DNA Sequences Are as Useful as Protein Sequences for Inferring Deep Phylogenies

Abstract

Inference of deep phylogenies has almost exclusively used protein rather than DNA sequences, based on the perception that protein sequences are less prone to homoplasy and saturation or to issues of compositional heterogeneity than DNA sequences. Here we analyze a model of codon evolution under an idealized genetic code and demonstrate that those perceptions may be misconceptions. We conduct a simulation study to assess the utility of protein versus DNA sequences for inferring deep phylogenies, with protein-coding data generated under models of heterogeneous substitution processes across sites in the sequence and among lineages on the tree, and then analyzed using nucleotide, amino acid, and codon models. Analysis of DNA sequences under nucleotide-substitution models (possibly with the third codon positions excluded) recovered the correct tree at least as often as analysis of the corresponding protein sequences under modern amino acid models. We also applied the different data-analysis strategies to an empirical dataset to infer the metazoan phylogeny. Our results from both simulated and real data suggest that DNA sequences may be as useful as proteins for inferring deep phylogenies and should not be excluded from such analyses. Analysis of DNA data under nucleotide models has a major computational advantage over protein-data analysis, potentially making it feasible to use advanced models that account for among-site and among-lineage heterogeneity in the nucleotide-substitution process in inference of deep phylogenies.

Genomic transfers help to decipher the ancient evolution of filoviruses and interactions with vertebrate hosts

“…Furthermore, significant differences in amino acid composition were identified between sequences linked to fish and those associated with tetrapods (Table S1). When the phylogram was based on methods that are less susceptible to LBA[32], (i.e., nucleotides with a partitioned codon model including all or only first and second positions), TAPV grouped with the sequences from genomes with a shared architecture, the mammalian MARV-like clade, instead of with the fish-associated virus, XILV (Fig 1B; Figs S1-S2). Another method that can reduce LBA due to amino acid site heterogeneity, PMSF (Posterior Mean Site Frequency profiles analysis [33]), found the same tree as the putative LBA phylogram (Fig S3).…”

“…We recommend that evolutionary analyses of divergent RNA viruses include codon-partitioned models. Such data uses the same alignment hypothesis as for amino acids but have three times as much characters and less susceptibility to LBA [32]. Adding divergent outgroups can reduce ingroup LBA (as occurred here with the L protein), but also introduce new sources of bias and LBA [35].…”

A PhyloFisher utility for nucleotide-based phylogenomic matrix construction;nucl_matrix_constructor.py

“…For deep evolutionary relationships, such as the origin of eukaryotes or the diversification of animals, it is common practice to use protein-based phylogenomic analyses (Burki et al, 2020; Simion et al, 2017). This is because protein sequences are much less prone to homoplasy compared to DNA sequences because of the slower rate of observable evolution and the larger alphabet (20 amino acids compared to 4 nucleotides) (Philippe et al, 2011, Kapli et al, 2023). This makes it helpful to use amino acid sequences rather than DNA sequences for ortholog and paralog determination.…”

“…This makes it helpful to use amino acid sequences rather than DNA sequences for ortholog and paralog determination. However, the usefulness of DNA sequence based phylogenetics for resolving relationships between closely related species is well documented, recent work suggests their utility in deep phylogenetics too (Kapli et al, 2023).…”