[5] Some excavates lack "classical" mitochondria, and are called "amitochondriate", although most retain a mitochondrial organelle in greatly modified form (e.g. a hydrogenosome or mitosome).

[3][6] However, various groups that lack these traits are considered to be derived excavates based on genetic evidence (primarily phylogenetic trees of molecular sequences).

[13] Excavates were thought to include multiple groups: Euglenozoa and Heterolobosea (Percolozoa) or Eozoa (as named by Cavalier-Smith[14]) appear to be particularly close relatives, and are united by the presence of discoid cristae within the mitochondria (Superphylum Discicristata).

A close relationship has been shown between Discicristata and Jakobida,[15] the latter having tubular cristae like most other protists, and hence were united under the taxon name Discoba, which was proposed for this supposedly monophyletic group.

The ancyromonad groove is not used for "suspension feeding", unlike in "typical excavates" (e.g. malawimonads, jakobids, Trimastix, Carpediemonas, Kiperferlia, etc).

[20] Caesar al Jewari and Sandra Baldauf argue instead that the Eukaryotes possibly started with an endosymbiosis event of a Deltaproteobacterium or Gammaproteobacterium, accounting for the otherwise unexplained presence of anaerobic bacterial enzymes in Metamonada.

In their scenario, the hydrogenosome and mitosome, both conventionally considered "mitochondrion-derived organelles", would predate the mitochondrion, and instead be derived from the earlier symbiotic bacterium.

Discoba and the rest of the Eukaryota appear to have emerged as sister taxon to the Preaxostyla, incorporating a single alphaproteobacterium as mitochondria by endosymbiosis.