On the desirability of models for inferring genome phylogenies

A Place for Viruses on the Tree of Life

Viruses are ubiquitous. They infect almost every species and are probably the most abundant biological entities on the planet, yet they are excluded from the Tree of Life (ToL). However, there can be no doubt that viruses play a significant role in evolution, the force that facilitates all life on Earth. Conceptually, viruses are regarded by many as non-living entities that hijack living cells in order to propagate. A strict separation between living and non-living entities places viruses far from the ToL, but this may be theoretically unsound. Advances in sequencing technology and comparative genomics have expanded our understanding of the evolutionary relationships between viruses and cellular organisms. Genomic and metagenomic data have revealed that co-evolution between viral and cellular genomes involves frequent horizontal gene transfer and the occasional co-option of novel functions over evolutionary time. From the giant, ameba-infecting marine viruses to the tiny Porcine circovirus harboring only two genes, viruses and their cellular hosts are ecologically and evolutionarily intertwined. When deciding how, if, and where viruses should be placed on the ToL, we should remember that the Tree functions best as a model of biological evolution on Earth, and it is important that models themselves evolve with our increasing understanding of biological systems.

On the artefactual parasitic eubacteria clan in conditioned logdet phylogenies: heterotachy and ortholog identification artefacts as explanations

BACKGROUND

Phylogenetic reconstruction methods based on gene content often place all the parasitic and endosymbiotic eubacteria (parasites for short) together in a clan. Many other lines of evidence point to this parasites clan being an artefact. This artefact could be a consequence of the methods used to construct ortholog databases (due to some unknown bias), the methods used to estimate the phylogeny, or both.We test the idea that the parasites clan is an ortholog identification artefact by analyzing three different ortholog databases (COG, TRIBES, and OFAM), which were constructed using different methods, and are thus unlikely to share the same biases. In each case, we estimate a phylogeny using an improved version of the conditioned logdet distance method. If the parasites clan appears in trees from all three databases, it is unlikely to be an ortholog identification artefact.Accelerated loss of a subset of gene families in parasites (a form of heterotachy) may contribute to the difficulty of estimating a phylogeny from gene content data. We test the idea that heterotachy is the underlying reason for the estimation of an artefactual parasites clan by applying two different mixture models (phylogenetic and non-phylogenetic), in combination with conditioned logdet. In these models, there are two categories of gene families, one of which has accelerated loss in parasites. Distances are estimated separately from each category by conditioned logdet. This should reduce the tendency for tree estimation methods to group the parasites together, if heterotachy is the underlying reason for estimation of the parasites clan.

RESULTS

The parasites clan appears in conditioned logdet trees estimated from all three databases. This makes it less likely to be an artefact of database construction. The non-phylogenetic mixture model gives trees without a parasites clan. However, the phylogenetic mixture model still results in a tree with a parasites clan. Thus, it is not entirely clear whether heterotachy is the underlying reason for the estimation of a parasites clan. Simulation studies suggest that the phylogenetic mixture model approach may be unsuccessful because the model of gene family gain and loss it uses does not adequately describe the real data.

CONCLUSIONS

The most successful methods for estimating a reliable phylogenetic tree for parasitic and endosymbiotic eubacteria from gene content data are still ad-hoc approaches such as the SHOT distance method. however, the improved conditioned logdet method we developed here may be useful for non-parasites and can be accessed at http://www.liv.ac.uk/~cgrbios/cond_logdet.html.

A phylogenetic mixture model for gene family loss in parasitic bacteria

Gene families are frequently gained and lost from prokaryotic genomes. It is widely believed that the rate of loss was accelerated for some but not all gene families in lineages that became parasites or endosymbionts. This leads to a form of heterotachy that may be responsible for the poor performance of phylogeny estimation based on gene content. We describe a mixture model that accounts for this heterotachy. We show that this model fits data on the distribution of gene families across bacteria from the COG database much better than previous models. However, it still favors an artifactual tree topology in which parasites form a clade over the more plausible 16S topology. In contrast to a previous model of genome dynamics, our model suggests that the ancestral bacterium had a small genome. We suggest that models of gene family gain and loss are likely to be more useful for understanding genome dynamics than for estimating phylogenetic trees.

The tree of genomes: an empirical comparison of genome-phylogeny reconstruction methods

BACKGROUND

In the past decade or more, the emphasis for reconstructing species phylogenies has moved from the analysis of a single gene to the analysis of multiple genes and even completed genomes. The simplest method of scaling up is to use familiar analysis methods on a larger scale and this is the most popular approach. However, duplications and losses of genes along with horizontal gene transfer (HGT) can lead to a situation where there is only an indirect relationship between gene and genome phylogenies. In this study we examine five widely-used approaches and their variants to see if indeed they are more-or-less saying the same thing. In particular, we focus on Conditioned Reconstruction as it is a method that is designed to work well even if HGT is present.

RESULTS

We confirm a previous suggestion that this method has a systematic bias. We show that no two methods produce the same results and most current methods of inferring genome phylogenies produce results that are significantly different to other methods.

CONCLUSION

We conclude that genome phylogenies need to be interpreted differently, depending on the method used to construct them.

Characterizing positive and negative selection and their phylogenetic effects

Total evidence and the use of large datasets to overcome uncertainty are the state of the art in systematic analysis. This assumes that the only true phylogenetic signal is ancestry and that functional, structural, and other factors will not add an alternative signal. Using gene families, where individual codon positions were sorted into bins based upon average-pairwise dN/dS ratio, we show that standard, common phylogenetic methods that were designed for stochastic, neutral, site-independent processes, generate less robust phylogenetic signal for bins with strong negative or positive selection. This was true for phylogenetic reconstruction with parsimony, distance, and likelihood methods. Further, we present a case for the potential existence of systematic functional or structural signal that competes with ancestral signal. For the example of positive selection, we simulate the evolution of sequences through three dimensional lattice constructs with folding constraint and changing binding functionality and show that total evidence for these lattice genes presents trees with functional signal, but that the neutral synonymous sites in these genes show the true ancestral signal. In this case, sequence convergence is promoted by functional convergence.

Supertrees Disentangle the Chimerical Origin of Eukaryotic Genomes

Eukaryotes are traditionally considered to be one of the three natural divisions of the tree of life and the sister group of the Archaebacteria. However, eukaryotic genomes are replete with genes of eubacterial ancestry, and more than 20 mutually incompatible hypotheses have been proposed to account for eukaryote origins. Here we test the predictions of these hypotheses using a novel supertree-based phylogenetic signal-stripping method, and recover supertrees of life based on phylogenies for up to 5,741 single gene families distributed across 185 genomes. Using our signal-stripping method, we show that there are three distinct phylogenetic signals in eukaryotic genomes. In order of strength, these link eukaryotes with the Cyanobacteria, the Proteobacteria, and the Thermoplasmatales, an archaebacterial (euryarchaeotes) group. These signals correspond to distinct symbiotic partners involved in eukaryote evolution: plastids, mitochondria, and the elusive host lineage. According to our whole-genome data, eukaryotes are hardly the sister group of the Archaebacteria, because up to 83% of eukaryotic genes with a prokaryotic homolog have eubacterial, not archaebacterial, origins. The results reject all but two of the current hypotheses for the origin of eukaryotes: those assuming a sulfur-dependent or hydrogen-dependent syntrophy for the origin of mitochondria.