Source
Pittis & Gabaldón (2016), Nature 531: 101–104, LIT-1595, read in full in NOTE-1240 from the PubMed Central author manuscript (PMC4780264), not the version of record. The main text, the Methods (“Measurement of the phylogenetic distance …: stem lengths” above all), Table 1, and the captions of Figures 1–3 and Extended Data Figures 1–8.
What was actually shown
- The measure. For each LECA family, the branch to its closest prokaryotic sister group, divided by the family’s median branch length inside eukaryotes. The correction removes rate differences only “Under the assumption that rates pre- and post-LECA are correlated”.
- The order. Bacterial families are shorter than archaeal (P = 1.38e-25). Alphaproteobacterial families are the shortest bacterial ones, with Verrucomicrobia/Chlamydiales, and shorter than other bacterial families at P = 4.4e-2. A mixture model puts alphaproteobacterial, mitochondrial and energy-production families in the shortest component, and archaeal, nuclear and ribosomal ones in the longest two.
- Origin, not function. Families matched for function, dN/dS, connectivity or expression still differ by origin. Raw stem lengths give the same order.
- Controls. Cyanobacterial families in plants have shorter stems than alphaproteobacterial ones, as the later plastid predicts. Halving alphaproteobacterial sampling, removing Rickettsiales and stricter LECA definitions leave the alphaproteobacterial result. Transfer scenarios are argued not to produce the non-alpha signal.
- Location. Alphaproteobacterial families associate with the mitochondrion (P ≤ 1e-6). Other bacterial families associate with endomembranes. Archaeal families associate with the nucleus.
What could have failed: bacterial families could have had similar stems whatever their source, as “a simple fusion model” predicts. They did not, though the alphaproteobacterial separation from other bacteria is narrow.
What this does not say
- Not that the host was structurally complex. What is shown is that its genes were of mixed origin before the mitochondrion, on the method’s assumptions. “Complex cell” in the conclusion means a genome with bacterial pathways in it.
- Not how the chimaera arose. “Serial symbiotic associations with different partners, the existence of prokaryotic consortia, or gradual waves of HGT” are all allowed.
- Not that stem length is time. It is time multiplied by an unknown ratio of rates, and it is distance to the closest sampled relative. The paper’s own Lokiarchaeota analysis shows that adding a closer genome shortens stems (30 families). Archaea were sampled with 130 genomes against 3,318 bacterial ones. The inference that sampling contributes to the archaeal–bacterial gap is mine.
- Not that mitochondria ended bacterial transfer or gave a “definitive selective advantage”. Both are asserted in the last paragraph and not analysed.
Connections
- QUESTION-277. It bears on whether eukaryogenesis was one merger or a sequence. It is evidence that genes from several bacterial sources reached the host lineage before the mitochondrion, by partners or by transfer. It does not address D, R, C or S.
- LIT-1566 (THEORY-1099). Kay et al. agree that the mitochondrion came late, by dating duplications with a relaxed clock. They find most non-alphaproteobacterial duplications after the mitochondrion and argue against serial endosymbiosis. Duplication is not acquisition, so this is a tension, not a contradiction. The comparison is mine.
Note of 2026-10-30: promoted, and the wider thesis filed beside it
The twenty-second status review promoted this reading under ADR-065: it reports what Pittis and Gabaldón’s analysis finds and how the authors read it, and NOTE-1240 read the paper in full. Whether the mitochondrion was acquired late by an already chimaeric host is THEORY-1206’s question, Proposed, which keeps this entry’s old promote_when. The record’s own critique, once in the title, is now CLAIM-509, Proposed.