoelomorph flatworms are deuterostomes related to Xenoturbella. Nature 470, 255–258 (2011). https://doi.org/10.1038/nature09676

5. Hejnol, A. et al. Assessing the root of bilaterian animals with scalable phylogenomic methods. Proc. Biol. Sci. 276, 4261–4270 (2009). https://doi.org/10.1098/rspb.2009.0896

6. Cannon, J. T. et al. Xenacoelomorpha is the sister group to Nephrozoa. Nature 530, 89–93 (2016). https://doi.org/10.1038/nature16520

7. Laumer, C. E. et al. Revisiting metazoan phylogeny with genomic sampling of all phyla. Proc. Biol. Sci. 286, 20190831 (2019). https://doi.org/10.1098/rspb.2019.0831

8. Philippe, H. et al. Mitigating anticipated effects of systematic errors supports sister-group relationship between Xenacoelomorpha and Ambulacraria. Curr. Biol. 29, 1818–1826.e6 (2019). https://doi.org/10.1016/j.cub.2019.04.009

9. Schiffer, P. H., Natsidis, P., Leite D. J., Robertson, H., Lapraz, F., Marlétaz, F., Fromm, B., Baudry, L., Simpson, F., Høye, E., Zakrzewski, A-C., Kapli, P., Hoff, K. J., Mueller, S., Marbouty, M., Marlow, H., Copley, R. R., Koszul, R., Sarkies, P. & Telford, M .J. The slow evolving genome of the xenacoelomorph worm Xenoturbella bocki. bioRxiv (2023), ver. 4 peer-reviewed and recommended by Peer Community in Genomics. https://doi.org/10.1101/2022.06.24.497508

10. Suga, H. et al. The Capsaspora genome reveals a complex unicellular prehistory of animals. Nat. Commun. 4, 2325 (2013). https://doi.org/10.1038/ncomms3325

11. Fernández, R. & Gabaldón, T. Gene gain and loss across the metazoan tree of life. Nat Ecol Evol 4, 524–533 (2020). https://doi.org/10.1038/s41559-019-1069-x