SVP abstracts 1 Asher et al. trust genomics over phenomics

From the Asher et al. 2020 abstract:
“Paleontology’s fundamental contribution to science includes understanding how extinct species relate to the living. Major questions about antiquity and diversification all depend on accurately reconstructing fossils on the Tree of Life.”

As longtime readers know already, an accurate (= self-healing due to taxon inclusion) Vertebrate Tree of Life, from headless Cambrian chordates to naked bipedal primates, can be found online here.

“However, fossils are poorly known compared to living species, having degraded with usually no traces of DNA or soft tissues.”

Yeah, but…if you don’t have fossils, many interrelationships will evade your studies. So don’t put the downer on fossils, especially in an SVP abstract.

“Molecular data are sometimes recovered from very recent fossils, but in general are exceedingly rare and will remain so for the foreseeable future. Here, we apply a method that incorporates genomic data from living species and, in turn, fundamentally reshapes the phylogenetic tree of living and fossil species alike.”

A bit of advice for Asher et al.: Stop what you’re doing. Go back to fossils. Genomics don’t work in deep time time studies. False positives are possible.

“This method does not require direct access to fossil biomolecules, but can nonetheless change our understanding of the evolutionary affinities of long-extinct fossil groups. We incorporate genomic signal by using morphology reconstructed from predicted ancestors on well-corroborated trees of living species, and re-include such ancestors as terminal taxa in phylogenetic analyses of fossils.”

I have not yet seen a deep time genomic study that makes sense or corroborates a phenomic study. No one has.

“With examples from Mesozoic mammals and maniraptoran dinosaurs, we show that this method recovers a signal supported by genomic data for living taxa using only morphological data. Examples from mammals include monophyletic Laurasiatheria, Afrotheria, Xenarthra, and Euarchontoglires, as well as glirids and Aplodontia among sciuromorph rodents.”

First, let’s define clades:

  1. Laurasiatheria = hedgehogs, even-toed ungulates, whales, bats, odd-toed ungulates, pangolins, and carnivorans
  2. Afrotheria = golden moles, elephant shrews (also known as sengis), tenrecs, aardvarks, hyraxes, elephants, and sea cows.
  3. Xenarthra = anteaters, tree sloths and armadillos.
  4. Euarchontoglires = rodents, lagomorphs, treeshrews, colugos and primates.

The LRT confirms some of these interrelationships (e.g. rodents and lagomorphs), but refutes others (e. g. whates and bats).

“For fossils, our results indicate that at least some anatomically well-known North American “condylarths” are more closely related to Laurasiatheria than Afrotheria.”

Stop right there. Three fourths of the above genomic clades are not supported.

“In addition, we demonstrate that morphological datasets applied to mammalian and maniraptoran phylogenetics vary in their ability to reconstruct well-corroborated clades. Some morphological datasets improve congruence of topologies derived from subsampled sequence alignments with the well-corroborated tree; others decrease congruence.”

Stop right there. The ‘well-corroborated tree’ means their genomic results match other genomic results (e.g. Flynn et al. 2005), not realizing that bats and whales cannot be closely related to one another. That’s bonkers! Stops that sort of thinking. You went to college! Genomics produce bogus results that must be rejected.

“We provide a set of criteria which we hope will be of use to the wider community in vetting the relative information content of morphological datasets relevant for paleontology, and hope to end at least some, seemingly intractable, debates about the phylogenetic affinities of long-extinct species lacking genomic data.”

If you really want to end the debate, add extinct and extant taxa to phenomic (= trait-based) studies that include fossils. In the LRT all sister taxa actually look alike down to the details, and all convergent clades nest apart from each other. The LRT provides a framework for this task.

Asher R, Beck R, Field D and  Benson R 2020. Fossils and the tree of life: Making genomic data informative for extinct vertebrates without direct access to molecules. SVP abstracts.
Flynn JJ, et al. 2005. Molecular phylogeny of the Carnivora (Mammalia): assessing the impact of increased sampling on resolving enigmatic relationships. Syst Biol 54:317–37.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.