An even larger genetic study of extant birds in Nature

Feng et al. 2020 bring us
yet another genomic study of extant birds, this time with a circular cladogram so dense it makes no attempt to list the 10,135 bird taxa in this study by dozens of authors.

This was my reply on the study,
copied from the Comments section on the Nature website. The asterisk and double asterisk are how the chicken* and finch** are located on the dense cladogram (their figure 1).

“How can one test the validity of genomic studies like this one? Earlier testing by Prum et al. 2015 nested flamingoes (Phoenicopterus) with dissimilar grebes (Aechmophorus) and yardbirds/chickens (Gallus) with dissimilar geese (Anser). Three of these four are not named in the present cladogram which lists and illustrates only a few sample genera. Among these, the genomically distant separation of the phenomically similar finch**, parrot (Agapornis) and chicken* duplicate what was discovered earlier in the Prum et al. 2015 study. Genomic deep time studies too often produce false positives that separate similar taxa and lump dissimilar taxa. By comparison, phenomic studies, like the one online at: http://reptileevolution.com lump similar taxa and separate dissimilar taxa, modeling evolutionary events while including fossil taxa.

Only in phenomic (trait-based) studies can one produce a cladogram in which all related taxa document a gradual accumulation of derived traits modeling actual events. If one is concerned about convergence, adding taxa to phenomic studies overcomes that problem.

Genomic studies have lumped bats with whales (Laurasiatheria) and golden moles with elephants (Afrotheria). Workers have to wake up to the sad fact that genetic studies work in criminal investigations, but not in cladograms.”

References
Feng et al. (dozens of co-authors) 2020. Dense sampling of bird diversity increases power of comparative genomics. Nature 587:252–257.